[2022-12-19 03:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 619): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 03:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 622): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: false IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 03:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 135): INFO Creating model:dcnv3_5_new3_with_meta/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 [2022-12-19 03:45:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 139): INFO DCNv3_5_new3_WithMeta( (patch_embed): ConvTokenizer( (conv1): Conv2d(3, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm1): Sequential( (0): to_channels_last() (1): LayerNorm((160,), eps=1e-06, elementwise_affine=True) (2): to_channels_first() ) (act): GELU() (conv2): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm2): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (pos_drop): Dropout(p=0.0, inplace=False) (levels): ModuleList( (0): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): Identity() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (1): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(640, 1280, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (2): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (6): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (7): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (8): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (9): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (10): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (11): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (12): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (13): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (14): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (15): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (16): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (17): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (18): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (19): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (20): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (21): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (22): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (23): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (24): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (25): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (26): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (27): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (28): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (29): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (30): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (31): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(1280, 2560, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (post_norms): ModuleList( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (2): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (3): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (4): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (meta_head_1): Sequential( (0): Linear(in_features=4, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (meta_head_2): Sequential( (0): Linear(in_features=3, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (dcnv3_head_x4): Sequential( (0): Conv2d(2560, 4096, kernel_size=(1, 1), stride=(1, 1)) (1): PixelShuffle(upscale_factor=2) ) (dcnv3_head_x3): Conv2d(1280, 1024, kernel_size=(1, 1), stride=(1, 1)) (clip_projector): AttentionPoolingBlock( (norm1_q): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_k): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_v): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (cross_attn): CrossAttention( (q): Linear(in_features=1024, out_features=1024, bias=False) (k): Linear(in_features=1024, out_features=1024, bias=False) (v): Linear(in_features=1024, out_features=1024, bias=False) (attn_drop): Dropout(p=0.0, inplace=False) (proj): Linear(in_features=1024, out_features=768, bias=True) (proj_drop): Dropout(p=0.0, inplace=False) ) (drop_path): Identity() ) (fc_norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True) (head): Linear(in_features=768, out_features=8142, bias=True) (meta_head): Linear(in_features=128, out_features=8142, bias=True) ) [2022-12-19 03:45:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 172): INFO Using native Torch AMP. Training in mixed precision. [2022-12-19 03:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 184): INFO using fp16_compress_hook! [2022-12-19 03:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 193): INFO number of params: 1087305720 [2022-12-19 03:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 223): INFO no checkpoint found in work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384, ignoring auto resume [2022-12-19 03:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 134): INFO ==============> Loading weight /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth for fine-tuning...... [2022-12-19 03:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 265): WARNING _IncompatibleKeys(missing_keys=['meta_head_1.0.weight', 'meta_head_1.0.bias', 'meta_head_1.2.weight', 'meta_head_1.2.bias', 'meta_head_1.3.norm_fn1.weight', 'meta_head_1.3.norm_fn1.bias', 'meta_head_1.3.norm_fn2.weight', 'meta_head_1.3.norm_fn2.bias', 'meta_head_1.3.w1.weight', 'meta_head_1.3.w1.bias', 'meta_head_1.3.w2.weight', 'meta_head_1.3.w2.bias', 'meta_head_2.0.weight', 'meta_head_2.0.bias', 'meta_head_2.2.weight', 'meta_head_2.2.bias', 'meta_head_2.3.norm_fn1.weight', 'meta_head_2.3.norm_fn1.bias', 'meta_head_2.3.norm_fn2.weight', 'meta_head_2.3.norm_fn2.bias', 'meta_head_2.3.w1.weight', 'meta_head_2.3.w1.bias', 'meta_head_2.3.w2.weight', 'meta_head_2.3.w2.bias', 'fc_norm.weight', 'fc_norm.bias', 'head.weight', 'head.bias', 'meta_head.weight', 'meta_head.bias'], unexpected_keys=[]) [2022-12-19 03:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 270): INFO => loaded successfully /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth [2022-12-19 03:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 619): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 03:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 622): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: false IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 03:49:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 135): INFO Creating model:dcnv3_5_new3_with_meta/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 139): INFO DCNv3_5_new3_WithMeta( (patch_embed): ConvTokenizer( (conv1): Conv2d(3, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm1): Sequential( (0): to_channels_last() (1): LayerNorm((160,), eps=1e-06, elementwise_affine=True) (2): to_channels_first() ) (act): GELU() (conv2): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm2): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (pos_drop): Dropout(p=0.0, inplace=False) (levels): ModuleList( (0): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): Identity() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (1): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(640, 1280, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (2): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (6): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (7): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (8): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (9): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (10): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (11): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (12): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (13): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (14): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (15): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (16): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (17): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (18): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (19): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (20): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (21): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (22): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (23): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (24): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (25): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (26): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (27): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (28): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (29): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (30): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (31): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(1280, 2560, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (post_norms): ModuleList( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (2): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (3): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (4): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (meta_head_1): Sequential( (0): Linear(in_features=4, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (meta_head_2): Sequential( (0): Linear(in_features=3, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (dcnv3_head_x4): Sequential( (0): Conv2d(2560, 4096, kernel_size=(1, 1), stride=(1, 1)) (1): PixelShuffle(upscale_factor=2) ) (dcnv3_head_x3): Conv2d(1280, 1024, kernel_size=(1, 1), stride=(1, 1)) (clip_projector): AttentionPoolingBlock( (norm1_q): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_k): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_v): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (cross_attn): CrossAttention( (q): Linear(in_features=1024, out_features=1024, bias=False) (k): Linear(in_features=1024, out_features=1024, bias=False) (v): Linear(in_features=1024, out_features=1024, bias=False) (attn_drop): Dropout(p=0.0, inplace=False) (proj): Linear(in_features=1024, out_features=768, bias=True) (proj_drop): Dropout(p=0.0, inplace=False) ) (drop_path): Identity() ) (fc_norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True) (head): Linear(in_features=768, out_features=8142, bias=True) (meta_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True) (meta_head): Linear(in_features=128, out_features=8142, bias=True) ) [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 172): INFO Using native Torch AMP. Training in mixed precision. [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 184): INFO using fp16_compress_hook! [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 193): INFO number of params: 1087305976 [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 223): INFO no checkpoint found in work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384, ignoring auto resume [2022-12-19 03:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 134): INFO ==============> Loading weight /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth for fine-tuning...... [2022-12-19 03:51:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 265): WARNING _IncompatibleKeys(missing_keys=['meta_head_1.0.weight', 'meta_head_1.0.bias', 'meta_head_1.2.weight', 'meta_head_1.2.bias', 'meta_head_1.3.norm_fn1.weight', 'meta_head_1.3.norm_fn1.bias', 'meta_head_1.3.norm_fn2.weight', 'meta_head_1.3.norm_fn2.bias', 'meta_head_1.3.w1.weight', 'meta_head_1.3.w1.bias', 'meta_head_1.3.w2.weight', 'meta_head_1.3.w2.bias', 'meta_head_2.0.weight', 'meta_head_2.0.bias', 'meta_head_2.2.weight', 'meta_head_2.2.bias', 'meta_head_2.3.norm_fn1.weight', 'meta_head_2.3.norm_fn1.bias', 'meta_head_2.3.norm_fn2.weight', 'meta_head_2.3.norm_fn2.bias', 'meta_head_2.3.w1.weight', 'meta_head_2.3.w1.bias', 'meta_head_2.3.w2.weight', 'meta_head_2.3.w2.bias', 'fc_norm.weight', 'fc_norm.bias', 'head.weight', 'head.bias', 'meta_norm.weight', 'meta_norm.bias', 'meta_head.weight', 'meta_head.bias'], unexpected_keys=[]) [2022-12-19 03:51:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 270): INFO => loaded successfully /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth [2022-12-19 03:51:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [0/85] Time 2.713 (2.713) Loss 9.1234 (9.1234) Acc@1 0.000 (0.000) Acc@5 0.347 (0.347) Mem 10077MB [2022-12-19 03:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [10/85] Time 0.294 (0.514) Loss 9.2359 (9.1860) Acc@1 0.000 (0.000) Acc@5 0.347 (0.095) Mem 10078MB [2022-12-19 03:51:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [20/85] Time 0.296 (0.409) Loss 9.1868 (9.1871) Acc@1 0.000 (0.000) Acc@5 0.000 (0.050) Mem 10078MB [2022-12-19 03:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [30/85] Time 0.300 (0.373) Loss 9.1972 (9.1854) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [40/85] Time 0.294 (0.354) Loss 9.2195 (9.1834) Acc@1 0.000 (0.000) Acc@5 0.000 (0.042) Mem 10078MB [2022-12-19 03:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [50/85] Time 0.296 (0.343) Loss 9.2102 (9.1878) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [60/85] Time 0.304 (0.336) Loss 9.2229 (9.1891) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:51:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [70/85] Time 0.296 (0.330) Loss 9.2167 (9.1914) Acc@1 0.000 (0.000) Acc@5 0.000 (0.029) Mem 10078MB [2022-12-19 03:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [80/85] Time 0.295 (0.326) Loss 9.1560 (9.1899) Acc@1 0.000 (0.000) Acc@5 0.000 (0.039) Mem 10078MB [2022-12-19 03:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 531): INFO * Acc@1 0.004 Acc@5 0.049 [2022-12-19 03:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 240): INFO Accuracy of the network on the 24426 test images: 0.0% [2022-12-19 03:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 263): INFO Start training [2022-12-19 03:51:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][0/1519] eta 4:50:06 lr 0.000034 time 11.4593 (11.4593) model_time 10.1401 (10.1401) loss 4.3789 (4.3789) grad_norm 0.0000 (0.0000/0.0000) mem 55588MB [2022-12-19 03:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][10/1519] eta 0:53:41 lr 0.000034 time 0.9355 (2.1347) model_time 0.9353 (2.0145) loss 4.5876 (4.5924) grad_norm nan (0.0000/0.0000) mem 59766MB [2022-12-19 03:52:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][20/1519] eta 0:40:28 lr 0.000034 time 0.9711 (1.6203) model_time 0.9709 (1.5571) loss 4.4827 (4.6015) grad_norm 16.8583 (15.3601/4.5636) mem 68057MB [2022-12-19 03:52:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][30/1519] eta 0:35:21 lr 0.000034 time 0.9344 (1.4247) model_time 0.9343 (1.3818) loss 4.4217 (4.6006) grad_norm 7.3432 (12.0417/4.4994) mem 68057MB [2022-12-19 03:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][40/1519] eta 0:32:38 lr 0.000034 time 0.9637 (1.3241) model_time 0.9636 (1.2916) loss 4.4071 (4.5721) grad_norm 7.6644 (10.2631/4.2365) mem 68057MB [2022-12-19 03:52:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][50/1519] eta 0:30:53 lr 0.000034 time 0.9389 (1.2614) model_time 0.9387 (1.2352) loss 4.3216 (4.5376) grad_norm 6.2467 (9.2662/3.9675) mem 68057MB [2022-12-19 03:52:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][60/1519] eta 0:29:40 lr 0.000034 time 0.9495 (1.2202) model_time 0.9494 (1.1983) loss 4.2967 (4.5051) grad_norm 6.1161 (8.5033/3.8242) mem 68057MB [2022-12-19 03:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 619): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 03:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 622): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: true IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 03:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 135): INFO Creating model:dcnv3_5_new3_with_meta/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 139): INFO DCNv3_5_new3_WithMeta( (patch_embed): ConvTokenizer( (conv1): Conv2d(3, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm1): Sequential( (0): to_channels_last() (1): LayerNorm((160,), eps=1e-06, elementwise_affine=True) (2): to_channels_first() ) (act): GELU() (conv2): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm2): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (pos_drop): Dropout(p=0.0, inplace=False) (levels): ModuleList( (0): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): Identity() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (1): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(640, 1280, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (2): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (6): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (7): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (8): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (9): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (10): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (11): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (12): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (13): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (14): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (15): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (16): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (17): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (18): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (19): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (20): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (21): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (22): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (23): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (24): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (25): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (26): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (27): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (28): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (29): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (30): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (31): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(1280, 2560, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (post_norms): ModuleList( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (2): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (3): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (4): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (meta_head_1): Sequential( (0): Linear(in_features=4, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (meta_head_2): Sequential( (0): Linear(in_features=3, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (dcnv3_head_x4): Sequential( (0): Conv2d(2560, 4096, kernel_size=(1, 1), stride=(1, 1)) (1): PixelShuffle(upscale_factor=2) ) (dcnv3_head_x3): Conv2d(1280, 1024, kernel_size=(1, 1), stride=(1, 1)) (clip_projector): AttentionPoolingBlock( (norm1_q): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_k): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_v): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (cross_attn): CrossAttention( (q): Linear(in_features=1024, out_features=1024, bias=False) (k): Linear(in_features=1024, out_features=1024, bias=False) (v): Linear(in_features=1024, out_features=1024, bias=False) (attn_drop): Dropout(p=0.0, inplace=False) (proj): Linear(in_features=1024, out_features=768, bias=True) (proj_drop): Dropout(p=0.0, inplace=False) ) (drop_path): Identity() ) (fc_norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True) (head): Linear(in_features=768, out_features=8142, bias=True) (meta_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True) (meta_head): Linear(in_features=128, out_features=8142, bias=True) ) [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 172): INFO Using native Torch AMP. Training in mixed precision. [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 184): INFO using fp16_compress_hook! [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 193): INFO number of params: 1087305976 [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 223): INFO no checkpoint found in work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384, ignoring auto resume [2022-12-19 03:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 134): INFO ==============> Loading weight /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth for fine-tuning...... [2022-12-19 03:56:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 265): WARNING _IncompatibleKeys(missing_keys=['meta_head_1.0.weight', 'meta_head_1.0.bias', 'meta_head_1.2.weight', 'meta_head_1.2.bias', 'meta_head_1.3.norm_fn1.weight', 'meta_head_1.3.norm_fn1.bias', 'meta_head_1.3.norm_fn2.weight', 'meta_head_1.3.norm_fn2.bias', 'meta_head_1.3.w1.weight', 'meta_head_1.3.w1.bias', 'meta_head_1.3.w2.weight', 'meta_head_1.3.w2.bias', 'meta_head_2.0.weight', 'meta_head_2.0.bias', 'meta_head_2.2.weight', 'meta_head_2.2.bias', 'meta_head_2.3.norm_fn1.weight', 'meta_head_2.3.norm_fn1.bias', 'meta_head_2.3.norm_fn2.weight', 'meta_head_2.3.norm_fn2.bias', 'meta_head_2.3.w1.weight', 'meta_head_2.3.w1.bias', 'meta_head_2.3.w2.weight', 'meta_head_2.3.w2.bias', 'fc_norm.weight', 'fc_norm.bias', 'head.weight', 'head.bias', 'meta_norm.weight', 'meta_norm.bias', 'meta_head.weight', 'meta_head.bias'], unexpected_keys=[]) [2022-12-19 03:56:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 270): INFO => loaded successfully /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth [2022-12-19 03:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [0/85] Time 4.817 (4.817) Loss 9.1234 (9.1234) Acc@1 0.000 (0.000) Acc@5 0.347 (0.347) Mem 10076MB [2022-12-19 03:56:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [10/85] Time 0.293 (0.709) Loss 9.2359 (9.1860) Acc@1 0.000 (0.000) Acc@5 0.347 (0.095) Mem 10078MB [2022-12-19 03:56:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [20/85] Time 0.297 (0.513) Loss 9.1868 (9.1871) Acc@1 0.000 (0.000) Acc@5 0.000 (0.050) Mem 10078MB [2022-12-19 03:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [30/85] Time 0.294 (0.443) Loss 9.1972 (9.1854) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [40/85] Time 0.298 (0.408) Loss 9.2195 (9.1834) Acc@1 0.000 (0.000) Acc@5 0.000 (0.042) Mem 10078MB [2022-12-19 03:56:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [50/85] Time 0.296 (0.386) Loss 9.2102 (9.1878) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [60/85] Time 0.298 (0.372) Loss 9.2229 (9.1891) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10078MB [2022-12-19 03:56:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [70/85] Time 0.300 (0.362) Loss 9.2167 (9.1914) Acc@1 0.000 (0.000) Acc@5 0.000 (0.029) Mem 10078MB [2022-12-19 03:56:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 520): INFO Test: [80/85] Time 0.303 (0.354) Loss 9.1560 (9.1899) Acc@1 0.000 (0.000) Acc@5 0.000 (0.039) Mem 10078MB [2022-12-19 03:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 531): INFO * Acc@1 0.004 Acc@5 0.049 [2022-12-19 03:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 240): INFO Accuracy of the network on the 24426 test images: 0.0% [2022-12-19 03:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 263): INFO Start training [2022-12-19 03:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][0/1519] eta 4:59:36 lr 0.000034 time 11.8344 (11.8344) model_time 10.2629 (10.2629) loss 4.5951 (4.5951) grad_norm 0.0000 (0.0000/0.0000) mem 55615MB [2022-12-19 03:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][10/1519] eta 0:56:29 lr 0.000034 time 1.5330 (2.2463) model_time 1.5328 (2.1031) loss 4.4952 (4.5551) grad_norm 11.6257 (11.6257/0.0000) mem 68105MB [2022-12-19 03:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 438): INFO Train: [0/100][20/1519] eta 0:41:05 lr 0.000034 time 0.9504 (1.6448) model_time 0.9502 (1.5695) loss 4.4840 (4.5534) grad_norm 12.2387 (12.3982/2.0602) mem 68105MB [2022-12-19 03:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 621): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 03:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 624): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: true IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 04:00:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 621): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 04:00:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 624): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: true IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 04:01:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 137): INFO Creating model:dcnv3_5_new3_with_meta/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 [2022-12-19 04:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 621): INFO Full config saved to work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/config.json [2022-12-19 04:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 624): INFO AMP_OPT_LEVEL: O1 AMP_TYPE: float16 AUG: AUTO_AUGMENT: rand-m9-mstd0.5-inc1 COLOR_JITTER: 0.4 CUTMIX: 0.0 CUTMIX_MINMAX: null MEAN: - 0.485 - 0.456 - 0.406 MIXUP: 0.0 MIXUP_MODE: batch MIXUP_PROB: 1.0 MIXUP_SWITCH_PROB: 0.5 RANDOM_RESIZED_CROP: false RECOUNT: 1 REMODE: pixel REPROB: 0.25 STD: - 0.229 - 0.224 - 0.225 BASE: - '' DATA: BATCH_SIZE: 12 CACHE_MODE: part DATASET: inat18 DATA_PATH: /mnt/petrelfs/share_data/chenzhe1/inat/ IMG_ON_MEMORY: false IMG_SIZE: 384 INTERPOLATION: bicubic NUM_WORKERS: 8 PIN_MEMORY: true ZIP_MODE: false EVAL_22K_TO_1K: false EVAL_FREQ: 1 EVAL_MODE: false LOCAL_RANK: 0 MODEL: CONVNEXT: CHANNELS: - 192 - 384 - 768 - 1536 IN_CHANNELS: 3 LAYERS: - 3 - 3 - 27 - 3 LAYER_SCALE: 1.0e-06 DAT: {} DCNV3: ACT_LAYER: GELU CLS_SCALE: 1.5 CPE_NORM_LAYER: BN DCNV3_CORE: MSDeformAttnGrid4_softmax_ab DEFORM_PADDING: true DEFORM_POINTS: 8 DEPTHS: - 6 - 6 - 32 - 6 DILATION_RATES: - 1 DW_KS: 5 EMBED_DIM: 320 LAYER_SCALE: null LEVEL2_POST_NORM: false LEVEL2_POST_NORM_BLOCKS_IDS: - 5 - 11 - 17 - 23 - 29 MLP_RATIO: 4.0 NORM_LAYER: LN NUM_HEADS: - 10 - 20 - 40 - 80 OFFSETS_SCALER: 1.0 REMOVE_PRE_NORM: false DROP_PATH_RATE: 0.6 DROP_PATH_TYPE: linear DROP_RATE: 0.0 FD_DCNV3: CLIP_MODEL: ViT-B/16 CLIP_PRETRAINED: /mnt/petrelfs/share_data/wangwenhai/clip/ViT-B-16.pt STUDENT_OUT_INDICES: - 3 WITH_FD: false LABEL_SMOOTHING: 0.1 NAME: dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 NUM_CLASSES: 1000 PRETRAINED: /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth REPLKNET: CHANNELS: - 192 - 384 - 768 - 1536 DW_RATIO: 1.0 FFN_RATIO: 4.0 IN_CHANNELS: 3 LARGE_KERNEL_SIZES: - 31 - 29 - 27 - 13 LAYERS: - 2 - 2 - 18 - 2 NORM_FEAT: false OUT_INDICES: null SMALL_KERNEL: 5 SMALL_KERNEL_MERGED: false USE_SYNC_BN: true RESUME: '' SWIN: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 QKV_BIAS: true QK_SCALE: null WINDOW_SIZE: 7 SWIN_MLP: APE: false DEPTHS: - 2 - 2 - 6 - 2 EMBED_DIM: 96 IN_CHANS: 3 MLP_RATIO: 4.0 NUM_HEADS: - 3 - 6 - 12 - 24 PATCH_NORM: true PATCH_SIZE: 4 WINDOW_SIZE: 7 TYPE: dcnv3_5_new3_with_meta OUTPUT: work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 PRINT_FREQ: 10 SAVE_CKPT_NUM: 1 SAVE_FREQ: 1 SEED: 0 TAG: default TEST: CROP: true SEQUENTIAL: false TTA: false THROUGHPUT_MODE: false TRAIN: ACCUMULATION_STEPS: 2 AUTO_RESUME: true BASE_LR: 3.375e-05 CLIP_GRAD: 5.0 EMA: DECAY: 0.9998 ENABLE: false EMS: ENABLE: false EPOCHS: 100 LR_LAYER_DECAY: true LR_LAYER_DECAY_RATIO: 0.8 LR_SCHEDULER: DECAY_EPOCHS: 30 DECAY_RATE: 0.1 NAME: cosine MIN_LR: 3.375e-07 OFFSET_SPECIAL_LR: false OPTIMIZER: BETAS: - 0.9 - 0.999 DCN_LR_MUL: null EPS: 1.0e-08 FREEZE_BACKBONE: null HEAD_WD_100: false MOMENTUM: 0.9 NAME: adamw USE_ZERO: false RAND_INIT_FT_HEAD: true START_EPOCH: 0 USE_CHECKPOINT: false WARMUP_EPOCHS: 0 WARMUP_LR: 3.3749999999999995e-08 WEIGHT_DECAY: 1.0e-08 [2022-12-19 04:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 137): INFO Creating model:dcnv3_5_new3_with_meta/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384 [2022-12-19 04:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 141): INFO DCNv3_5_new3_WithMeta( (patch_embed): ConvTokenizer( (conv1): Conv2d(3, 160, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm1): Sequential( (0): to_channels_last() (1): LayerNorm((160,), eps=1e-06, elementwise_affine=True) (2): to_channels_first() ) (act): GELU() (conv2): Conv2d(160, 320, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) (norm2): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (pos_drop): Dropout(p=0.0, inplace=False) (levels): ModuleList( (0): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): Identity() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(320, 320, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=320) (1): Sequential( (0): to_channels_last() (1): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=320, out_features=160, bias=True) (attention_weights): Linear(in_features=320, out_features=80, bias=True) (value_proj): Linear(in_features=320, out_features=320, bias=True) (output_proj): Linear(in_features=320, out_features=320, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=320, out_features=1280, bias=True) (act): GELU() (fc2): Linear(in_features=1280, out_features=320, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((320,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(320, 640, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (1): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(640, 640, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=640) (1): Sequential( (0): to_channels_last() (1): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=640, out_features=320, bias=True) (attention_weights): Linear(in_features=640, out_features=160, bias=True) (value_proj): Linear(in_features=640, out_features=640, bias=True) (output_proj): Linear(in_features=640, out_features=640, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=640, out_features=2560, bias=True) (act): GELU() (fc2): Linear(in_features=2560, out_features=640, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((640,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(640, 1280, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (2): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (6): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (7): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (8): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (9): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (10): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (11): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (12): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (13): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (14): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (15): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (16): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (17): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (18): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (19): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (20): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (21): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (22): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (23): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (24): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (25): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (26): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (27): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (28): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (29): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (30): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (31): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(1280, 1280, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=1280) (1): Sequential( (0): to_channels_last() (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=1280, out_features=640, bias=True) (attention_weights): Linear(in_features=1280, out_features=320, bias=True) (value_proj): Linear(in_features=1280, out_features=1280, bias=True) (output_proj): Linear(in_features=1280, out_features=1280, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=1280, out_features=5120, bias=True) (act): GELU() (fc2): Linear(in_features=5120, out_features=1280, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) (downsample): ConvDownsampler( (conv): Conv2d(1280, 2560, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (norm): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (post_norms): ModuleList( (0): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (1): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (2): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (3): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) (4): LayerNorm((1280,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Block( (blocks): ModuleList( (0): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (1): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (2): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (3): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (4): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) (5): DCNv3Layer( (norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (attn): MSDeformAttnGrid4_softmax_ab( (dw_conv): Sequential( (0): Conv2d(2560, 2560, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2), groups=2560) (1): Sequential( (0): to_channels_last() (1): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (2): GELU() ) (sampling_offsets): Linear(in_features=2560, out_features=1280, bias=True) (attention_weights): Linear(in_features=2560, out_features=640, bias=True) (value_proj): Linear(in_features=2560, out_features=2560, bias=True) (output_proj): Linear(in_features=2560, out_features=2560, bias=True) (center_feature_scale_module): CenterFeatureScaleModule() ) (drop_path): DropPath() (norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (mlp): MLP( (fc1): Linear(in_features=2560, out_features=10240, bias=True) (act): GELU() (fc2): Linear(in_features=10240, out_features=2560, bias=True) (drop): Dropout(p=0.0, inplace=False) ) (res_post_norm1): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) (res_post_norm2): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (norm): Sequential( (0): LayerNorm((2560,), eps=1e-06, elementwise_affine=True) ) ) ) (meta_head_1): Sequential( (0): Linear(in_features=4, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (meta_head_2): Sequential( (0): Linear(in_features=3, out_features=64, bias=True) (1): ReLU(inplace=True) (2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (3): MetaEncoder( (nonlin1): ReLU(inplace=True) (nonlin2): ReLU(inplace=True) (norm_fn1): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (norm_fn2): LayerNorm((64,), eps=1e-05, elementwise_affine=True) (w1): Linear(in_features=64, out_features=64, bias=True) (w2): Linear(in_features=64, out_features=64, bias=True) ) ) (dcnv3_head_x4): Sequential( (0): Conv2d(2560, 4096, kernel_size=(1, 1), stride=(1, 1)) (1): PixelShuffle(upscale_factor=2) ) (dcnv3_head_x3): Conv2d(1280, 1024, kernel_size=(1, 1), stride=(1, 1)) (clip_projector): AttentionPoolingBlock( (norm1_q): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_k): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (norm1_v): LayerNorm((1024,), eps=1e-06, elementwise_affine=True) (cross_attn): CrossAttention( (q): Linear(in_features=1024, out_features=1024, bias=False) (k): Linear(in_features=1024, out_features=1024, bias=False) (v): Linear(in_features=1024, out_features=1024, bias=False) (attn_drop): Dropout(p=0.0, inplace=False) (proj): Linear(in_features=1024, out_features=768, bias=True) (proj_drop): Dropout(p=0.0, inplace=False) ) (drop_path): Identity() ) (fc_norm): LayerNorm((768,), eps=1e-06, elementwise_affine=True) (head): Linear(in_features=768, out_features=8142, bias=True) (meta_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True) (meta_head): Linear(in_features=128, out_features=8142, bias=True) ) [2022-12-19 04:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 174): INFO Using native Torch AMP. Training in mixed precision. [2022-12-19 04:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 186): INFO using fp16_compress_hook! [2022-12-19 04:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 195): INFO number of params: 1087305976 [2022-12-19 04:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 225): INFO no checkpoint found in work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384, ignoring auto resume [2022-12-19 04:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 134): INFO ==============> Loading weight /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth for fine-tuning...... [2022-12-19 04:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 265): WARNING _IncompatibleKeys(missing_keys=['meta_head_1.0.weight', 'meta_head_1.0.bias', 'meta_head_1.2.weight', 'meta_head_1.2.bias', 'meta_head_1.3.norm_fn1.weight', 'meta_head_1.3.norm_fn1.bias', 'meta_head_1.3.norm_fn2.weight', 'meta_head_1.3.norm_fn2.bias', 'meta_head_1.3.w1.weight', 'meta_head_1.3.w1.bias', 'meta_head_1.3.w2.weight', 'meta_head_1.3.w2.bias', 'meta_head_2.0.weight', 'meta_head_2.0.bias', 'meta_head_2.2.weight', 'meta_head_2.2.bias', 'meta_head_2.3.norm_fn1.weight', 'meta_head_2.3.norm_fn1.bias', 'meta_head_2.3.norm_fn2.weight', 'meta_head_2.3.norm_fn2.bias', 'meta_head_2.3.w1.weight', 'meta_head_2.3.w1.bias', 'meta_head_2.3.w2.weight', 'meta_head_2.3.w2.bias', 'fc_norm.weight', 'fc_norm.bias', 'head.weight', 'head.bias', 'meta_norm.weight', 'meta_norm.bias', 'meta_head.weight', 'meta_head.bias'], unexpected_keys=[]) [2022-12-19 04:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 270): INFO => loaded successfully /mnt/petrelfs/wangwenhai/workspace_swj/work_dir/sim/configs/pjlab/imagenet22k_30ep_init_from_pretrain_mm_1_bs304x120_resume384_lr3e_5_320gpus_lr1e_5.sh/checkpoint-latest.pth [2022-12-19 04:05:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 6.424 (6.424) Loss 9.1234 (9.1234) Acc@1 0.000 (0.000) Acc@5 0.347 (0.347) Mem 10077MB [2022-12-19 04:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.852) Loss 9.2359 (9.1860) Acc@1 0.000 (0.000) Acc@5 0.347 (0.095) Mem 10077MB [2022-12-19 04:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.587) Loss 9.1868 (9.1871) Acc@1 0.000 (0.000) Acc@5 0.000 (0.050) Mem 10077MB [2022-12-19 04:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.300 (0.494) Loss 9.1972 (9.1854) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10077MB [2022-12-19 04:05:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.293 (0.445) Loss 9.2195 (9.1834) Acc@1 0.000 (0.000) Acc@5 0.000 (0.042) Mem 10077MB [2022-12-19 04:05:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.416) Loss 9.2102 (9.1878) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10077MB [2022-12-19 04:05:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.397) Loss 9.2229 (9.1891) Acc@1 0.000 (0.000) Acc@5 0.000 (0.034) Mem 10077MB [2022-12-19 04:05:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.292 (0.382) Loss 9.2167 (9.1914) Acc@1 0.000 (0.000) Acc@5 0.000 (0.029) Mem 10077MB [2022-12-19 04:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.372) Loss 9.1560 (9.1899) Acc@1 0.000 (0.000) Acc@5 0.000 (0.039) Mem 10077MB [2022-12-19 04:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 533): INFO * Acc@1 0.004 Acc@5 0.049 [2022-12-19 04:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 242): INFO Accuracy of the network on the 24426 test images: 0.0% [2022-12-19 04:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 265): INFO Start training [2022-12-19 04:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][0/1519] eta 4:50:52 lr 0.000034 time 11.4893 (11.4893) model_time 10.2592 (10.2592) loss 4.3789 (4.3789) grad_norm 0.0000 (0.0000/0.0000) mem 55617MB [2022-12-19 04:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][10/1519] eta 0:53:47 lr 0.000034 time 0.9304 (2.1390) model_time 0.9301 (2.0268) loss 4.5876 (4.5924) grad_norm nan (0.0000/0.0000) mem 59815MB [2022-12-19 04:06:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][20/1519] eta 0:40:28 lr 0.000034 time 0.9340 (1.6198) model_time 0.9339 (1.5609) loss 4.4803 (4.6014) grad_norm 18.9634 (16.7294/5.4857) mem 68106MB [2022-12-19 04:06:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][30/1519] eta 0:35:18 lr 0.000034 time 0.9365 (1.4230) model_time 0.9363 (1.3829) loss 4.4180 (4.6000) grad_norm 10.3237 (12.7700/4.7697) mem 68106MB [2022-12-19 04:06:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][40/1519] eta 0:32:32 lr 0.000034 time 0.9376 (1.3201) model_time 0.9374 (1.2898) loss 4.4562 (4.5744) grad_norm 7.0805 (11.4023/4.3666) mem 68106MB [2022-12-19 04:06:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][50/1519] eta 0:30:47 lr 0.000034 time 0.9320 (1.2577) model_time 0.9319 (1.2332) loss 4.3212 (4.5389) grad_norm 4.3133 (10.1872/4.2958) mem 68106MB [2022-12-19 04:06:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][60/1519] eta 0:29:34 lr 0.000034 time 0.9386 (1.2162) model_time 0.9385 (1.1957) loss 4.3046 (4.5073) grad_norm 4.8809 (9.1621/4.2812) mem 68106MB [2022-12-19 04:07:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][70/1519] eta 0:28:38 lr 0.000034 time 0.9331 (1.1863) model_time 0.9329 (1.1686) loss 4.4930 (4.4863) grad_norm 4.6871 (8.4595/4.1751) mem 68106MB [2022-12-19 04:07:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][80/1519] eta 0:27:54 lr 0.000034 time 0.9406 (1.1640) model_time 0.9405 (1.1484) loss 4.3352 (4.4578) grad_norm 4.0429 (7.7000/4.2473) mem 68106MB [2022-12-19 04:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][90/1519] eta 0:27:18 lr 0.000034 time 0.9265 (1.1467) model_time 0.9264 (1.1328) loss 4.4081 (4.4342) grad_norm 3.9532 (7.1683/4.1881) mem 68106MB [2022-12-19 04:07:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][100/1519] eta 0:26:46 lr 0.000034 time 0.9137 (1.1322) model_time 0.9136 (1.1196) loss 4.1249 (4.4188) grad_norm 4.0739 (6.7945/4.0712) mem 68106MB [2022-12-19 04:07:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][110/1519] eta 0:26:18 lr 0.000034 time 0.9320 (1.1206) model_time 0.9319 (1.1092) loss 4.2114 (4.4057) grad_norm 4.7155 (6.4902/3.9586) mem 68106MB [2022-12-19 04:07:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][120/1519] eta 0:25:53 lr 0.000034 time 0.9236 (1.1105) model_time 0.9235 (1.1000) loss 3.9557 (4.3833) grad_norm 3.2857 (6.1921/3.8795) mem 68106MB [2022-12-19 04:08:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][130/1519] eta 0:25:30 lr 0.000034 time 0.9312 (1.1020) model_time 0.9311 (1.0923) loss 4.1290 (4.3606) grad_norm 6.7553 (6.0520/3.7521) mem 68106MB [2022-12-19 04:08:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][140/1519] eta 0:25:09 lr 0.000034 time 0.9350 (1.0948) model_time 0.9344 (1.0857) loss 4.3867 (4.3417) grad_norm 7.0367 (6.0413/3.6103) mem 68106MB [2022-12-19 04:08:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][150/1519] eta 0:24:52 lr 0.000034 time 0.9305 (1.0901) model_time 0.9302 (1.0816) loss 4.0547 (4.3232) grad_norm 4.2463 (5.9545/3.4995) mem 68106MB [2022-12-19 04:08:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][160/1519] eta 0:24:33 lr 0.000034 time 0.9336 (1.0845) model_time 0.9335 (1.0765) loss 3.8518 (4.3050) grad_norm 5.9939 (5.9805/3.4617) mem 68106MB [2022-12-19 04:08:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][170/1519] eta 0:24:16 lr 0.000034 time 0.9260 (1.0798) model_time 0.9259 (1.0722) loss 4.0804 (4.2941) grad_norm 8.8784 (6.1187/3.4045) mem 68106MB [2022-12-19 04:08:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][180/1519] eta 0:23:59 lr 0.000034 time 0.9311 (1.0753) model_time 0.9309 (1.0682) loss 3.9327 (4.2798) grad_norm 10.7152 (6.2641/3.4091) mem 68106MB [2022-12-19 04:09:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][190/1519] eta 0:23:43 lr 0.000034 time 0.9313 (1.0713) model_time 0.9312 (1.0645) loss 3.6821 (4.2606) grad_norm 9.7234 (6.2942/3.3449) mem 68106MB [2022-12-19 04:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][200/1519] eta 0:23:28 lr 0.000034 time 0.9345 (1.0679) model_time 0.9344 (1.0615) loss 4.1679 (4.2460) grad_norm 9.5027 (6.4511/3.3782) mem 68106MB [2022-12-19 04:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][210/1519] eta 0:23:13 lr 0.000034 time 0.9555 (1.0648) model_time 0.9554 (1.0586) loss 4.2822 (4.2288) grad_norm 7.9687 (6.6356/3.4640) mem 68106MB [2022-12-19 04:09:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][220/1519] eta 0:22:59 lr 0.000034 time 0.9373 (1.0619) model_time 0.9371 (1.0560) loss 3.9777 (4.2115) grad_norm 6.0042 (6.6316/3.3876) mem 68106MB [2022-12-19 04:09:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][230/1519] eta 0:22:45 lr 0.000034 time 0.9357 (1.0594) model_time 0.9355 (1.0537) loss 3.6713 (4.1913) grad_norm 6.2192 (6.6388/3.3153) mem 68106MB [2022-12-19 04:09:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][240/1519] eta 0:22:31 lr 0.000034 time 0.9418 (1.0571) model_time 0.9417 (1.0516) loss 3.8449 (4.1720) grad_norm 9.3656 (6.7555/3.3052) mem 68106MB [2022-12-19 04:10:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][250/1519] eta 0:22:18 lr 0.000034 time 0.9334 (1.0549) model_time 0.9333 (1.0497) loss 3.7198 (4.1564) grad_norm 12.8102 (6.8337/3.2985) mem 68106MB [2022-12-19 04:10:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][260/1519] eta 0:22:05 lr 0.000034 time 0.9369 (1.0529) model_time 0.9368 (1.0478) loss 3.6858 (4.1398) grad_norm 4.8136 (6.9286/3.4135) mem 68106MB [2022-12-19 04:10:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][270/1519] eta 0:21:52 lr 0.000034 time 0.9284 (1.0509) model_time 0.9282 (1.0460) loss 3.7863 (4.1255) grad_norm 9.6576 (7.0061/3.3977) mem 68106MB [2022-12-19 04:10:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][280/1519] eta 0:21:39 lr 0.000034 time 0.9338 (1.0492) model_time 0.9337 (1.0444) loss 3.7282 (4.1073) grad_norm 11.0451 (7.0475/3.3620) mem 68106MB [2022-12-19 04:10:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][290/1519] eta 0:21:27 lr 0.000034 time 0.9389 (1.0477) model_time 0.9388 (1.0431) loss 3.1866 (4.0933) grad_norm 8.5308 (7.1502/3.4400) mem 68106MB [2022-12-19 04:10:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][300/1519] eta 0:21:15 lr 0.000034 time 0.9298 (1.0461) model_time 0.9297 (1.0417) loss 3.2381 (4.0774) grad_norm 6.3836 (7.2241/3.4376) mem 68106MB [2022-12-19 04:11:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][310/1519] eta 0:21:03 lr 0.000034 time 0.9472 (1.0447) model_time 0.9471 (1.0404) loss 3.7180 (4.0648) grad_norm 7.2454 (7.2557/3.4081) mem 68106MB [2022-12-19 04:11:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][320/1519] eta 0:20:50 lr 0.000034 time 0.9354 (1.0433) model_time 0.9353 (1.0391) loss 3.3764 (4.0508) grad_norm 5.0045 (7.2836/3.3808) mem 68106MB [2022-12-19 04:11:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][330/1519] eta 0:20:38 lr 0.000034 time 0.9333 (1.0419) model_time 0.9331 (1.0379) loss 3.5080 (4.0389) grad_norm 13.6091 (7.4933/3.5565) mem 68106MB [2022-12-19 04:11:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][340/1519] eta 0:20:27 lr 0.000034 time 0.9348 (1.0408) model_time 0.9347 (1.0368) loss 3.2874 (4.0261) grad_norm 4.9561 (7.6424/3.8207) mem 68106MB [2022-12-19 04:11:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][350/1519] eta 0:20:15 lr 0.000034 time 0.9351 (1.0395) model_time 0.9349 (1.0357) loss 3.2230 (4.0096) grad_norm 7.6933 (7.6672/3.7837) mem 68106MB [2022-12-19 04:11:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][360/1519] eta 0:20:03 lr 0.000034 time 0.9249 (1.0384) model_time 0.9248 (1.0346) loss 3.6493 (3.9971) grad_norm 5.8453 (7.6853/3.7457) mem 68106MB [2022-12-19 04:12:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][370/1519] eta 0:19:51 lr 0.000034 time 0.9293 (1.0373) model_time 0.9291 (1.0336) loss 3.4751 (3.9857) grad_norm 6.3276 (7.6585/3.7046) mem 68106MB [2022-12-19 04:12:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][380/1519] eta 0:19:40 lr 0.000034 time 0.9365 (1.0363) model_time 0.9364 (1.0327) loss 3.2363 (3.9697) grad_norm 5.3549 (7.6423/3.6579) mem 68106MB [2022-12-19 04:12:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][390/1519] eta 0:19:28 lr 0.000034 time 0.9332 (1.0354) model_time 0.9331 (1.0319) loss 3.6925 (3.9567) grad_norm 7.3931 (7.5993/3.6227) mem 68106MB [2022-12-19 04:12:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][400/1519] eta 0:19:17 lr 0.000034 time 0.9013 (1.0347) model_time 0.9011 (1.0313) loss 3.5422 (3.9446) grad_norm 8.5122 (7.6314/3.6061) mem 68106MB [2022-12-19 04:12:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][410/1519] eta 0:19:06 lr 0.000034 time 0.9308 (1.0339) model_time 0.9306 (1.0306) loss 3.3326 (3.9321) grad_norm 5.8365 (7.6155/3.5657) mem 68106MB [2022-12-19 04:12:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][420/1519] eta 0:18:55 lr 0.000034 time 0.9299 (1.0331) model_time 0.9297 (1.0299) loss 3.3596 (3.9191) grad_norm 5.3844 (7.6421/3.5583) mem 68106MB [2022-12-19 04:13:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][430/1519] eta 0:18:44 lr 0.000034 time 0.9327 (1.0323) model_time 0.9325 (1.0292) loss 3.5128 (3.9050) grad_norm 10.3085 (7.6695/3.5298) mem 68106MB [2022-12-19 04:13:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][440/1519] eta 0:18:33 lr 0.000034 time 0.9369 (1.0316) model_time 0.9368 (1.0285) loss 3.5494 (3.8959) grad_norm 8.9577 (7.6573/3.4942) mem 68106MB [2022-12-19 04:13:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][450/1519] eta 0:18:22 lr 0.000034 time 0.9354 (1.0309) model_time 0.9353 (1.0279) loss 3.4108 (3.8849) grad_norm 5.7778 (7.6690/3.4738) mem 68106MB [2022-12-19 04:13:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][460/1519] eta 0:18:11 lr 0.000034 time 0.9368 (1.0308) model_time 0.9367 (1.0278) loss 3.6306 (3.8719) grad_norm 8.0449 (7.6700/3.4381) mem 68106MB [2022-12-19 04:13:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][470/1519] eta 0:18:00 lr 0.000034 time 0.9453 (1.0302) model_time 0.9451 (1.0272) loss 3.5905 (3.8636) grad_norm 7.3146 (7.6791/3.4071) mem 68106MB [2022-12-19 04:13:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][480/1519] eta 0:17:49 lr 0.000034 time 0.9376 (1.0296) model_time 0.9374 (1.0267) loss 3.5266 (3.8560) grad_norm 10.8516 (7.7450/3.4476) mem 68106MB [2022-12-19 04:14:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][490/1519] eta 0:17:38 lr 0.000034 time 0.9329 (1.0290) model_time 0.9328 (1.0262) loss 2.7585 (3.8420) grad_norm 10.9342 (7.7768/3.4222) mem 68106MB [2022-12-19 04:14:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][500/1519] eta 0:17:27 lr 0.000034 time 0.9368 (1.0283) model_time 0.9365 (1.0255) loss 3.3798 (3.8282) grad_norm 16.4882 (7.8434/3.4558) mem 68106MB [2022-12-19 04:14:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][510/1519] eta 0:17:17 lr 0.000034 time 0.9416 (1.0278) model_time 0.9415 (1.0251) loss 3.0315 (3.8168) grad_norm 15.4600 (7.8586/3.4601) mem 68106MB [2022-12-19 04:14:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][520/1519] eta 0:17:06 lr 0.000034 time 0.9362 (1.0273) model_time 0.9360 (1.0246) loss 3.0334 (3.8028) grad_norm 7.2623 (7.9115/3.5112) mem 68106MB [2022-12-19 04:14:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][530/1519] eta 0:16:55 lr 0.000034 time 0.9369 (1.0269) model_time 0.9367 (1.0242) loss 2.9698 (3.7932) grad_norm 12.2221 (7.9613/3.5827) mem 68106MB [2022-12-19 04:14:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][540/1519] eta 0:16:44 lr 0.000034 time 0.9417 (1.0265) model_time 0.9416 (1.0238) loss 3.3375 (3.7821) grad_norm 9.7208 (8.0471/3.7464) mem 68106MB [2022-12-19 04:15:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][550/1519] eta 0:16:34 lr 0.000034 time 0.9389 (1.0260) model_time 0.9387 (1.0235) loss 3.5074 (3.7722) grad_norm 23.2414 (8.0908/3.8298) mem 68106MB [2022-12-19 04:15:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][560/1519] eta 0:16:23 lr 0.000034 time 0.9375 (1.0256) model_time 0.9374 (1.0231) loss 3.0303 (3.7644) grad_norm 14.7906 (8.1702/3.8580) mem 68106MB [2022-12-19 04:15:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][570/1519] eta 0:16:12 lr 0.000034 time 0.9400 (1.0252) model_time 0.9398 (1.0227) loss 3.3231 (3.7558) grad_norm 13.7286 (8.1949/3.8489) mem 68106MB [2022-12-19 04:15:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][580/1519] eta 0:16:02 lr 0.000034 time 0.9345 (1.0248) model_time 0.9344 (1.0223) loss 3.1066 (3.7473) grad_norm 14.0711 (8.2242/3.8393) mem 68106MB [2022-12-19 04:15:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][590/1519] eta 0:15:51 lr 0.000034 time 0.9375 (1.0245) model_time 0.9373 (1.0220) loss 3.3623 (3.7352) grad_norm 12.7093 (8.2954/3.8521) mem 68106MB [2022-12-19 04:15:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][600/1519] eta 0:15:41 lr 0.000034 time 0.9426 (1.0241) model_time 0.9424 (1.0217) loss 3.6351 (3.7251) grad_norm 12.1528 (8.3007/3.8285) mem 68106MB [2022-12-19 04:16:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][610/1519] eta 0:15:30 lr 0.000034 time 0.9312 (1.0238) model_time 0.9310 (1.0214) loss 3.5875 (3.7152) grad_norm 19.9488 (8.3565/3.8640) mem 68106MB [2022-12-19 04:16:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][620/1519] eta 0:15:20 lr 0.000034 time 0.9328 (1.0234) model_time 0.9326 (1.0211) loss 3.2634 (3.7040) grad_norm 14.6993 (8.3854/3.8557) mem 68106MB [2022-12-19 04:16:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][630/1519] eta 0:15:09 lr 0.000034 time 0.9347 (1.0230) model_time 0.9345 (1.0207) loss 3.6129 (3.6950) grad_norm 7.9327 (8.3167/3.8081) mem 68106MB [2022-12-19 04:16:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][640/1519] eta 0:14:58 lr 0.000034 time 0.9389 (1.0228) model_time 0.9386 (1.0205) loss 2.9619 (3.6855) grad_norm 6.5206 (8.3222/3.8503) mem 68106MB [2022-12-19 04:16:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][650/1519] eta 0:14:48 lr 0.000034 time 0.9345 (1.0224) model_time 0.9344 (1.0202) loss 2.8955 (3.6741) grad_norm 9.7969 (8.3114/3.8590) mem 68106MB [2022-12-19 04:16:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][660/1519] eta 0:14:37 lr 0.000034 time 0.9371 (1.0221) model_time 0.9369 (1.0199) loss 3.5222 (3.6650) grad_norm 12.1502 (8.4055/3.8972) mem 68106MB [2022-12-19 04:17:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][670/1519] eta 0:14:27 lr 0.000034 time 0.9353 (1.0218) model_time 0.9351 (1.0196) loss 3.1067 (3.6575) grad_norm 6.4890 (8.4490/3.8808) mem 68106MB [2022-12-19 04:17:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][680/1519] eta 0:14:17 lr 0.000034 time 0.9312 (1.0215) model_time 0.9310 (1.0193) loss 3.0295 (3.6488) grad_norm 12.1853 (8.5376/3.8604) mem 68106MB [2022-12-19 04:17:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][690/1519] eta 0:14:06 lr 0.000034 time 0.9363 (1.0213) model_time 0.9361 (1.0191) loss 2.9738 (3.6404) grad_norm 6.9115 (8.6615/3.8661) mem 68106MB [2022-12-19 04:17:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][700/1519] eta 0:13:56 lr 0.000034 time 0.9313 (1.0210) model_time 0.9312 (1.0188) loss 2.7373 (3.6312) grad_norm 10.1174 (8.7929/3.9055) mem 68106MB [2022-12-19 04:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][710/1519] eta 0:13:45 lr 0.000034 time 0.9330 (1.0207) model_time 0.9327 (1.0186) loss 2.8218 (3.6218) grad_norm 12.4151 (8.9162/3.8651) mem 68106MB [2022-12-19 04:17:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][720/1519] eta 0:13:35 lr 0.000034 time 0.9351 (1.0205) model_time 0.9348 (1.0184) loss 2.6438 (3.6150) grad_norm 7.4614 (9.0300/3.8276) mem 68106MB [2022-12-19 04:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][730/1519] eta 0:13:24 lr 0.000034 time 0.9399 (1.0202) model_time 0.9398 (1.0182) loss 3.3590 (3.6093) grad_norm 11.6573 (9.1476/3.7851) mem 68106MB [2022-12-19 04:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][740/1519] eta 0:13:14 lr 0.000034 time 0.9250 (1.0200) model_time 0.9249 (1.0179) loss 3.3883 (3.5993) grad_norm 18.1869 (9.2925/3.8807) mem 68106MB [2022-12-19 04:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][750/1519] eta 0:13:04 lr 0.000034 time 0.9331 (1.0197) model_time 0.9329 (1.0177) loss 2.2473 (3.5901) grad_norm 8.5277 (9.4499/3.9399) mem 68106MB [2022-12-19 04:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][760/1519] eta 0:12:53 lr 0.000034 time 0.9379 (1.0194) model_time 0.9378 (1.0174) loss 2.8815 (3.5844) grad_norm 9.0221 (9.5140/3.9010) mem 68106MB [2022-12-19 04:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][770/1519] eta 0:12:43 lr 0.000034 time 0.9361 (1.0195) model_time 0.9360 (1.0176) loss 2.2309 (3.5752) grad_norm 17.6953 (9.6006/3.9312) mem 68106MB [2022-12-19 04:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][780/1519] eta 0:12:33 lr 0.000034 time 0.9335 (1.0193) model_time 0.9333 (1.0173) loss 2.6688 (3.5694) grad_norm 9.9809 (9.7193/4.0110) mem 68106MB [2022-12-19 04:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][790/1519] eta 0:12:22 lr 0.000034 time 0.9335 (1.0190) model_time 0.9333 (1.0171) loss 2.6912 (3.5617) grad_norm 9.3037 (9.7481/3.9897) mem 68106MB [2022-12-19 04:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][800/1519] eta 0:12:12 lr 0.000034 time 0.9355 (1.0188) model_time 0.9354 (1.0169) loss 2.8182 (3.5534) grad_norm 21.7736 (9.8256/4.0845) mem 68106MB [2022-12-19 04:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][810/1519] eta 0:12:02 lr 0.000034 time 0.9356 (1.0186) model_time 0.9354 (1.0167) loss 3.0387 (3.5440) grad_norm 8.2283 (9.8320/4.0781) mem 68106MB [2022-12-19 04:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][820/1519] eta 0:11:51 lr 0.000034 time 0.9333 (1.0183) model_time 0.9332 (1.0165) loss 2.9272 (3.5357) grad_norm 5.7327 (9.8566/4.0754) mem 68106MB [2022-12-19 04:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][830/1519] eta 0:11:41 lr 0.000034 time 0.9380 (1.0181) model_time 0.9378 (1.0163) loss 3.3893 (3.5298) grad_norm 16.6164 (9.9883/4.1228) mem 68106MB [2022-12-19 04:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][840/1519] eta 0:11:31 lr 0.000034 time 0.9286 (1.0179) model_time 0.9285 (1.0161) loss 2.5354 (3.5231) grad_norm 16.0413 (10.0323/4.1450) mem 68106MB [2022-12-19 04:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][850/1519] eta 0:11:20 lr 0.000034 time 0.9310 (1.0177) model_time 0.9308 (1.0159) loss 2.6902 (3.5157) grad_norm 12.0485 (10.0774/4.1324) mem 68106MB [2022-12-19 04:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][860/1519] eta 0:11:10 lr 0.000034 time 0.9348 (1.0175) model_time 0.9346 (1.0157) loss 2.2147 (3.5066) grad_norm 10.8400 (10.1673/4.2018) mem 68106MB [2022-12-19 04:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][870/1519] eta 0:11:00 lr 0.000034 time 0.9335 (1.0173) model_time 0.9333 (1.0155) loss 2.7601 (3.4993) grad_norm 26.1358 (10.2735/4.3062) mem 68106MB [2022-12-19 04:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][880/1519] eta 0:10:49 lr 0.000034 time 0.9374 (1.0171) model_time 0.9372 (1.0153) loss 2.2957 (3.4918) grad_norm 11.0394 (10.2939/4.3058) mem 68106MB [2022-12-19 04:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][890/1519] eta 0:10:39 lr 0.000034 time 0.9348 (1.0169) model_time 0.9346 (1.0151) loss 2.7501 (3.4853) grad_norm 14.2544 (10.2895/4.2856) mem 68106MB [2022-12-19 04:20:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][900/1519] eta 0:10:29 lr 0.000034 time 0.9314 (1.0167) model_time 0.9313 (1.0149) loss 2.5675 (3.4806) grad_norm 12.2632 (10.3816/4.3963) mem 68106MB [2022-12-19 04:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][910/1519] eta 0:10:19 lr 0.000034 time 0.9401 (1.0165) model_time 0.9399 (1.0148) loss 2.9248 (3.4720) grad_norm 12.8158 (10.4573/4.3951) mem 68106MB [2022-12-19 04:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][920/1519] eta 0:10:08 lr 0.000034 time 0.9366 (1.0163) model_time 0.9364 (1.0146) loss 2.5267 (3.4643) grad_norm 11.1791 (10.5430/4.4335) mem 68106MB [2022-12-19 04:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][930/1519] eta 0:09:58 lr 0.000034 time 0.9347 (1.0161) model_time 0.9346 (1.0144) loss 2.3780 (3.4583) grad_norm 31.2708 (10.6845/4.7194) mem 68106MB [2022-12-19 04:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][940/1519] eta 0:09:48 lr 0.000034 time 0.9381 (1.0159) model_time 0.9379 (1.0143) loss 2.6396 (3.4514) grad_norm 6.3173 (10.6545/4.6633) mem 68106MB [2022-12-19 04:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][950/1519] eta 0:09:38 lr 0.000034 time 0.9325 (1.0160) model_time 0.9323 (1.0144) loss 3.2247 (3.4454) grad_norm 20.2706 (10.7073/4.6792) mem 68106MB [2022-12-19 04:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][960/1519] eta 0:09:27 lr 0.000034 time 0.9312 (1.0159) model_time 0.9310 (1.0142) loss 2.6707 (3.4378) grad_norm 12.9728 (10.7407/4.6847) mem 68106MB [2022-12-19 04:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][970/1519] eta 0:09:17 lr 0.000034 time 0.9352 (1.0157) model_time 0.9350 (1.0141) loss 2.5469 (3.4315) grad_norm 10.8879 (10.8115/4.6575) mem 68106MB [2022-12-19 04:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][980/1519] eta 0:09:07 lr 0.000034 time 0.9237 (1.0155) model_time 0.9236 (1.0139) loss 2.6722 (3.4252) grad_norm 9.8454 (10.9694/4.8029) mem 68106MB [2022-12-19 04:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][990/1519] eta 0:08:57 lr 0.000034 time 0.9350 (1.0154) model_time 0.9348 (1.0138) loss 3.1160 (3.4214) grad_norm 10.7782 (11.0816/4.7660) mem 68106MB [2022-12-19 04:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1000/1519] eta 0:08:46 lr 0.000034 time 0.9354 (1.0152) model_time 0.9353 (1.0136) loss 2.6251 (3.4134) grad_norm 14.9928 (11.0828/4.7751) mem 68106MB [2022-12-19 04:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1010/1519] eta 0:08:36 lr 0.000034 time 0.9352 (1.0151) model_time 0.9350 (1.0135) loss 2.9209 (3.4064) grad_norm 21.2470 (11.1934/4.7982) mem 68106MB [2022-12-19 04:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1020/1519] eta 0:08:26 lr 0.000034 time 0.9300 (1.0149) model_time 0.9298 (1.0133) loss 2.7961 (3.4013) grad_norm 13.8234 (11.2331/4.7875) mem 68106MB [2022-12-19 04:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1030/1519] eta 0:08:16 lr 0.000034 time 0.9309 (1.0148) model_time 0.9308 (1.0133) loss 2.1718 (3.3947) grad_norm 12.7847 (11.2860/4.7880) mem 68106MB [2022-12-19 04:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1040/1519] eta 0:08:06 lr 0.000034 time 0.9243 (1.0147) model_time 0.9241 (1.0131) loss 3.3160 (3.3890) grad_norm 11.3724 (11.4220/5.0217) mem 68106MB [2022-12-19 04:23:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1050/1519] eta 0:07:55 lr 0.000034 time 0.9304 (1.0146) model_time 0.9303 (1.0130) loss 3.0242 (3.3849) grad_norm 16.1796 (11.4671/5.0180) mem 68106MB [2022-12-19 04:23:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1060/1519] eta 0:07:45 lr 0.000034 time 0.9298 (1.0144) model_time 0.9297 (1.0129) loss 2.3017 (3.3789) grad_norm 13.3475 (11.5141/5.0010) mem 68106MB [2022-12-19 04:23:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1070/1519] eta 0:07:35 lr 0.000034 time 0.9355 (1.0143) model_time 0.9354 (1.0127) loss 3.0856 (3.3738) grad_norm 16.8867 (11.6331/5.0359) mem 68106MB [2022-12-19 04:23:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1080/1519] eta 0:07:25 lr 0.000034 time 0.9316 (1.0141) model_time 0.9315 (1.0126) loss 2.9221 (3.3665) grad_norm 9.0497 (11.6623/5.0172) mem 68106MB [2022-12-19 04:24:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1090/1519] eta 0:07:15 lr 0.000034 time 0.9281 (1.0140) model_time 0.9278 (1.0125) loss 2.4636 (3.3624) grad_norm 14.9328 (11.7139/5.0205) mem 68106MB [2022-12-19 04:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1100/1519] eta 0:07:04 lr 0.000034 time 0.9329 (1.0139) model_time 0.9327 (1.0124) loss 2.6973 (3.3559) grad_norm 8.6201 (11.6655/5.0251) mem 68106MB [2022-12-19 04:24:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1110/1519] eta 0:06:54 lr 0.000034 time 0.9348 (1.0137) model_time 0.9346 (1.0123) loss 2.7626 (3.3493) grad_norm 6.9164 (11.7009/5.0656) mem 68106MB [2022-12-19 04:24:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1120/1519] eta 0:06:44 lr 0.000034 time 0.9340 (1.0136) model_time 0.9338 (1.0122) loss 2.2502 (3.3425) grad_norm 8.5699 (11.6815/5.0356) mem 68106MB [2022-12-19 04:24:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1130/1519] eta 0:06:34 lr 0.000034 time 0.9342 (1.0135) model_time 0.9340 (1.0120) loss 2.2805 (3.3357) grad_norm 11.3846 (11.6707/4.9873) mem 68106MB [2022-12-19 04:24:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1140/1519] eta 0:06:24 lr 0.000034 time 0.9350 (1.0134) model_time 0.9348 (1.0119) loss 3.2930 (3.3328) grad_norm 12.7655 (11.6403/4.9006) mem 68106MB [2022-12-19 04:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1150/1519] eta 0:06:13 lr 0.000034 time 0.9360 (1.0132) model_time 0.9356 (1.0118) loss 2.6410 (3.3274) grad_norm 13.4347 (11.6778/4.8387) mem 68106MB [2022-12-19 04:25:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1160/1519] eta 0:06:03 lr 0.000034 time 0.9375 (1.0131) model_time 0.9373 (1.0117) loss 2.2533 (3.3210) grad_norm 15.3789 (11.6601/4.8358) mem 68106MB [2022-12-19 04:25:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1170/1519] eta 0:05:53 lr 0.000034 time 0.9285 (1.0130) model_time 0.9284 (1.0116) loss 2.9649 (3.3170) grad_norm 7.8965 (11.6473/4.8341) mem 68106MB [2022-12-19 04:25:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1180/1519] eta 0:05:43 lr 0.000034 time 0.9357 (1.0129) model_time 0.9354 (1.0115) loss 2.6439 (3.3118) grad_norm 9.0632 (11.6933/4.8256) mem 68106MB [2022-12-19 04:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1190/1519] eta 0:05:33 lr 0.000034 time 0.9354 (1.0128) model_time 0.9352 (1.0114) loss 2.5665 (3.3063) grad_norm 8.0365 (11.6562/4.8281) mem 68106MB [2022-12-19 04:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1200/1519] eta 0:05:23 lr 0.000034 time 0.9408 (1.0127) model_time 0.9407 (1.0113) loss 2.7334 (3.3009) grad_norm 8.8099 (11.7042/4.8122) mem 68106MB [2022-12-19 04:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1210/1519] eta 0:05:12 lr 0.000034 time 0.9390 (1.0128) model_time 0.9388 (1.0114) loss 2.9864 (3.2955) grad_norm 17.5908 (11.6999/4.8100) mem 68106MB [2022-12-19 04:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1220/1519] eta 0:05:02 lr 0.000034 time 0.9290 (1.0127) model_time 0.9289 (1.0113) loss 2.4652 (3.2891) grad_norm 10.1472 (11.6815/4.7806) mem 68106MB [2022-12-19 04:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1230/1519] eta 0:04:52 lr 0.000034 time 0.9344 (1.0126) model_time 0.9343 (1.0112) loss 2.2650 (3.2832) grad_norm 15.9086 (11.7748/4.8199) mem 68106MB [2022-12-19 04:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1240/1519] eta 0:04:42 lr 0.000034 time 0.9314 (1.0125) model_time 0.9312 (1.0112) loss 2.8400 (3.2785) grad_norm 13.2467 (11.8782/4.8615) mem 68106MB [2022-12-19 04:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1250/1519] eta 0:04:32 lr 0.000034 time 0.9352 (1.0124) model_time 0.9350 (1.0111) loss 2.4085 (3.2735) grad_norm 12.8280 (11.9248/4.8275) mem 68106MB [2022-12-19 04:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1260/1519] eta 0:04:22 lr 0.000034 time 0.9161 (1.0125) model_time 0.9159 (1.0111) loss 3.3303 (3.2676) grad_norm 12.4244 (11.9273/4.8055) mem 68106MB [2022-12-19 04:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1270/1519] eta 0:04:12 lr 0.000034 time 0.9310 (1.0124) model_time 0.9308 (1.0110) loss 3.0732 (3.2619) grad_norm 12.5265 (11.9918/4.7857) mem 68106MB [2022-12-19 04:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1280/1519] eta 0:04:01 lr 0.000034 time 0.9323 (1.0123) model_time 0.9321 (1.0109) loss 2.6770 (3.2578) grad_norm 14.9021 (12.0624/4.7624) mem 68106MB [2022-12-19 04:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1290/1519] eta 0:03:51 lr 0.000034 time 0.9329 (1.0122) model_time 0.9328 (1.0109) loss 2.2347 (3.2521) grad_norm 12.0903 (12.0712/4.7457) mem 68106MB [2022-12-19 04:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1300/1519] eta 0:03:41 lr 0.000034 time 0.9326 (1.0121) model_time 0.9324 (1.0108) loss 2.8019 (3.2477) grad_norm 12.3764 (12.0436/4.7035) mem 68106MB [2022-12-19 04:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1310/1519] eta 0:03:31 lr 0.000034 time 0.9370 (1.0120) model_time 0.9369 (1.0107) loss 2.7946 (3.2431) grad_norm 8.7859 (12.0540/4.7720) mem 68106MB [2022-12-19 04:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1320/1519] eta 0:03:21 lr 0.000034 time 0.9308 (1.0119) model_time 0.9307 (1.0106) loss 3.2323 (3.2382) grad_norm 25.1528 (12.0724/4.8352) mem 68106MB [2022-12-19 04:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1330/1519] eta 0:03:11 lr 0.000034 time 0.9309 (1.0118) model_time 0.9307 (1.0105) loss 2.5975 (3.2336) grad_norm 7.5670 (12.0787/4.8470) mem 68106MB [2022-12-19 04:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1340/1519] eta 0:03:01 lr 0.000034 time 0.9356 (1.0117) model_time 0.9355 (1.0104) loss 3.0123 (3.2300) grad_norm 7.6197 (11.9830/4.8105) mem 68106MB [2022-12-19 04:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1350/1519] eta 0:02:50 lr 0.000034 time 0.9325 (1.0116) model_time 0.9323 (1.0104) loss 2.3275 (3.2247) grad_norm 6.8833 (11.9648/4.8292) mem 68106MB [2022-12-19 04:28:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1360/1519] eta 0:02:40 lr 0.000034 time 0.9346 (1.0115) model_time 0.9345 (1.0103) loss 2.4774 (3.2191) grad_norm 11.6141 (11.9746/4.8297) mem 68106MB [2022-12-19 04:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1370/1519] eta 0:02:30 lr 0.000034 time 0.9393 (1.0114) model_time 0.9392 (1.0102) loss 3.0026 (3.2138) grad_norm 18.5443 (12.0188/4.8974) mem 68106MB [2022-12-19 04:28:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1380/1519] eta 0:02:20 lr 0.000034 time 0.9332 (1.0114) model_time 0.9331 (1.0101) loss 2.5254 (3.2084) grad_norm 5.9737 (11.9172/4.8842) mem 68106MB [2022-12-19 04:29:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1390/1519] eta 0:02:10 lr 0.000034 time 0.9342 (1.0113) model_time 0.9341 (1.0100) loss 3.1783 (3.2037) grad_norm 11.3564 (11.9455/4.8804) mem 68106MB [2022-12-19 04:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1400/1519] eta 0:02:00 lr 0.000034 time 0.9333 (1.0112) model_time 0.9331 (1.0099) loss 2.3211 (3.1999) grad_norm 13.1799 (11.8732/4.8268) mem 68106MB [2022-12-19 04:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1410/1519] eta 0:01:50 lr 0.000034 time 0.9319 (1.0111) model_time 0.9318 (1.0099) loss 1.8694 (3.1956) grad_norm 7.8725 (11.8554/4.8316) mem 68106MB [2022-12-19 04:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1420/1519] eta 0:01:40 lr 0.000034 time 0.9355 (1.0110) model_time 0.9353 (1.0098) loss 2.6280 (3.1922) grad_norm 21.0445 (11.9456/4.8455) mem 68106MB [2022-12-19 04:29:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1430/1519] eta 0:01:29 lr 0.000034 time 0.9360 (1.0109) model_time 0.9359 (1.0097) loss 2.8012 (3.1862) grad_norm 13.6429 (11.8775/4.8197) mem 68106MB [2022-12-19 04:29:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1440/1519] eta 0:01:19 lr 0.000034 time 0.9319 (1.0109) model_time 0.9318 (1.0096) loss 3.0631 (3.1820) grad_norm 8.2670 (11.8617/4.8305) mem 68106MB [2022-12-19 04:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1450/1519] eta 0:01:09 lr 0.000034 time 0.9322 (1.0108) model_time 0.9321 (1.0096) loss 2.4557 (3.1771) grad_norm 8.4034 (11.8606/4.8323) mem 68106MB [2022-12-19 04:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1460/1519] eta 0:00:59 lr 0.000034 time 0.9260 (1.0107) model_time 0.9257 (1.0095) loss 2.4144 (3.1723) grad_norm 7.2949 (11.7289/4.7899) mem 68106MB [2022-12-19 04:30:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1470/1519] eta 0:00:49 lr 0.000034 time 0.9340 (1.0106) model_time 0.9339 (1.0094) loss 1.8807 (3.1673) grad_norm 12.1680 (11.6824/4.7008) mem 68106MB [2022-12-19 04:30:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1480/1519] eta 0:00:39 lr 0.000034 time 0.9259 (1.0105) model_time 0.9257 (1.0094) loss 3.1711 (3.1631) grad_norm 8.2680 (11.7362/4.7178) mem 68106MB [2022-12-19 04:30:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1490/1519] eta 0:00:29 lr 0.000034 time 0.9312 (1.0105) model_time 0.9310 (1.0093) loss 2.3311 (3.1580) grad_norm 19.5894 (11.7920/4.7428) mem 68106MB [2022-12-19 04:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1500/1519] eta 0:00:19 lr 0.000034 time 0.9334 (1.0104) model_time 0.9333 (1.0092) loss 2.1836 (3.1545) grad_norm 16.3777 (11.7823/4.7077) mem 68106MB [2022-12-19 04:31:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [0/100][1510/1519] eta 0:00:09 lr 0.000034 time 0.9298 (1.0104) model_time 0.9297 (1.0092) loss 2.0148 (3.1488) grad_norm 9.4058 (11.7298/4.7153) mem 68106MB [2022-12-19 04:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 0 training takes 0:25:34 [2022-12-19 04:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_0.pth saving...... [2022-12-19 04:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_0.pth saved !!! [2022-12-19 04:31:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.688 (0.688) Loss 6.4980 (6.4980) Acc@1 9.375 (9.375) Acc@5 19.097 (19.097) Mem 68106MB [2022-12-19 04:31:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.333) Loss 6.9615 (6.7404) Acc@1 3.472 (7.828) Acc@5 12.847 (16.604) Mem 68106MB [2022-12-19 04:31:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.315) Loss 6.6287 (6.7462) Acc@1 7.639 (7.639) Acc@5 16.667 (16.237) Mem 68106MB [2022-12-19 04:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.310) Loss 6.7813 (6.7489) Acc@1 6.597 (7.460) Acc@5 13.194 (16.006) Mem 68106MB [2022-12-19 04:31:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.302 (0.308) Loss 6.6407 (6.7384) Acc@1 6.944 (7.613) Acc@5 15.278 (16.048) Mem 68106MB [2022-12-19 04:31:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 6.6983 (6.7367) Acc@1 8.333 (7.503) Acc@5 17.361 (16.122) Mem 68106MB [2022-12-19 04:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 6.8785 (6.7432) Acc@1 6.597 (7.417) Acc@5 13.194 (15.961) Mem 68106MB [2022-12-19 04:31:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 6.9886 (6.7500) Acc@1 4.167 (7.414) Acc@5 14.583 (15.953) Mem 68106MB [2022-12-19 04:32:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.303) Loss 6.6048 (6.7453) Acc@1 8.333 (7.407) Acc@5 17.361 (15.989) Mem 68106MB [2022-12-19 04:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:0] * Acc@1 7.404 Acc@5 15.930 [2022-12-19 04:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 7.4% [2022-12-19 04:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 04:32:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 04:32:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 7.40% [2022-12-19 04:32:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][0/1519] eta 0:35:54 lr 0.000034 time 1.4181 (1.4181) model_time 0.9737 (0.9737) loss 2.0603 (2.0603) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 04:32:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][10/1519] eta 0:26:03 lr 0.000034 time 0.9259 (1.0360) model_time 0.9258 (0.9952) loss 2.5880 (2.3196) grad_norm 10.3029 (15.0113/8.0188) mem 68106MB [2022-12-19 04:32:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][20/1519] eta 0:25:30 lr 0.000034 time 0.9299 (1.0207) model_time 0.9298 (0.9992) loss 2.8630 (2.5113) grad_norm 18.8057 (12.4046/7.0629) mem 68106MB [2022-12-19 04:32:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][30/1519] eta 0:25:09 lr 0.000034 time 0.9325 (1.0137) model_time 0.9324 (0.9990) loss 1.3341 (2.5255) grad_norm 9.4504 (12.1772/6.2625) mem 68106MB [2022-12-19 04:33:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][40/1519] eta 0:24:53 lr 0.000034 time 0.9331 (1.0098) model_time 0.9330 (0.9986) loss 2.0872 (2.4938) grad_norm 6.9404 (11.4946/5.6797) mem 68106MB [2022-12-19 04:33:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][50/1519] eta 0:24:40 lr 0.000034 time 0.9320 (1.0075) model_time 0.9319 (0.9984) loss 2.5009 (2.4932) grad_norm 12.2089 (11.5230/5.1691) mem 68106MB [2022-12-19 04:33:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][60/1519] eta 0:24:31 lr 0.000034 time 0.9348 (1.0087) model_time 0.9346 (1.0010) loss 2.5108 (2.5040) grad_norm 11.6784 (11.8300/5.4626) mem 68106MB [2022-12-19 04:33:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][70/1519] eta 0:24:19 lr 0.000034 time 0.9260 (1.0074) model_time 0.9258 (1.0007) loss 2.8540 (2.5085) grad_norm 11.3770 (11.7629/5.2092) mem 68106MB [2022-12-19 04:33:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][80/1519] eta 0:24:08 lr 0.000034 time 0.9358 (1.0063) model_time 0.9357 (1.0004) loss 2.3894 (2.4812) grad_norm 8.2024 (11.3995/4.9862) mem 68106MB [2022-12-19 04:33:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][90/1519] eta 0:23:56 lr 0.000034 time 0.9334 (1.0056) model_time 0.9333 (1.0003) loss 2.6182 (2.4727) grad_norm 11.0466 (11.3838/4.8457) mem 68106MB [2022-12-19 04:34:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][100/1519] eta 0:23:46 lr 0.000034 time 0.9337 (1.0051) model_time 0.9336 (1.0003) loss 2.5678 (2.4873) grad_norm 10.5189 (11.2796/4.6861) mem 68106MB [2022-12-19 04:34:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][110/1519] eta 0:23:35 lr 0.000034 time 0.9306 (1.0045) model_time 0.9305 (1.0001) loss 2.1967 (2.4678) grad_norm 10.2559 (11.1749/4.5144) mem 68106MB [2022-12-19 04:34:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][120/1519] eta 0:23:24 lr 0.000034 time 0.9323 (1.0041) model_time 0.9320 (1.0000) loss 2.5323 (2.4836) grad_norm 10.2067 (11.0971/4.4327) mem 68106MB [2022-12-19 04:34:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][130/1519] eta 0:23:14 lr 0.000034 time 0.9357 (1.0039) model_time 0.9355 (1.0002) loss 1.9943 (2.4688) grad_norm 6.6123 (11.0671/4.4421) mem 68106MB [2022-12-19 04:34:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][140/1519] eta 0:23:04 lr 0.000034 time 0.9364 (1.0037) model_time 0.9361 (1.0001) loss 2.7903 (2.4790) grad_norm 8.2096 (10.9508/4.3948) mem 68106MB [2022-12-19 04:34:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][150/1519] eta 0:22:53 lr 0.000034 time 0.9402 (1.0034) model_time 0.9400 (1.0000) loss 2.3081 (2.4714) grad_norm 12.0572 (10.9300/4.3806) mem 68106MB [2022-12-19 04:35:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][160/1519] eta 0:22:43 lr 0.000034 time 0.9377 (1.0031) model_time 0.9376 (0.9999) loss 2.3032 (2.4576) grad_norm 9.2397 (10.8115/4.3124) mem 68106MB [2022-12-19 04:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][170/1519] eta 0:22:33 lr 0.000034 time 0.9320 (1.0030) model_time 0.9319 (1.0001) loss 2.3422 (2.4442) grad_norm 7.0934 (11.3293/5.2621) mem 68106MB [2022-12-19 04:35:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][180/1519] eta 0:22:22 lr 0.000034 time 0.9305 (1.0029) model_time 0.9304 (1.0001) loss 2.1803 (2.4368) grad_norm 10.9116 (11.5561/5.5496) mem 68106MB [2022-12-19 04:35:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][190/1519] eta 0:22:12 lr 0.000034 time 0.9362 (1.0028) model_time 0.9361 (1.0001) loss 2.2249 (2.4317) grad_norm 7.3274 (11.5643/5.4405) mem 68106MB [2022-12-19 04:35:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][200/1519] eta 0:22:02 lr 0.000034 time 0.9254 (1.0025) model_time 0.9253 (1.0000) loss 2.3964 (2.4326) grad_norm 8.0579 (11.4624/5.3386) mem 68106MB [2022-12-19 04:35:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][210/1519] eta 0:21:52 lr 0.000034 time 0.9324 (1.0024) model_time 0.9323 (0.9999) loss 2.1357 (2.4245) grad_norm 11.2070 (11.5341/5.2499) mem 68106MB [2022-12-19 04:36:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][220/1519] eta 0:21:41 lr 0.000034 time 0.9374 (1.0022) model_time 0.9373 (0.9998) loss 2.3779 (2.4255) grad_norm 6.7601 (11.4301/5.1714) mem 68106MB [2022-12-19 04:36:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][230/1519] eta 0:21:31 lr 0.000034 time 0.9411 (1.0021) model_time 0.9409 (0.9999) loss 2.3483 (2.4214) grad_norm 9.9735 (11.4580/5.1437) mem 68106MB [2022-12-19 04:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][240/1519] eta 0:21:22 lr 0.000034 time 0.9381 (1.0029) model_time 0.9379 (1.0007) loss 2.4833 (2.4189) grad_norm 9.9397 (11.4635/5.0934) mem 68106MB [2022-12-19 04:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][250/1519] eta 0:21:12 lr 0.000034 time 0.9331 (1.0028) model_time 0.9329 (1.0007) loss 2.0754 (2.4159) grad_norm 11.8307 (11.4130/5.0090) mem 68106MB [2022-12-19 04:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][260/1519] eta 0:21:02 lr 0.000034 time 0.9374 (1.0027) model_time 0.9372 (1.0007) loss 2.3307 (2.4099) grad_norm 12.2150 (11.5257/4.9686) mem 68106MB [2022-12-19 04:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][270/1519] eta 0:20:52 lr 0.000034 time 0.9365 (1.0026) model_time 0.9363 (1.0006) loss 2.3609 (2.4048) grad_norm 10.7324 (11.5766/4.9453) mem 68106MB [2022-12-19 04:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][280/1519] eta 0:20:42 lr 0.000034 time 0.9347 (1.0026) model_time 0.9346 (1.0007) loss 1.8508 (2.4058) grad_norm 10.9935 (11.5680/4.8971) mem 68106MB [2022-12-19 04:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][290/1519] eta 0:20:32 lr 0.000034 time 0.9422 (1.0026) model_time 0.9421 (1.0007) loss 2.3080 (2.3994) grad_norm 24.3676 (11.6869/5.0384) mem 68106MB [2022-12-19 04:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][300/1519] eta 0:20:22 lr 0.000034 time 0.9466 (1.0026) model_time 0.9464 (1.0008) loss 1.8041 (2.3949) grad_norm 9.5676 (11.6486/4.9710) mem 68106MB [2022-12-19 04:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][310/1519] eta 0:20:12 lr 0.000034 time 0.9319 (1.0025) model_time 0.9318 (1.0007) loss 2.3176 (2.4018) grad_norm 20.9589 (11.6944/4.9534) mem 68106MB [2022-12-19 04:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][320/1519] eta 0:20:01 lr 0.000034 time 0.9306 (1.0025) model_time 0.9305 (1.0007) loss 2.4786 (2.4057) grad_norm 7.2708 (11.6945/4.9312) mem 68106MB [2022-12-19 04:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][330/1519] eta 0:19:51 lr 0.000034 time 0.9305 (1.0025) model_time 0.9304 (1.0008) loss 2.3127 (2.4036) grad_norm 14.8532 (11.6749/4.8961) mem 68106MB [2022-12-19 04:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][340/1519] eta 0:19:41 lr 0.000034 time 0.9333 (1.0024) model_time 0.9332 (1.0007) loss 2.0652 (2.3934) grad_norm 8.8081 (11.5668/4.8730) mem 68106MB [2022-12-19 04:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][350/1519] eta 0:19:31 lr 0.000034 time 0.9370 (1.0024) model_time 0.9368 (1.0007) loss 2.4049 (2.3888) grad_norm 6.4942 (11.4413/4.8631) mem 68106MB [2022-12-19 04:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][360/1519] eta 0:19:21 lr 0.000034 time 0.9425 (1.0024) model_time 0.9424 (1.0008) loss 2.6404 (2.3867) grad_norm 6.2936 (11.4116/4.8229) mem 68106MB [2022-12-19 04:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][370/1519] eta 0:19:11 lr 0.000034 time 0.9282 (1.0023) model_time 0.9280 (1.0007) loss 2.3087 (2.3808) grad_norm 10.6804 (11.3772/4.7923) mem 68106MB [2022-12-19 04:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][380/1519] eta 0:19:01 lr 0.000034 time 0.9362 (1.0022) model_time 0.9361 (1.0007) loss 1.9750 (2.3780) grad_norm 7.1381 (11.2790/4.7702) mem 68106MB [2022-12-19 04:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][390/1519] eta 0:18:51 lr 0.000034 time 0.9255 (1.0021) model_time 0.9254 (1.0006) loss 1.8822 (2.3766) grad_norm 7.7744 (11.1931/4.7567) mem 68106MB [2022-12-19 04:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][400/1519] eta 0:18:41 lr 0.000034 time 0.9328 (1.0021) model_time 0.9326 (1.0006) loss 2.6438 (2.3748) grad_norm 11.0122 (11.2277/4.7403) mem 68106MB [2022-12-19 04:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][410/1519] eta 0:18:31 lr 0.000034 time 0.9353 (1.0020) model_time 0.9351 (1.0006) loss 2.5136 (2.3736) grad_norm 6.7595 (11.2299/4.7311) mem 68106MB [2022-12-19 04:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][420/1519] eta 0:18:21 lr 0.000034 time 0.9297 (1.0020) model_time 0.9295 (1.0006) loss 2.7528 (2.3751) grad_norm 12.2450 (11.2941/4.7288) mem 68106MB [2022-12-19 04:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][430/1519] eta 0:18:11 lr 0.000034 time 0.9407 (1.0020) model_time 0.9405 (1.0006) loss 2.1429 (2.3732) grad_norm 16.6799 (11.4024/4.8839) mem 68106MB [2022-12-19 04:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][440/1519] eta 0:18:01 lr 0.000034 time 0.9473 (1.0019) model_time 0.9472 (1.0006) loss 2.1604 (2.3692) grad_norm 8.5748 (11.4398/4.8870) mem 68106MB [2022-12-19 04:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][450/1519] eta 0:17:51 lr 0.000034 time 0.9377 (1.0019) model_time 0.9376 (1.0006) loss 1.5000 (2.3650) grad_norm 14.6820 (11.4216/4.8551) mem 68106MB [2022-12-19 04:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][460/1519] eta 0:17:40 lr 0.000034 time 0.9394 (1.0019) model_time 0.9393 (1.0005) loss 2.1820 (2.3642) grad_norm 6.8255 (11.3715/4.8197) mem 68106MB [2022-12-19 04:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][470/1519] eta 0:17:30 lr 0.000034 time 0.9412 (1.0019) model_time 0.9410 (1.0006) loss 2.0464 (2.3634) grad_norm 6.2409 (11.3227/4.8046) mem 68106MB [2022-12-19 04:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][480/1519] eta 0:17:20 lr 0.000034 time 0.9360 (1.0018) model_time 0.9359 (1.0005) loss 2.0243 (2.3573) grad_norm 4.5399 (11.3092/4.7875) mem 68106MB [2022-12-19 04:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][490/1519] eta 0:17:10 lr 0.000034 time 0.9292 (1.0018) model_time 0.9290 (1.0006) loss 2.0040 (2.3532) grad_norm 15.2685 (11.3146/4.7511) mem 68106MB [2022-12-19 04:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][500/1519] eta 0:17:00 lr 0.000034 time 0.9459 (1.0018) model_time 0.9457 (1.0006) loss 2.7773 (2.3550) grad_norm 15.9357 (11.3215/4.7418) mem 68106MB [2022-12-19 04:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][510/1519] eta 0:16:51 lr 0.000034 time 0.9355 (1.0020) model_time 0.9354 (1.0008) loss 1.8929 (2.3519) grad_norm 7.7700 (11.4447/5.0703) mem 68106MB [2022-12-19 04:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][520/1519] eta 0:16:41 lr 0.000034 time 0.9334 (1.0020) model_time 0.9333 (1.0008) loss 2.0575 (2.3497) grad_norm 10.5419 (11.4370/5.0270) mem 68106MB [2022-12-19 04:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][530/1519] eta 0:16:31 lr 0.000034 time 0.9474 (1.0020) model_time 0.9473 (1.0008) loss 3.1214 (2.3500) grad_norm 9.2379 (11.4056/5.0195) mem 68106MB [2022-12-19 04:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][540/1519] eta 0:16:20 lr 0.000034 time 0.9345 (1.0020) model_time 0.9344 (1.0008) loss 2.4870 (2.3489) grad_norm 8.0536 (11.4036/4.9911) mem 68106MB [2022-12-19 04:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][550/1519] eta 0:16:11 lr 0.000034 time 0.9372 (1.0023) model_time 0.9371 (1.0011) loss 1.8930 (2.3461) grad_norm 8.9580 (11.3917/4.9618) mem 68106MB [2022-12-19 04:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][560/1519] eta 0:16:01 lr 0.000034 time 0.9316 (1.0022) model_time 0.9314 (1.0011) loss 1.9568 (2.3456) grad_norm 5.5961 (11.3266/4.9451) mem 68106MB [2022-12-19 04:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][570/1519] eta 0:15:51 lr 0.000034 time 0.9381 (1.0023) model_time 0.9380 (1.0011) loss 2.4726 (2.3430) grad_norm 8.2840 (11.3108/4.9385) mem 68106MB [2022-12-19 04:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][580/1519] eta 0:15:41 lr 0.000034 time 0.9347 (1.0022) model_time 0.9345 (1.0011) loss 3.0236 (2.3408) grad_norm 10.8264 (11.2872/4.9089) mem 68106MB [2022-12-19 04:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][590/1519] eta 0:15:31 lr 0.000034 time 0.9305 (1.0022) model_time 0.9303 (1.0011) loss 2.4100 (2.3388) grad_norm 11.2402 (11.2727/4.8754) mem 68106MB [2022-12-19 04:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][600/1519] eta 0:15:20 lr 0.000034 time 0.9368 (1.0022) model_time 0.9366 (1.0011) loss 1.9503 (2.3360) grad_norm 5.5423 (11.2433/4.8618) mem 68106MB [2022-12-19 04:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][610/1519] eta 0:15:10 lr 0.000034 time 0.9313 (1.0021) model_time 0.9312 (1.0010) loss 2.2877 (2.3398) grad_norm 8.9889 (11.1701/4.7471) mem 68106MB [2022-12-19 04:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][620/1519] eta 0:15:00 lr 0.000034 time 0.9433 (1.0021) model_time 0.9432 (1.0010) loss 1.3553 (2.3372) grad_norm 13.4741 (11.1447/4.7342) mem 68106MB [2022-12-19 04:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][630/1519] eta 0:14:50 lr 0.000034 time 0.9340 (1.0020) model_time 0.9339 (1.0010) loss 2.1316 (2.3359) grad_norm 14.2996 (11.1582/4.7494) mem 68106MB [2022-12-19 04:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][640/1519] eta 0:14:40 lr 0.000034 time 0.9348 (1.0020) model_time 0.9345 (1.0010) loss 2.1624 (2.3342) grad_norm 12.3055 (11.2255/4.7954) mem 68106MB [2022-12-19 04:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][650/1519] eta 0:14:30 lr 0.000034 time 0.9296 (1.0020) model_time 0.9294 (1.0010) loss 1.6985 (2.3296) grad_norm 8.2474 (11.2075/4.8342) mem 68106MB [2022-12-19 04:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][660/1519] eta 0:14:20 lr 0.000034 time 0.9432 (1.0020) model_time 0.9431 (1.0010) loss 1.9351 (2.3265) grad_norm 8.4831 (11.1748/4.8076) mem 68106MB [2022-12-19 04:43:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][670/1519] eta 0:14:10 lr 0.000034 time 0.9375 (1.0020) model_time 0.9374 (1.0010) loss 1.7385 (2.3242) grad_norm 11.3149 (11.2167/4.8114) mem 68106MB [2022-12-19 04:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][680/1519] eta 0:14:00 lr 0.000034 time 0.9335 (1.0020) model_time 0.9333 (1.0010) loss 1.8281 (2.3191) grad_norm 12.5392 (11.2333/4.8126) mem 68106MB [2022-12-19 04:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][690/1519] eta 0:13:50 lr 0.000034 time 0.9384 (1.0019) model_time 0.9382 (1.0009) loss 2.5887 (2.3166) grad_norm 9.6087 (11.1996/4.8034) mem 68106MB [2022-12-19 04:44:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][700/1519] eta 0:13:40 lr 0.000034 time 0.9368 (1.0019) model_time 0.9367 (1.0010) loss 2.7937 (2.3157) grad_norm 15.6055 (11.1967/4.8075) mem 68106MB [2022-12-19 04:44:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][710/1519] eta 0:13:30 lr 0.000034 time 0.9340 (1.0020) model_time 0.9338 (1.0010) loss 1.9342 (2.3157) grad_norm 17.3189 (11.2162/4.8199) mem 68106MB [2022-12-19 04:44:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][720/1519] eta 0:13:20 lr 0.000034 time 0.9307 (1.0019) model_time 0.9306 (1.0010) loss 2.4280 (2.3147) grad_norm 11.1631 (11.2739/4.8344) mem 68106MB [2022-12-19 04:44:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][730/1519] eta 0:13:10 lr 0.000034 time 0.9331 (1.0020) model_time 0.9330 (1.0010) loss 2.1322 (2.3137) grad_norm 8.8325 (11.2745/4.8068) mem 68106MB [2022-12-19 04:44:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][740/1519] eta 0:13:00 lr 0.000034 time 0.9341 (1.0019) model_time 0.9340 (1.0010) loss 1.9951 (2.3117) grad_norm 12.1487 (11.3683/4.8754) mem 68106MB [2022-12-19 04:45:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][750/1519] eta 0:12:50 lr 0.000034 time 0.9366 (1.0019) model_time 0.9365 (1.0010) loss 1.8442 (2.3082) grad_norm 8.8020 (11.3779/4.8566) mem 68106MB [2022-12-19 04:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][760/1519] eta 0:12:40 lr 0.000034 time 0.9365 (1.0019) model_time 0.9364 (1.0010) loss 1.3940 (2.3032) grad_norm 9.3426 (11.3682/4.8534) mem 68106MB [2022-12-19 04:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][770/1519] eta 0:12:30 lr 0.000034 time 0.9375 (1.0019) model_time 0.9373 (1.0010) loss 2.4479 (2.3013) grad_norm 10.6955 (11.1770/4.5766) mem 68106MB [2022-12-19 04:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][780/1519] eta 0:12:20 lr 0.000034 time 0.9388 (1.0019) model_time 0.9386 (1.0009) loss 2.1724 (2.2989) grad_norm 13.1725 (11.1612/4.5055) mem 68106MB [2022-12-19 04:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][790/1519] eta 0:12:10 lr 0.000034 time 0.9407 (1.0018) model_time 0.9406 (1.0009) loss 1.8687 (2.2965) grad_norm 17.9017 (11.1742/4.5197) mem 68106MB [2022-12-19 04:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][800/1519] eta 0:12:00 lr 0.000034 time 0.9361 (1.0018) model_time 0.9360 (1.0009) loss 2.1544 (2.2934) grad_norm 8.8890 (11.1384/4.5357) mem 68106MB [2022-12-19 04:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][810/1519] eta 0:11:50 lr 0.000034 time 0.9398 (1.0018) model_time 0.9397 (1.0009) loss 1.9486 (2.2917) grad_norm 8.6936 (11.1109/4.5628) mem 68106MB [2022-12-19 04:46:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][820/1519] eta 0:11:40 lr 0.000034 time 0.9369 (1.0019) model_time 0.9368 (1.0010) loss 1.9813 (2.2870) grad_norm 10.9227 (11.1107/4.5706) mem 68106MB [2022-12-19 04:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][830/1519] eta 0:11:30 lr 0.000034 time 0.9347 (1.0018) model_time 0.9345 (1.0010) loss 1.6753 (2.2832) grad_norm 8.8703 (11.0956/4.5395) mem 68106MB [2022-12-19 04:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][840/1519] eta 0:11:20 lr 0.000034 time 0.9358 (1.0018) model_time 0.9357 (1.0009) loss 2.0428 (2.2813) grad_norm 20.0240 (11.1171/4.5515) mem 68106MB [2022-12-19 04:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][850/1519] eta 0:11:10 lr 0.000034 time 0.9315 (1.0018) model_time 0.9314 (1.0009) loss 2.3648 (2.2796) grad_norm 8.6216 (11.0892/4.5636) mem 68106MB [2022-12-19 04:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][860/1519] eta 0:11:00 lr 0.000034 time 0.9338 (1.0021) model_time 0.9336 (1.0012) loss 2.6718 (2.2785) grad_norm 10.0216 (11.0147/4.5365) mem 68106MB [2022-12-19 04:47:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][870/1519] eta 0:10:50 lr 0.000034 time 0.9318 (1.0021) model_time 0.9316 (1.0012) loss 1.6071 (2.2755) grad_norm 10.3875 (10.9518/4.5134) mem 68106MB [2022-12-19 04:47:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][880/1519] eta 0:10:40 lr 0.000034 time 0.9333 (1.0020) model_time 0.9331 (1.0012) loss 1.2305 (2.2736) grad_norm 11.4256 (10.9942/4.6046) mem 68106MB [2022-12-19 04:47:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][890/1519] eta 0:10:30 lr 0.000034 time 0.9361 (1.0020) model_time 0.9359 (1.0011) loss 1.8630 (2.2735) grad_norm 10.7779 (10.9219/4.4942) mem 68106MB [2022-12-19 04:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][900/1519] eta 0:10:20 lr 0.000034 time 0.9348 (1.0019) model_time 0.9347 (1.0011) loss 1.4637 (2.2695) grad_norm 11.7410 (10.9247/4.4964) mem 68106MB [2022-12-19 04:47:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][910/1519] eta 0:10:10 lr 0.000034 time 0.9317 (1.0019) model_time 0.9315 (1.0011) loss 2.0819 (2.2668) grad_norm 7.3614 (10.9171/4.6586) mem 68106MB [2022-12-19 04:47:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][920/1519] eta 0:10:00 lr 0.000034 time 0.9374 (1.0019) model_time 0.9372 (1.0010) loss 1.8781 (2.2650) grad_norm 24.3828 (10.9640/4.6990) mem 68106MB [2022-12-19 04:48:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][930/1519] eta 0:09:50 lr 0.000034 time 0.9344 (1.0018) model_time 0.9343 (1.0010) loss 2.4019 (2.2637) grad_norm 10.8008 (10.9302/4.6982) mem 68106MB [2022-12-19 04:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][940/1519] eta 0:09:40 lr 0.000034 time 0.9363 (1.0018) model_time 0.9362 (1.0010) loss 2.2336 (2.2640) grad_norm 8.0795 (10.9769/4.7107) mem 68106MB [2022-12-19 04:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][950/1519] eta 0:09:30 lr 0.000034 time 0.9354 (1.0018) model_time 0.9353 (1.0010) loss 2.4299 (2.2609) grad_norm 6.8149 (11.0148/4.7257) mem 68106MB [2022-12-19 04:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][960/1519] eta 0:09:20 lr 0.000034 time 0.9275 (1.0018) model_time 0.9274 (1.0010) loss 2.0816 (2.2583) grad_norm 8.0977 (11.0857/4.9065) mem 68106MB [2022-12-19 04:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][970/1519] eta 0:09:10 lr 0.000034 time 0.9378 (1.0018) model_time 0.9376 (1.0010) loss 2.0260 (2.2557) grad_norm 28.1147 (11.1289/4.9955) mem 68106MB [2022-12-19 04:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][980/1519] eta 0:08:59 lr 0.000034 time 0.9344 (1.0018) model_time 0.9342 (1.0010) loss 1.5317 (2.2540) grad_norm 9.2348 (11.1716/4.9784) mem 68106MB [2022-12-19 04:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][990/1519] eta 0:08:49 lr 0.000034 time 0.9340 (1.0018) model_time 0.9339 (1.0010) loss 2.6203 (2.2542) grad_norm 9.8658 (11.2643/5.0005) mem 68106MB [2022-12-19 04:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1000/1519] eta 0:08:39 lr 0.000034 time 0.9330 (1.0018) model_time 0.9329 (1.0010) loss 1.8752 (2.2521) grad_norm 10.5592 (11.2574/5.0104) mem 68106MB [2022-12-19 04:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1010/1519] eta 0:08:29 lr 0.000034 time 0.9364 (1.0018) model_time 0.9363 (1.0010) loss 1.9911 (2.2490) grad_norm 10.1399 (11.2232/4.9923) mem 68106MB [2022-12-19 04:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1020/1519] eta 0:08:19 lr 0.000034 time 0.9321 (1.0018) model_time 0.9320 (1.0010) loss 1.9324 (2.2478) grad_norm 17.8830 (11.2507/4.9986) mem 68106MB [2022-12-19 04:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1030/1519] eta 0:08:09 lr 0.000034 time 0.9422 (1.0018) model_time 0.9420 (1.0010) loss 1.6793 (2.2450) grad_norm 9.0848 (11.1859/4.8879) mem 68106MB [2022-12-19 04:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1040/1519] eta 0:07:59 lr 0.000034 time 0.9308 (1.0018) model_time 0.9307 (1.0010) loss 2.1741 (2.2452) grad_norm 9.6681 (11.2264/4.8974) mem 68106MB [2022-12-19 04:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1050/1519] eta 0:07:49 lr 0.000034 time 0.9283 (1.0018) model_time 0.9281 (1.0010) loss 1.6033 (2.2433) grad_norm 15.4738 (11.2229/4.8961) mem 68106MB [2022-12-19 04:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1060/1519] eta 0:07:39 lr 0.000034 time 0.9373 (1.0018) model_time 0.9372 (1.0010) loss 1.6216 (2.2417) grad_norm 21.0413 (11.2888/4.9290) mem 68106MB [2022-12-19 04:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1070/1519] eta 0:07:29 lr 0.000034 time 0.9402 (1.0018) model_time 0.9400 (1.0010) loss 1.4783 (2.2389) grad_norm 12.4906 (11.3187/4.9091) mem 68106MB [2022-12-19 04:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1080/1519] eta 0:07:19 lr 0.000034 time 0.9378 (1.0017) model_time 0.9377 (1.0010) loss 2.5324 (2.2369) grad_norm 8.5515 (11.3178/4.8916) mem 68106MB [2022-12-19 04:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1090/1519] eta 0:07:09 lr 0.000034 time 0.9387 (1.0017) model_time 0.9384 (1.0010) loss 2.1247 (2.2343) grad_norm 7.3749 (11.3003/4.8946) mem 68106MB [2022-12-19 04:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1100/1519] eta 0:06:59 lr 0.000034 time 0.9418 (1.0017) model_time 0.9416 (1.0010) loss 2.5756 (2.2313) grad_norm 11.8057 (11.2687/4.8741) mem 68106MB [2022-12-19 04:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1110/1519] eta 0:06:49 lr 0.000034 time 0.9290 (1.0017) model_time 0.9288 (1.0010) loss 1.9904 (2.2301) grad_norm 6.3045 (11.1200/4.5613) mem 68106MB [2022-12-19 04:51:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1120/1519] eta 0:06:39 lr 0.000034 time 0.9392 (1.0017) model_time 0.9391 (1.0009) loss 2.0334 (2.2296) grad_norm 18.5585 (11.1620/4.5948) mem 68106MB [2022-12-19 04:51:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1130/1519] eta 0:06:29 lr 0.000034 time 0.8880 (1.0017) model_time 0.8879 (1.0010) loss 1.1528 (2.2258) grad_norm 9.1383 (11.1694/4.5642) mem 68106MB [2022-12-19 04:51:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1140/1519] eta 0:06:19 lr 0.000034 time 0.9368 (1.0017) model_time 0.9367 (1.0010) loss 2.0058 (2.2254) grad_norm 10.4348 (11.1245/4.5609) mem 68106MB [2022-12-19 04:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1150/1519] eta 0:06:09 lr 0.000034 time 0.9290 (1.0017) model_time 0.9289 (1.0009) loss 2.4242 (2.2250) grad_norm 22.4491 (11.2061/4.6199) mem 68106MB [2022-12-19 04:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1160/1519] eta 0:05:59 lr 0.000034 time 0.9320 (1.0017) model_time 0.9319 (1.0009) loss 2.3847 (2.2236) grad_norm 13.4744 (11.2427/4.6092) mem 68106MB [2022-12-19 04:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1170/1519] eta 0:05:49 lr 0.000034 time 1.0794 (1.0018) model_time 1.0792 (1.0010) loss 1.9911 (2.2214) grad_norm 8.6826 (11.2405/4.5782) mem 68106MB [2022-12-19 04:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1180/1519] eta 0:05:39 lr 0.000034 time 0.9372 (1.0017) model_time 0.9371 (1.0010) loss 1.9152 (2.2203) grad_norm 6.0617 (11.2351/4.5863) mem 68106MB [2022-12-19 04:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1190/1519] eta 0:05:29 lr 0.000034 time 0.9246 (1.0017) model_time 0.9245 (1.0010) loss 2.4794 (2.2197) grad_norm 8.7249 (11.2282/4.5861) mem 68106MB [2022-12-19 04:52:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1200/1519] eta 0:05:19 lr 0.000034 time 0.9300 (1.0017) model_time 0.9298 (1.0010) loss 1.7574 (2.2174) grad_norm 7.5573 (11.2102/4.5868) mem 68106MB [2022-12-19 04:52:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1210/1519] eta 0:05:09 lr 0.000034 time 0.9345 (1.0017) model_time 0.9343 (1.0010) loss 2.1592 (2.2164) grad_norm 6.7369 (11.1642/4.5919) mem 68106MB [2022-12-19 04:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1220/1519] eta 0:04:59 lr 0.000034 time 0.9317 (1.0017) model_time 0.9314 (1.0010) loss 2.3113 (2.2140) grad_norm 19.4983 (11.2403/4.6074) mem 68106MB [2022-12-19 04:53:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1230/1519] eta 0:04:49 lr 0.000034 time 0.9326 (1.0017) model_time 0.9325 (1.0010) loss 2.4156 (2.2135) grad_norm 9.2305 (11.2017/4.5842) mem 68106MB [2022-12-19 04:53:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1240/1519] eta 0:04:39 lr 0.000034 time 0.9416 (1.0017) model_time 0.9415 (1.0010) loss 1.8093 (2.2106) grad_norm 17.7668 (11.2105/4.5774) mem 68106MB [2022-12-19 04:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1250/1519] eta 0:04:29 lr 0.000034 time 0.9335 (1.0017) model_time 0.9333 (1.0010) loss 2.2994 (2.2100) grad_norm 9.5270 (11.2404/4.5409) mem 68106MB [2022-12-19 04:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1260/1519] eta 0:04:19 lr 0.000034 time 0.9318 (1.0017) model_time 0.9316 (1.0010) loss 1.4475 (2.2074) grad_norm 9.3252 (11.2253/4.5087) mem 68106MB [2022-12-19 04:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1270/1519] eta 0:04:09 lr 0.000034 time 0.9358 (1.0017) model_time 0.9356 (1.0010) loss 2.5735 (2.2061) grad_norm 7.6958 (11.2002/4.5176) mem 68106MB [2022-12-19 04:53:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1280/1519] eta 0:03:59 lr 0.000034 time 0.9466 (1.0017) model_time 0.9465 (1.0010) loss 2.7788 (2.2060) grad_norm 8.1402 (11.2014/4.5206) mem 68106MB [2022-12-19 04:54:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1290/1519] eta 0:03:49 lr 0.000034 time 0.9304 (1.0017) model_time 0.9302 (1.0010) loss 2.5189 (2.2042) grad_norm 20.5819 (11.2700/4.5568) mem 68106MB [2022-12-19 04:54:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1300/1519] eta 0:03:39 lr 0.000034 time 0.9320 (1.0017) model_time 0.9319 (1.0010) loss 1.9137 (2.2021) grad_norm 9.9983 (11.2983/4.5406) mem 68106MB [2022-12-19 04:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1310/1519] eta 0:03:29 lr 0.000034 time 0.9458 (1.0016) model_time 0.9455 (1.0010) loss 1.2838 (2.1992) grad_norm 8.6192 (11.2719/4.5253) mem 68106MB [2022-12-19 04:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1320/1519] eta 0:03:19 lr 0.000034 time 0.9343 (1.0016) model_time 0.9342 (1.0009) loss 2.1630 (2.1978) grad_norm 15.3053 (11.2399/4.4986) mem 68106MB [2022-12-19 04:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1330/1519] eta 0:03:09 lr 0.000034 time 0.9360 (1.0016) model_time 0.9359 (1.0009) loss 1.7830 (2.1954) grad_norm 9.2739 (11.2220/4.4994) mem 68106MB [2022-12-19 04:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1340/1519] eta 0:02:59 lr 0.000034 time 0.9374 (1.0016) model_time 0.9372 (1.0009) loss 2.1936 (2.1936) grad_norm 16.3285 (11.1450/4.4163) mem 68106MB [2022-12-19 04:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1350/1519] eta 0:02:49 lr 0.000034 time 1.1371 (1.0017) model_time 1.1370 (1.0010) loss 2.4630 (2.1906) grad_norm 14.9953 (11.1482/4.4198) mem 68106MB [2022-12-19 04:55:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1360/1519] eta 0:02:39 lr 0.000034 time 0.9371 (1.0017) model_time 0.9370 (1.0010) loss 1.5251 (2.1890) grad_norm 9.2847 (11.1738/4.4143) mem 68106MB [2022-12-19 04:55:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1370/1519] eta 0:02:29 lr 0.000034 time 0.9236 (1.0017) model_time 0.9234 (1.0010) loss 2.3440 (2.1887) grad_norm 15.7401 (11.2301/4.4229) mem 68106MB [2022-12-19 04:55:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1380/1519] eta 0:02:19 lr 0.000034 time 0.9269 (1.0017) model_time 0.9267 (1.0010) loss 1.5061 (2.1874) grad_norm 9.3459 (11.1644/4.3357) mem 68106MB [2022-12-19 04:55:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1390/1519] eta 0:02:09 lr 0.000034 time 0.9362 (1.0017) model_time 0.9361 (1.0010) loss 2.2295 (2.1866) grad_norm 10.4358 (11.1175/4.3223) mem 68106MB [2022-12-19 04:55:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1400/1519] eta 0:01:59 lr 0.000034 time 0.9246 (1.0017) model_time 0.9245 (1.0010) loss 2.0183 (2.1848) grad_norm 8.2387 (11.1562/4.3092) mem 68106MB [2022-12-19 04:56:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1410/1519] eta 0:01:49 lr 0.000034 time 0.9371 (1.0017) model_time 0.9370 (1.0010) loss 1.2445 (2.1825) grad_norm 9.1833 (11.1452/4.2656) mem 68106MB [2022-12-19 04:56:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1420/1519] eta 0:01:39 lr 0.000034 time 0.9359 (1.0017) model_time 0.9357 (1.0011) loss 2.3401 (2.1820) grad_norm 8.7677 (11.1501/4.2497) mem 68106MB [2022-12-19 04:56:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1430/1519] eta 0:01:29 lr 0.000034 time 0.9443 (1.0017) model_time 0.9442 (1.0011) loss 2.6432 (2.1813) grad_norm 13.3215 (11.1875/4.2746) mem 68106MB [2022-12-19 04:56:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1440/1519] eta 0:01:19 lr 0.000034 time 0.9367 (1.0017) model_time 0.9366 (1.0011) loss 2.3371 (2.1813) grad_norm 11.1977 (11.1340/4.2479) mem 68106MB [2022-12-19 04:56:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1450/1519] eta 0:01:09 lr 0.000034 time 0.9379 (1.0018) model_time 0.9378 (1.0012) loss 1.3033 (2.1799) grad_norm 6.6926 (11.1527/4.2407) mem 68106MB [2022-12-19 04:56:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1460/1519] eta 0:00:59 lr 0.000034 time 0.9328 (1.0018) model_time 0.9327 (1.0012) loss 1.8684 (2.1787) grad_norm 7.3521 (11.1401/4.2576) mem 68106MB [2022-12-19 04:57:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1470/1519] eta 0:00:49 lr 0.000034 time 0.9298 (1.0018) model_time 0.9296 (1.0012) loss 1.4805 (2.1777) grad_norm 10.9358 (11.1749/4.2586) mem 68106MB [2022-12-19 04:57:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1480/1519] eta 0:00:39 lr 0.000034 time 0.9334 (1.0018) model_time 0.9333 (1.0011) loss 1.6731 (2.1758) grad_norm 15.0642 (11.1683/4.1912) mem 68106MB [2022-12-19 04:57:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1490/1519] eta 0:00:29 lr 0.000034 time 0.9290 (1.0018) model_time 0.9288 (1.0011) loss 1.1894 (2.1734) grad_norm 12.3475 (11.1719/4.1842) mem 68106MB [2022-12-19 04:57:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1500/1519] eta 0:00:19 lr 0.000034 time 0.9359 (1.0018) model_time 0.9357 (1.0011) loss 1.3783 (2.1703) grad_norm 5.3915 (11.1397/4.2041) mem 68106MB [2022-12-19 04:57:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [1/100][1510/1519] eta 0:00:09 lr 0.000034 time 0.9291 (1.0017) model_time 0.9290 (1.0011) loss 1.9453 (2.1695) grad_norm 14.7241 (11.1019/3.9923) mem 68106MB [2022-12-19 04:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 1 training takes 0:25:21 [2022-12-19 04:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_1.pth saving...... [2022-12-19 04:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_1.pth saved !!! [2022-12-19 04:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.644 (0.644) Loss 4.7927 (4.7927) Acc@1 18.750 (18.750) Acc@5 39.931 (39.931) Mem 68106MB [2022-12-19 04:58:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.330) Loss 5.2056 (5.0047) Acc@1 14.236 (16.572) Acc@5 32.639 (36.016) Mem 68106MB [2022-12-19 04:58:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.315) Loss 4.8645 (5.0099) Acc@1 16.667 (16.336) Acc@5 37.847 (36.376) Mem 68106MB [2022-12-19 04:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 5.1506 (5.0186) Acc@1 12.153 (16.252) Acc@5 30.556 (36.010) Mem 68106MB [2022-12-19 04:58:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.307) Loss 4.9177 (5.0005) Acc@1 14.583 (16.319) Acc@5 39.236 (36.408) Mem 68106MB [2022-12-19 04:58:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 5.0158 (4.9967) Acc@1 18.056 (16.319) Acc@5 34.375 (36.540) Mem 68106MB [2022-12-19 04:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.304) Loss 5.1636 (5.0051) Acc@1 14.236 (16.217) Acc@5 29.861 (36.362) Mem 68106MB [2022-12-19 04:58:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.298 (0.304) Loss 5.2117 (5.0114) Acc@1 11.458 (16.129) Acc@5 32.292 (36.082) Mem 68106MB [2022-12-19 04:58:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 4.9390 (5.0089) Acc@1 16.667 (16.135) Acc@5 37.847 (36.120) Mem 68106MB [2022-12-19 04:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:1] * Acc@1 16.016 Acc@5 36.072 [2022-12-19 04:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 16.0% [2022-12-19 04:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 04:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 04:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 16.02% [2022-12-19 04:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][0/1519] eta 0:33:59 lr 0.000034 time 1.3424 (1.3424) model_time 0.9457 (0.9457) loss 1.7809 (1.7809) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 04:59:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][10/1519] eta 0:25:55 lr 0.000034 time 0.9360 (1.0305) model_time 0.9358 (0.9941) loss 1.6432 (1.7984) grad_norm 20.8041 (16.1666/3.8519) mem 68106MB [2022-12-19 04:59:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][20/1519] eta 0:25:25 lr 0.000034 time 0.9343 (1.0180) model_time 0.9342 (0.9987) loss 2.2739 (1.9065) grad_norm 15.3702 (15.8077/5.6768) mem 68106MB [2022-12-19 04:59:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][30/1519] eta 0:25:08 lr 0.000034 time 0.9312 (1.0134) model_time 0.9310 (1.0002) loss 1.7508 (1.8933) grad_norm 11.4063 (14.5887/5.2086) mem 68106MB [2022-12-19 04:59:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][40/1519] eta 0:24:54 lr 0.000034 time 0.9389 (1.0103) model_time 0.9387 (1.0002) loss 1.4834 (1.8788) grad_norm 8.1563 (14.3679/4.9669) mem 68106MB [2022-12-19 04:59:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][50/1519] eta 0:24:40 lr 0.000034 time 0.9353 (1.0080) model_time 0.9351 (0.9997) loss 2.2777 (1.8817) grad_norm 17.2982 (14.0195/4.8184) mem 68106MB [2022-12-19 05:00:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][60/1519] eta 0:24:29 lr 0.000034 time 0.9278 (1.0073) model_time 0.9277 (1.0003) loss 2.4199 (1.9085) grad_norm 11.3235 (13.7908/4.6077) mem 68106MB [2022-12-19 05:00:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][70/1519] eta 0:24:18 lr 0.000034 time 0.9337 (1.0063) model_time 0.9335 (1.0003) loss 1.6115 (1.8904) grad_norm 9.3610 (13.3614/4.5041) mem 68106MB [2022-12-19 05:00:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][80/1519] eta 0:24:07 lr 0.000034 time 0.9250 (1.0058) model_time 0.9248 (1.0005) loss 1.5462 (1.8948) grad_norm 12.8666 (13.0908/4.4641) mem 68106MB [2022-12-19 05:00:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][90/1519] eta 0:23:56 lr 0.000034 time 0.9240 (1.0050) model_time 0.9239 (1.0003) loss 2.1886 (1.9049) grad_norm 9.1012 (12.8678/4.3069) mem 68106MB [2022-12-19 05:00:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][100/1519] eta 0:23:45 lr 0.000034 time 0.9370 (1.0045) model_time 0.9369 (1.0001) loss 1.6887 (1.9041) grad_norm 11.9613 (12.7192/4.3613) mem 68106MB [2022-12-19 05:00:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][110/1519] eta 0:23:34 lr 0.000034 time 0.9270 (1.0040) model_time 0.9269 (1.0000) loss 1.6400 (1.9134) grad_norm 6.8221 (12.2967/4.4054) mem 68106MB [2022-12-19 05:01:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][120/1519] eta 0:23:24 lr 0.000034 time 0.9424 (1.0037) model_time 0.9420 (1.0000) loss 1.5595 (1.9172) grad_norm 6.9523 (12.2561/4.3129) mem 68106MB [2022-12-19 05:01:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][130/1519] eta 0:23:13 lr 0.000034 time 0.9294 (1.0032) model_time 0.9292 (0.9998) loss 2.0894 (1.9023) grad_norm 17.4597 (12.5067/4.3517) mem 68106MB [2022-12-19 05:01:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][140/1519] eta 0:23:03 lr 0.000034 time 0.9610 (1.0032) model_time 0.9608 (1.0000) loss 2.1292 (1.9022) grad_norm 13.4421 (12.3941/4.4403) mem 68106MB [2022-12-19 05:01:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][150/1519] eta 0:22:54 lr 0.000034 time 0.9370 (1.0039) model_time 0.9369 (1.0008) loss 1.5398 (1.8903) grad_norm 9.7572 (12.1783/4.3777) mem 68106MB [2022-12-19 05:01:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][160/1519] eta 0:22:43 lr 0.000034 time 0.9354 (1.0036) model_time 0.9352 (1.0007) loss 2.0711 (1.8848) grad_norm 14.2746 (12.2148/4.4094) mem 68106MB [2022-12-19 05:01:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][170/1519] eta 0:22:33 lr 0.000034 time 0.9342 (1.0035) model_time 0.9340 (1.0007) loss 1.7980 (1.8803) grad_norm 9.2982 (11.9702/4.4073) mem 68106MB [2022-12-19 05:02:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][180/1519] eta 0:22:23 lr 0.000034 time 0.9325 (1.0033) model_time 0.9324 (1.0007) loss 1.5693 (1.8810) grad_norm 15.0196 (11.9998/4.3320) mem 68106MB [2022-12-19 05:02:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][190/1519] eta 0:22:13 lr 0.000034 time 0.9508 (1.0034) model_time 0.9506 (1.0009) loss 1.7748 (1.8781) grad_norm 6.3969 (11.9795/4.4442) mem 68106MB [2022-12-19 05:02:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][200/1519] eta 0:22:03 lr 0.000034 time 0.9343 (1.0032) model_time 0.9341 (1.0008) loss 1.8904 (1.8755) grad_norm 9.4684 (12.0398/4.4660) mem 68106MB [2022-12-19 05:02:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][210/1519] eta 0:21:53 lr 0.000034 time 0.9401 (1.0031) model_time 0.9400 (1.0008) loss 1.7242 (1.8801) grad_norm 8.5361 (11.9960/4.4329) mem 68106MB [2022-12-19 05:02:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][220/1519] eta 0:21:42 lr 0.000034 time 0.9346 (1.0030) model_time 0.9343 (1.0007) loss 1.9410 (1.8759) grad_norm 12.3658 (11.9116/4.3942) mem 68106MB [2022-12-19 05:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][230/1519] eta 0:21:33 lr 0.000034 time 0.9425 (1.0031) model_time 0.9424 (1.0010) loss 1.4173 (1.8663) grad_norm 9.1712 (11.7791/4.3612) mem 68106MB [2022-12-19 05:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][240/1519] eta 0:21:22 lr 0.000034 time 0.9115 (1.0030) model_time 0.9113 (1.0010) loss 1.3437 (1.8689) grad_norm 12.7803 (11.6926/4.3158) mem 68106MB [2022-12-19 05:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][250/1519] eta 0:21:12 lr 0.000034 time 0.9327 (1.0029) model_time 0.9326 (1.0009) loss 1.4973 (1.8639) grad_norm 14.1349 (11.5868/4.3052) mem 68106MB [2022-12-19 05:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][260/1519] eta 0:21:03 lr 0.000034 time 0.9308 (1.0032) model_time 0.9307 (1.0013) loss 2.1116 (1.8674) grad_norm 11.5925 (11.5348/4.2554) mem 68106MB [2022-12-19 05:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][270/1519] eta 0:20:52 lr 0.000034 time 0.9283 (1.0031) model_time 0.9281 (1.0012) loss 1.8008 (1.8736) grad_norm 12.3559 (11.4493/4.2324) mem 68106MB [2022-12-19 05:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][280/1519] eta 0:20:42 lr 0.000034 time 0.9349 (1.0030) model_time 0.9347 (1.0012) loss 1.6276 (1.8677) grad_norm 7.2468 (11.3959/4.1736) mem 68106MB [2022-12-19 05:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][290/1519] eta 0:20:32 lr 0.000034 time 0.9271 (1.0028) model_time 0.9270 (1.0011) loss 1.7909 (1.8666) grad_norm 16.7422 (11.3769/4.1444) mem 68106MB [2022-12-19 05:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][300/1519] eta 0:20:22 lr 0.000034 time 0.9371 (1.0028) model_time 0.9369 (1.0011) loss 1.9721 (1.8683) grad_norm 19.8283 (11.4599/4.2527) mem 68106MB [2022-12-19 05:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][310/1519] eta 0:20:12 lr 0.000034 time 0.9444 (1.0027) model_time 0.9443 (1.0010) loss 2.4591 (1.8692) grad_norm 7.2906 (11.4293/4.2671) mem 68106MB [2022-12-19 05:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][320/1519] eta 0:20:02 lr 0.000034 time 0.9316 (1.0026) model_time 0.9313 (1.0010) loss 2.1196 (1.8705) grad_norm 7.9664 (11.4001/4.2316) mem 68106MB [2022-12-19 05:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][330/1519] eta 0:19:52 lr 0.000034 time 0.9289 (1.0034) model_time 0.9287 (1.0018) loss 1.8695 (1.8767) grad_norm 11.9235 (11.3797/4.1993) mem 68106MB [2022-12-19 05:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][340/1519] eta 0:19:42 lr 0.000034 time 0.9367 (1.0033) model_time 0.9366 (1.0017) loss 1.9628 (1.8728) grad_norm 6.4506 (11.3760/4.1987) mem 68106MB [2022-12-19 05:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][350/1519] eta 0:19:32 lr 0.000034 time 0.9384 (1.0033) model_time 0.9382 (1.0018) loss 1.9373 (1.8722) grad_norm 4.5144 (11.4066/4.2103) mem 68106MB [2022-12-19 05:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][360/1519] eta 0:19:22 lr 0.000034 time 0.9328 (1.0032) model_time 0.9327 (1.0017) loss 2.0173 (1.8697) grad_norm 13.3441 (11.4030/4.1876) mem 68106MB [2022-12-19 05:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][370/1519] eta 0:19:12 lr 0.000034 time 0.9303 (1.0031) model_time 0.9301 (1.0016) loss 1.7988 (1.8709) grad_norm 6.3895 (11.4776/4.2385) mem 68106MB [2022-12-19 05:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][380/1519] eta 0:19:02 lr 0.000034 time 0.9283 (1.0029) model_time 0.9282 (1.0015) loss 1.7145 (1.8713) grad_norm 7.7734 (11.4272/4.2426) mem 68106MB [2022-12-19 05:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][390/1519] eta 0:18:52 lr 0.000034 time 0.9323 (1.0029) model_time 0.9322 (1.0015) loss 1.9155 (1.8722) grad_norm 10.1123 (11.4544/4.2388) mem 68106MB [2022-12-19 05:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][400/1519] eta 0:18:42 lr 0.000034 time 0.9328 (1.0027) model_time 0.9326 (1.0014) loss 2.1282 (1.8688) grad_norm 5.4062 (11.3792/4.2218) mem 68106MB [2022-12-19 05:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][410/1519] eta 0:18:32 lr 0.000034 time 0.9330 (1.0027) model_time 0.9329 (1.0014) loss 1.6284 (1.8682) grad_norm 9.4798 (11.3720/4.1825) mem 68106MB [2022-12-19 05:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][420/1519] eta 0:18:21 lr 0.000034 time 0.9309 (1.0026) model_time 0.9307 (1.0013) loss 1.9886 (1.8727) grad_norm 17.3601 (11.4288/4.2436) mem 68106MB [2022-12-19 05:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][430/1519] eta 0:18:11 lr 0.000034 time 0.9266 (1.0026) model_time 0.9264 (1.0013) loss 1.7165 (1.8724) grad_norm 9.1526 (11.3799/4.2188) mem 68106MB [2022-12-19 05:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][440/1519] eta 0:18:01 lr 0.000034 time 0.9374 (1.0026) model_time 0.9373 (1.0014) loss 1.9441 (1.8713) grad_norm 10.9460 (11.3425/4.1854) mem 68106MB [2022-12-19 05:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][450/1519] eta 0:17:51 lr 0.000034 time 0.9329 (1.0025) model_time 0.9328 (1.0013) loss 1.6481 (1.8694) grad_norm 19.7806 (11.3315/4.2009) mem 68106MB [2022-12-19 05:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][460/1519] eta 0:17:41 lr 0.000034 time 0.9297 (1.0026) model_time 0.9296 (1.0014) loss 2.3871 (1.8716) grad_norm 9.7524 (11.4025/4.2438) mem 68106MB [2022-12-19 05:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][470/1519] eta 0:17:31 lr 0.000034 time 0.9328 (1.0025) model_time 0.9326 (1.0013) loss 1.9618 (1.8713) grad_norm 9.3401 (11.4104/4.2324) mem 68106MB [2022-12-19 05:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][480/1519] eta 0:17:21 lr 0.000034 time 0.9288 (1.0025) model_time 0.9286 (1.0013) loss 1.6666 (1.8693) grad_norm 9.0436 (11.3546/4.2185) mem 68106MB [2022-12-19 05:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][490/1519] eta 0:17:11 lr 0.000034 time 0.9393 (1.0024) model_time 0.9392 (1.0012) loss 1.6391 (1.8688) grad_norm 7.0390 (11.3480/4.1871) mem 68106MB [2022-12-19 05:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][500/1519] eta 0:17:01 lr 0.000034 time 0.9362 (1.0024) model_time 0.9360 (1.0012) loss 1.4484 (1.8650) grad_norm 7.5890 (11.3325/4.2008) mem 68106MB [2022-12-19 05:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][510/1519] eta 0:16:51 lr 0.000034 time 0.9370 (1.0024) model_time 0.9369 (1.0012) loss 2.3669 (1.8670) grad_norm 11.3509 (11.2948/4.1956) mem 68106MB [2022-12-19 05:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][520/1519] eta 0:16:41 lr 0.000034 time 0.9552 (1.0025) model_time 0.9551 (1.0014) loss 1.5621 (1.8661) grad_norm 12.0481 (11.2589/4.1719) mem 68106MB [2022-12-19 05:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][530/1519] eta 0:16:31 lr 0.000034 time 0.9343 (1.0025) model_time 0.9342 (1.0014) loss 2.0536 (1.8650) grad_norm 12.4805 (11.2055/4.1662) mem 68106MB [2022-12-19 05:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][540/1519] eta 0:16:21 lr 0.000034 time 0.9362 (1.0025) model_time 0.9361 (1.0014) loss 1.7055 (1.8651) grad_norm 11.2440 (11.2020/4.1305) mem 68106MB [2022-12-19 05:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][550/1519] eta 0:16:11 lr 0.000034 time 0.9323 (1.0024) model_time 0.9321 (1.0013) loss 1.7828 (1.8661) grad_norm 6.0755 (11.1973/4.1372) mem 68106MB [2022-12-19 05:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][560/1519] eta 0:16:01 lr 0.000034 time 0.9358 (1.0023) model_time 0.9357 (1.0013) loss 2.0996 (1.8688) grad_norm 6.5481 (11.1727/4.1141) mem 68106MB [2022-12-19 05:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][570/1519] eta 0:15:51 lr 0.000034 time 0.9196 (1.0023) model_time 0.9195 (1.0012) loss 2.4484 (1.8691) grad_norm 12.5178 (11.1502/4.0961) mem 68106MB [2022-12-19 05:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][580/1519] eta 0:15:41 lr 0.000034 time 0.9303 (1.0022) model_time 0.9302 (1.0012) loss 1.4972 (1.8674) grad_norm 14.0895 (11.1505/4.0920) mem 68106MB [2022-12-19 05:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][590/1519] eta 0:15:31 lr 0.000034 time 0.9330 (1.0022) model_time 0.9329 (1.0012) loss 2.2931 (1.8682) grad_norm 15.4174 (11.2054/4.0841) mem 68106MB [2022-12-19 05:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][600/1519] eta 0:15:21 lr 0.000034 time 0.9413 (1.0022) model_time 0.9412 (1.0012) loss 1.9317 (1.8675) grad_norm 8.7774 (11.2220/4.0683) mem 68106MB [2022-12-19 05:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][610/1519] eta 0:15:10 lr 0.000034 time 0.9381 (1.0022) model_time 0.9380 (1.0012) loss 1.3262 (1.8689) grad_norm 15.8875 (11.2012/4.0620) mem 68106MB [2022-12-19 05:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][620/1519] eta 0:15:00 lr 0.000034 time 0.9371 (1.0021) model_time 0.9369 (1.0011) loss 1.6915 (1.8700) grad_norm 6.4435 (11.1038/3.9331) mem 68106MB [2022-12-19 05:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][630/1519] eta 0:14:50 lr 0.000034 time 0.9299 (1.0021) model_time 0.9298 (1.0011) loss 1.7407 (1.8696) grad_norm 10.6402 (11.0748/3.9167) mem 68106MB [2022-12-19 05:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][640/1519] eta 0:14:41 lr 0.000034 time 0.9290 (1.0024) model_time 0.9288 (1.0015) loss 2.2958 (1.8680) grad_norm 8.8269 (11.0338/3.9322) mem 68106MB [2022-12-19 05:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][650/1519] eta 0:14:31 lr 0.000034 time 0.9348 (1.0024) model_time 0.9347 (1.0014) loss 1.8956 (1.8692) grad_norm 6.1240 (11.0183/3.9366) mem 68106MB [2022-12-19 05:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][660/1519] eta 0:14:21 lr 0.000034 time 0.9296 (1.0024) model_time 0.9295 (1.0014) loss 1.6474 (1.8686) grad_norm 7.0701 (10.9708/3.9211) mem 68106MB [2022-12-19 05:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][670/1519] eta 0:14:10 lr 0.000034 time 0.9372 (1.0023) model_time 0.9370 (1.0014) loss 2.0125 (1.8686) grad_norm 7.5140 (10.9675/3.9680) mem 68106MB [2022-12-19 05:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][680/1519] eta 0:14:00 lr 0.000034 time 0.9364 (1.0023) model_time 0.9363 (1.0014) loss 2.0905 (1.8674) grad_norm 11.5104 (10.9809/3.9628) mem 68106MB [2022-12-19 05:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][690/1519] eta 0:13:50 lr 0.000034 time 0.9363 (1.0023) model_time 0.9362 (1.0013) loss 1.1616 (1.8660) grad_norm 12.1902 (10.9550/3.9648) mem 68106MB [2022-12-19 05:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][700/1519] eta 0:13:40 lr 0.000034 time 0.9366 (1.0022) model_time 0.9364 (1.0013) loss 1.4201 (1.8640) grad_norm 10.3597 (10.9053/3.9376) mem 68106MB [2022-12-19 05:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][710/1519] eta 0:13:30 lr 0.000034 time 0.9332 (1.0022) model_time 0.9331 (1.0013) loss 2.0075 (1.8653) grad_norm 8.6403 (10.9203/3.9313) mem 68106MB [2022-12-19 05:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][720/1519] eta 0:13:20 lr 0.000034 time 0.9374 (1.0022) model_time 0.9372 (1.0013) loss 1.9525 (1.8633) grad_norm 10.5458 (10.9235/3.9239) mem 68106MB [2022-12-19 05:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][730/1519] eta 0:13:10 lr 0.000034 time 0.9326 (1.0021) model_time 0.9325 (1.0012) loss 1.6927 (1.8590) grad_norm 12.3299 (10.8274/3.8644) mem 68106MB [2022-12-19 05:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][740/1519] eta 0:13:00 lr 0.000034 time 0.9373 (1.0021) model_time 0.9371 (1.0012) loss 1.5161 (1.8578) grad_norm 39.2350 (10.9729/4.2196) mem 68106MB [2022-12-19 05:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][750/1519] eta 0:12:50 lr 0.000034 time 0.9303 (1.0021) model_time 0.9302 (1.0013) loss 1.6690 (1.8575) grad_norm 8.3544 (11.0233/4.2377) mem 68106MB [2022-12-19 05:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][760/1519] eta 0:12:40 lr 0.000034 time 0.9354 (1.0021) model_time 0.9352 (1.0012) loss 1.6712 (1.8580) grad_norm 6.7492 (10.9600/4.2041) mem 68106MB [2022-12-19 05:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][770/1519] eta 0:12:30 lr 0.000034 time 0.9364 (1.0021) model_time 0.9363 (1.0012) loss 2.1784 (1.8565) grad_norm 9.1774 (11.0566/4.2468) mem 68106MB [2022-12-19 05:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][780/1519] eta 0:12:20 lr 0.000034 time 0.9328 (1.0021) model_time 0.9326 (1.0012) loss 1.6249 (1.8547) grad_norm 8.8932 (11.0548/4.2515) mem 68106MB [2022-12-19 05:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][790/1519] eta 0:12:10 lr 0.000034 time 0.9269 (1.0021) model_time 0.9268 (1.0012) loss 2.2529 (1.8550) grad_norm 8.5384 (10.9951/4.2073) mem 68106MB [2022-12-19 05:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][800/1519] eta 0:12:00 lr 0.000034 time 0.9324 (1.0021) model_time 0.9323 (1.0012) loss 1.5258 (1.8528) grad_norm 7.2987 (10.9623/4.1645) mem 68106MB [2022-12-19 05:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][810/1519] eta 0:11:50 lr 0.000034 time 0.9288 (1.0020) model_time 0.9286 (1.0012) loss 1.8369 (1.8530) grad_norm 10.1119 (11.0105/4.2437) mem 68106MB [2022-12-19 05:12:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][820/1519] eta 0:11:40 lr 0.000034 time 0.9331 (1.0020) model_time 0.9330 (1.0012) loss 1.4146 (1.8527) grad_norm 20.9655 (11.0933/4.3155) mem 68106MB [2022-12-19 05:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][830/1519] eta 0:11:30 lr 0.000034 time 0.9346 (1.0020) model_time 0.9344 (1.0012) loss 2.0283 (1.8507) grad_norm 9.9101 (11.2633/4.6337) mem 68106MB [2022-12-19 05:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][840/1519] eta 0:11:20 lr 0.000034 time 0.9369 (1.0020) model_time 0.9368 (1.0012) loss 1.6544 (1.8482) grad_norm 9.4404 (11.2870/4.6382) mem 68106MB [2022-12-19 05:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][850/1519] eta 0:11:10 lr 0.000034 time 0.9338 (1.0020) model_time 0.9337 (1.0012) loss 1.6377 (1.8472) grad_norm 9.4292 (11.3198/4.6205) mem 68106MB [2022-12-19 05:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][860/1519] eta 0:11:00 lr 0.000034 time 0.9322 (1.0020) model_time 0.9320 (1.0012) loss 2.0710 (1.8475) grad_norm 12.1930 (11.3409/4.6616) mem 68106MB [2022-12-19 05:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][870/1519] eta 0:10:50 lr 0.000034 time 0.9359 (1.0020) model_time 0.9358 (1.0012) loss 1.7602 (1.8473) grad_norm 10.3049 (11.3613/4.6485) mem 68106MB [2022-12-19 05:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][880/1519] eta 0:10:40 lr 0.000034 time 0.9368 (1.0020) model_time 0.9367 (1.0012) loss 1.9192 (1.8466) grad_norm 8.3440 (11.3761/4.6889) mem 68106MB [2022-12-19 05:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][890/1519] eta 0:10:30 lr 0.000034 time 0.9379 (1.0020) model_time 0.9378 (1.0012) loss 1.9426 (1.8475) grad_norm 10.8310 (11.3833/4.6767) mem 68106MB [2022-12-19 05:14:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][900/1519] eta 0:10:20 lr 0.000034 time 0.9351 (1.0020) model_time 0.9349 (1.0012) loss 1.0756 (1.8445) grad_norm 16.8181 (11.3802/4.6510) mem 68106MB [2022-12-19 05:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][910/1519] eta 0:10:10 lr 0.000034 time 0.9376 (1.0020) model_time 0.9374 (1.0012) loss 1.7919 (1.8452) grad_norm 15.0720 (11.4098/4.6228) mem 68106MB [2022-12-19 05:14:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][920/1519] eta 0:10:00 lr 0.000034 time 0.9389 (1.0020) model_time 0.9387 (1.0012) loss 1.5362 (1.8444) grad_norm 8.8207 (11.4194/4.6305) mem 68106MB [2022-12-19 05:14:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][930/1519] eta 0:09:50 lr 0.000034 time 0.9349 (1.0020) model_time 0.9348 (1.0012) loss 1.5581 (1.8445) grad_norm 13.2456 (11.4353/4.6169) mem 68106MB [2022-12-19 05:14:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][940/1519] eta 0:09:40 lr 0.000034 time 0.9287 (1.0020) model_time 0.9286 (1.0012) loss 1.9270 (1.8453) grad_norm 6.7658 (11.3846/4.6063) mem 68106MB [2022-12-19 05:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][950/1519] eta 0:09:30 lr 0.000034 time 1.1657 (1.0023) model_time 1.1655 (1.0016) loss 1.8790 (1.8466) grad_norm 7.0825 (11.3737/4.6452) mem 68106MB [2022-12-19 05:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][960/1519] eta 0:09:20 lr 0.000034 time 0.9391 (1.0023) model_time 0.9390 (1.0015) loss 1.2140 (1.8430) grad_norm 12.6514 (11.4044/4.6524) mem 68106MB [2022-12-19 05:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][970/1519] eta 0:09:10 lr 0.000034 time 0.9282 (1.0023) model_time 0.9281 (1.0015) loss 1.8890 (1.8426) grad_norm 10.4221 (11.3452/4.6230) mem 68106MB [2022-12-19 05:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][980/1519] eta 0:09:00 lr 0.000034 time 0.9359 (1.0023) model_time 0.9358 (1.0015) loss 1.9708 (1.8408) grad_norm 12.1067 (11.3539/4.6051) mem 68106MB [2022-12-19 05:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][990/1519] eta 0:08:50 lr 0.000034 time 0.9370 (1.0023) model_time 0.9368 (1.0015) loss 1.6895 (1.8392) grad_norm 7.6189 (11.2939/4.5922) mem 68106MB [2022-12-19 05:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1000/1519] eta 0:08:40 lr 0.000034 time 0.9374 (1.0023) model_time 0.9373 (1.0015) loss 2.0876 (1.8383) grad_norm 17.0877 (11.3603/4.5917) mem 68106MB [2022-12-19 05:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1010/1519] eta 0:08:30 lr 0.000034 time 0.9344 (1.0022) model_time 0.9343 (1.0015) loss 1.0229 (1.8377) grad_norm 27.6568 (11.4116/4.6880) mem 68106MB [2022-12-19 05:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1020/1519] eta 0:08:20 lr 0.000034 time 0.9334 (1.0022) model_time 0.9333 (1.0015) loss 1.8759 (1.8375) grad_norm 7.8752 (11.3831/4.6415) mem 68106MB [2022-12-19 05:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1030/1519] eta 0:08:10 lr 0.000034 time 0.9346 (1.0022) model_time 0.9344 (1.0015) loss 2.0311 (1.8378) grad_norm 10.0831 (11.4316/4.6520) mem 68106MB [2022-12-19 05:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1040/1519] eta 0:08:00 lr 0.000034 time 0.9296 (1.0022) model_time 0.9294 (1.0015) loss 1.9049 (1.8375) grad_norm 20.6326 (11.4696/4.6827) mem 68106MB [2022-12-19 05:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1050/1519] eta 0:07:50 lr 0.000034 time 0.9320 (1.0022) model_time 0.9319 (1.0014) loss 1.6255 (1.8366) grad_norm 9.8051 (11.4420/4.6567) mem 68106MB [2022-12-19 05:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1060/1519] eta 0:07:40 lr 0.000034 time 0.8864 (1.0022) model_time 0.8863 (1.0015) loss 2.1215 (1.8359) grad_norm 10.1261 (11.3422/4.6125) mem 68106MB [2022-12-19 05:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1070/1519] eta 0:07:29 lr 0.000034 time 0.9425 (1.0022) model_time 0.9424 (1.0014) loss 1.5136 (1.8354) grad_norm 9.9197 (11.3208/4.6016) mem 68106MB [2022-12-19 05:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1080/1519] eta 0:07:19 lr 0.000034 time 0.9417 (1.0022) model_time 0.9415 (1.0014) loss 2.1148 (1.8358) grad_norm 10.8239 (11.3597/4.6035) mem 68106MB [2022-12-19 05:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1090/1519] eta 0:07:09 lr 0.000034 time 0.9358 (1.0021) model_time 0.9357 (1.0014) loss 1.7152 (1.8337) grad_norm 14.6914 (11.3279/4.6208) mem 68106MB [2022-12-19 05:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1100/1519] eta 0:06:59 lr 0.000034 time 0.9353 (1.0021) model_time 0.9351 (1.0014) loss 2.2994 (1.8327) grad_norm 10.8604 (11.2942/4.5962) mem 68106MB [2022-12-19 05:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1110/1519] eta 0:06:49 lr 0.000034 time 0.9406 (1.0021) model_time 0.9405 (1.0014) loss 1.6100 (1.8309) grad_norm 8.0509 (11.3875/4.6906) mem 68106MB [2022-12-19 05:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1120/1519] eta 0:06:39 lr 0.000034 time 0.9294 (1.0021) model_time 0.9293 (1.0014) loss 1.7976 (1.8309) grad_norm 7.6257 (11.4586/4.7561) mem 68106MB [2022-12-19 05:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1130/1519] eta 0:06:29 lr 0.000034 time 0.9367 (1.0021) model_time 0.9365 (1.0014) loss 1.1161 (1.8304) grad_norm 7.1744 (11.5016/4.7659) mem 68106MB [2022-12-19 05:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1140/1519] eta 0:06:19 lr 0.000034 time 0.9314 (1.0021) model_time 0.9312 (1.0014) loss 2.1280 (1.8304) grad_norm 10.1617 (11.5530/4.8251) mem 68106MB [2022-12-19 05:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1150/1519] eta 0:06:09 lr 0.000034 time 0.9344 (1.0021) model_time 0.9343 (1.0014) loss 2.1731 (1.8304) grad_norm 14.6381 (11.5499/4.8097) mem 68106MB [2022-12-19 05:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1160/1519] eta 0:05:59 lr 0.000034 time 0.9371 (1.0020) model_time 0.9370 (1.0013) loss 1.9099 (1.8294) grad_norm 9.2703 (11.5407/4.8100) mem 68106MB [2022-12-19 05:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1170/1519] eta 0:05:49 lr 0.000034 time 0.9353 (1.0020) model_time 0.9352 (1.0013) loss 1.2025 (1.8283) grad_norm 9.3003 (11.5328/4.8052) mem 68106MB [2022-12-19 05:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1180/1519] eta 0:05:39 lr 0.000034 time 0.9336 (1.0020) model_time 0.9335 (1.0013) loss 1.8020 (1.8272) grad_norm 13.0427 (11.4880/4.8094) mem 68106MB [2022-12-19 05:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1190/1519] eta 0:05:29 lr 0.000034 time 0.9368 (1.0020) model_time 0.9366 (1.0013) loss 1.8580 (1.8272) grad_norm 10.3352 (11.4051/4.8006) mem 68106MB [2022-12-19 05:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1200/1519] eta 0:05:19 lr 0.000034 time 0.9359 (1.0019) model_time 0.9358 (1.0013) loss 2.3166 (1.8284) grad_norm 5.4802 (11.4330/4.8865) mem 68106MB [2022-12-19 05:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1210/1519] eta 0:05:09 lr 0.000034 time 0.9325 (1.0019) model_time 0.9324 (1.0012) loss 1.2635 (1.8273) grad_norm 9.7546 (11.3418/4.8350) mem 68106MB [2022-12-19 05:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1220/1519] eta 0:04:59 lr 0.000034 time 0.9368 (1.0019) model_time 0.9367 (1.0012) loss 2.5842 (1.8273) grad_norm 15.3162 (11.4221/4.8582) mem 68106MB [2022-12-19 05:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1230/1519] eta 0:04:49 lr 0.000034 time 0.9348 (1.0019) model_time 0.9346 (1.0012) loss 1.7064 (1.8262) grad_norm 8.7234 (11.4130/4.8865) mem 68106MB [2022-12-19 05:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1240/1519] eta 0:04:39 lr 0.000034 time 0.9389 (1.0019) model_time 0.9387 (1.0012) loss 2.1356 (1.8255) grad_norm 17.0022 (11.4320/4.8537) mem 68106MB [2022-12-19 05:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1250/1519] eta 0:04:29 lr 0.000034 time 0.9371 (1.0019) model_time 0.9370 (1.0012) loss 1.9758 (1.8257) grad_norm 13.4669 (11.3732/4.8513) mem 68106MB [2022-12-19 05:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1260/1519] eta 0:04:19 lr 0.000034 time 0.9448 (1.0019) model_time 0.9446 (1.0012) loss 1.4687 (1.8249) grad_norm 20.4945 (11.4300/4.9357) mem 68106MB [2022-12-19 05:20:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1270/1519] eta 0:04:09 lr 0.000034 time 0.9356 (1.0020) model_time 0.9355 (1.0013) loss 2.0471 (1.8242) grad_norm 33.0988 (11.4797/5.0598) mem 68106MB [2022-12-19 05:20:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1280/1519] eta 0:03:59 lr 0.000034 time 0.9320 (1.0020) model_time 0.9318 (1.0013) loss 1.5590 (1.8225) grad_norm 26.4337 (11.5173/5.1290) mem 68106MB [2022-12-19 05:20:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1290/1519] eta 0:03:49 lr 0.000034 time 0.9572 (1.0020) model_time 0.9571 (1.0013) loss 1.6768 (1.8209) grad_norm 13.9162 (11.5671/5.1258) mem 68106MB [2022-12-19 05:20:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1300/1519] eta 0:03:39 lr 0.000034 time 0.9339 (1.0020) model_time 0.9338 (1.0013) loss 1.9022 (1.8197) grad_norm 8.3518 (11.7098/5.2237) mem 68106MB [2022-12-19 05:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1310/1519] eta 0:03:29 lr 0.000034 time 0.9339 (1.0020) model_time 0.9338 (1.0013) loss 1.9087 (1.8200) grad_norm 14.5914 (11.7497/5.2142) mem 68106MB [2022-12-19 05:21:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1320/1519] eta 0:03:19 lr 0.000034 time 0.9345 (1.0019) model_time 0.9343 (1.0013) loss 1.6908 (1.8198) grad_norm 16.6771 (11.7456/5.2201) mem 68106MB [2022-12-19 05:21:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1330/1519] eta 0:03:09 lr 0.000034 time 0.9340 (1.0019) model_time 0.9338 (1.0013) loss 1.5164 (1.8182) grad_norm 8.0452 (11.7482/5.2230) mem 68106MB [2022-12-19 05:21:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1340/1519] eta 0:02:59 lr 0.000034 time 0.9349 (1.0019) model_time 0.9348 (1.0013) loss 1.8373 (1.8168) grad_norm 10.2377 (11.6359/4.9934) mem 68106MB [2022-12-19 05:21:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1350/1519] eta 0:02:49 lr 0.000034 time 0.9258 (1.0019) model_time 0.9257 (1.0012) loss 1.5077 (1.8155) grad_norm 9.2915 (11.5716/4.9923) mem 68106MB [2022-12-19 05:21:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1360/1519] eta 0:02:39 lr 0.000034 time 0.9366 (1.0019) model_time 0.9365 (1.0013) loss 1.9766 (1.8153) grad_norm 24.1289 (11.7352/5.1359) mem 68106MB [2022-12-19 05:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1370/1519] eta 0:02:29 lr 0.000034 time 0.9295 (1.0019) model_time 0.9294 (1.0012) loss 1.3798 (1.8133) grad_norm 7.4129 (11.6694/5.1062) mem 68106MB [2022-12-19 05:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1380/1519] eta 0:02:19 lr 0.000034 time 0.9361 (1.0019) model_time 0.9359 (1.0013) loss 1.7732 (1.8124) grad_norm 7.3246 (11.6255/5.1149) mem 68106MB [2022-12-19 05:22:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1390/1519] eta 0:02:09 lr 0.000034 time 0.9355 (1.0019) model_time 0.9353 (1.0013) loss 2.0485 (1.8114) grad_norm 11.6435 (11.6397/5.0997) mem 68106MB [2022-12-19 05:22:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1400/1519] eta 0:01:59 lr 0.000034 time 0.9392 (1.0019) model_time 0.9390 (1.0013) loss 1.6408 (1.8115) grad_norm 14.9635 (11.6558/5.1085) mem 68106MB [2022-12-19 05:22:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1410/1519] eta 0:01:49 lr 0.000034 time 0.9352 (1.0019) model_time 0.9351 (1.0012) loss 1.3895 (1.8098) grad_norm 10.5693 (11.6119/5.0295) mem 68106MB [2022-12-19 05:22:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1420/1519] eta 0:01:39 lr 0.000034 time 0.9384 (1.0019) model_time 0.9382 (1.0012) loss 1.3575 (1.8083) grad_norm 16.2940 (11.5500/4.9712) mem 68106MB [2022-12-19 05:22:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1430/1519] eta 0:01:29 lr 0.000034 time 0.9318 (1.0018) model_time 0.9317 (1.0012) loss 2.1682 (1.8077) grad_norm 5.7424 (11.3957/4.7052) mem 68106MB [2022-12-19 05:23:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1440/1519] eta 0:01:19 lr 0.000034 time 0.9472 (1.0019) model_time 0.9471 (1.0013) loss 1.8633 (1.8069) grad_norm 9.7850 (11.3739/4.6993) mem 68106MB [2022-12-19 05:23:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1450/1519] eta 0:01:09 lr 0.000034 time 0.9299 (1.0021) model_time 0.9298 (1.0015) loss 2.5381 (1.8070) grad_norm 19.3197 (11.3930/4.7199) mem 68106MB [2022-12-19 05:23:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1460/1519] eta 0:00:59 lr 0.000034 time 0.9307 (1.0021) model_time 0.9305 (1.0014) loss 1.9594 (1.8061) grad_norm 12.7133 (11.3492/4.6895) mem 68106MB [2022-12-19 05:23:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1470/1519] eta 0:00:49 lr 0.000034 time 0.9392 (1.0021) model_time 0.9391 (1.0014) loss 1.9581 (1.8058) grad_norm 11.0300 (11.3482/4.6915) mem 68106MB [2022-12-19 05:23:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1480/1519] eta 0:00:39 lr 0.000034 time 0.9367 (1.0021) model_time 0.9366 (1.0014) loss 1.3169 (1.8044) grad_norm 12.0908 (11.3137/4.6628) mem 68106MB [2022-12-19 05:23:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1490/1519] eta 0:00:29 lr 0.000034 time 0.9401 (1.0020) model_time 0.9399 (1.0014) loss 1.3469 (1.8027) grad_norm 18.5703 (11.3075/4.6860) mem 68106MB [2022-12-19 05:24:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1500/1519] eta 0:00:19 lr 0.000034 time 0.9358 (1.0020) model_time 0.9356 (1.0014) loss 0.9725 (1.8024) grad_norm 8.2099 (11.2223/4.6465) mem 68106MB [2022-12-19 05:24:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [2/100][1510/1519] eta 0:00:09 lr 0.000034 time 0.9352 (1.0020) model_time 0.9351 (1.0014) loss 1.3720 (1.8013) grad_norm 8.3080 (11.1736/4.6504) mem 68106MB [2022-12-19 05:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 2 training takes 0:25:22 [2022-12-19 05:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_2.pth saving...... [2022-12-19 05:24:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_2.pth saved !!! [2022-12-19 05:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 3.5581 (3.5581) Acc@1 34.028 (34.028) Acc@5 64.583 (64.583) Mem 68106MB [2022-12-19 05:24:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.332) Loss 3.9254 (3.7342) Acc@1 26.389 (28.977) Acc@5 51.389 (58.365) Mem 68106MB [2022-12-19 05:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.316) Loss 3.5682 (3.7323) Acc@1 29.167 (28.968) Acc@5 62.500 (58.614) Mem 68106MB [2022-12-19 05:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.310 (0.311) Loss 3.8420 (3.7410) Acc@1 23.611 (28.394) Acc@5 55.556 (58.681) Mem 68106MB [2022-12-19 05:25:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.306 (0.308) Loss 3.6389 (3.7214) Acc@1 28.125 (28.667) Acc@5 60.764 (59.036) Mem 68106MB [2022-12-19 05:25:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 3.8018 (3.7163) Acc@1 25.347 (28.772) Acc@5 56.250 (59.055) Mem 68106MB [2022-12-19 05:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 3.9204 (3.7247) Acc@1 26.042 (28.632) Acc@5 51.042 (58.641) Mem 68106MB [2022-12-19 05:25:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 3.8936 (3.7327) Acc@1 29.514 (28.404) Acc@5 54.861 (58.470) Mem 68106MB [2022-12-19 05:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.307 (0.303) Loss 3.6771 (3.7311) Acc@1 27.431 (28.515) Acc@5 56.944 (58.582) Mem 68106MB [2022-12-19 05:25:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:2] * Acc@1 28.540 Acc@5 58.563 [2022-12-19 05:25:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 28.5% [2022-12-19 05:25:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 05:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 05:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 28.54% [2022-12-19 05:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][0/1519] eta 0:33:06 lr 0.000034 time 1.3077 (1.3077) model_time 0.9154 (0.9154) loss 1.8843 (1.8843) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 05:25:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][10/1519] eta 0:25:49 lr 0.000034 time 0.9274 (1.0270) model_time 0.9273 (0.9909) loss 1.2415 (1.5618) grad_norm 7.9669 (8.9984/2.4540) mem 68106MB [2022-12-19 05:26:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][20/1519] eta 0:25:19 lr 0.000034 time 0.9261 (1.0138) model_time 0.9259 (0.9948) loss 1.8964 (1.5846) grad_norm 17.4197 (12.1580/6.3821) mem 68106MB [2022-12-19 05:26:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][30/1519] eta 0:25:02 lr 0.000034 time 0.9246 (1.0090) model_time 0.9245 (0.9959) loss 1.0954 (1.5588) grad_norm 7.1558 (11.1450/5.6171) mem 68106MB [2022-12-19 05:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][40/1519] eta 0:24:48 lr 0.000034 time 0.9209 (1.0065) model_time 0.9207 (0.9965) loss 1.3819 (1.5828) grad_norm 7.9102 (11.1538/5.1583) mem 68106MB [2022-12-19 05:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][50/1519] eta 0:24:37 lr 0.000034 time 0.9329 (1.0055) model_time 0.9328 (0.9974) loss 1.7084 (1.5965) grad_norm 9.0756 (11.0507/4.7476) mem 68106MB [2022-12-19 05:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][60/1519] eta 0:24:26 lr 0.000034 time 0.9345 (1.0055) model_time 0.9344 (0.9986) loss 1.6389 (1.5971) grad_norm 7.0868 (12.4197/6.9606) mem 68106MB [2022-12-19 05:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][70/1519] eta 0:24:15 lr 0.000034 time 0.9349 (1.0044) model_time 0.9348 (0.9984) loss 1.8686 (1.6034) grad_norm 8.2634 (12.0134/6.5419) mem 68106MB [2022-12-19 05:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][80/1519] eta 0:24:04 lr 0.000034 time 0.9199 (1.0039) model_time 0.9197 (0.9986) loss 1.9679 (1.6241) grad_norm 7.6970 (11.7081/6.2249) mem 68106MB [2022-12-19 05:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][90/1519] eta 0:23:54 lr 0.000034 time 0.8844 (1.0040) model_time 0.8843 (0.9992) loss 1.2107 (1.6222) grad_norm 13.8600 (11.9715/6.1596) mem 68106MB [2022-12-19 05:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][100/1519] eta 0:23:43 lr 0.000034 time 0.9332 (1.0034) model_time 0.9330 (0.9991) loss 1.7599 (1.6180) grad_norm 10.9531 (11.7164/5.9316) mem 68106MB [2022-12-19 05:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][110/1519] eta 0:23:33 lr 0.000034 time 0.9395 (1.0032) model_time 0.9393 (0.9993) loss 1.8584 (1.6250) grad_norm 11.0746 (11.7230/5.8956) mem 68106MB [2022-12-19 05:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][120/1519] eta 0:23:22 lr 0.000034 time 0.9334 (1.0028) model_time 0.9332 (0.9992) loss 1.1618 (1.6184) grad_norm 15.3202 (11.6933/5.7643) mem 68106MB [2022-12-19 05:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][130/1519] eta 0:23:12 lr 0.000034 time 0.9348 (1.0025) model_time 0.9346 (0.9991) loss 1.8646 (1.6240) grad_norm 24.8352 (12.1670/6.1020) mem 68106MB [2022-12-19 05:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][140/1519] eta 0:23:02 lr 0.000034 time 0.9335 (1.0022) model_time 0.9332 (0.9990) loss 2.1708 (1.6345) grad_norm 10.6608 (12.3669/6.0506) mem 68106MB [2022-12-19 05:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][150/1519] eta 0:22:52 lr 0.000034 time 0.9344 (1.0022) model_time 0.9343 (0.9992) loss 1.3006 (1.6338) grad_norm 8.0534 (12.1242/5.9739) mem 68106MB [2022-12-19 05:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][160/1519] eta 0:22:42 lr 0.000034 time 0.9450 (1.0022) model_time 0.9448 (0.9994) loss 1.8616 (1.6502) grad_norm 10.0745 (11.9379/5.8990) mem 68106MB [2022-12-19 05:28:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][170/1519] eta 0:22:32 lr 0.000034 time 0.9495 (1.0024) model_time 0.9492 (0.9997) loss 1.2306 (1.6404) grad_norm 12.8359 (12.0800/5.8731) mem 68106MB [2022-12-19 05:28:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][180/1519] eta 0:22:22 lr 0.000034 time 0.9318 (1.0024) model_time 0.9317 (0.9999) loss 1.8540 (1.6469) grad_norm 7.7476 (11.9944/5.7598) mem 68106MB [2022-12-19 05:28:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][190/1519] eta 0:22:12 lr 0.000034 time 0.9408 (1.0026) model_time 0.9406 (1.0002) loss 2.0322 (1.6469) grad_norm 9.9984 (11.9583/5.6343) mem 68106MB [2022-12-19 05:29:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][200/1519] eta 0:22:02 lr 0.000034 time 0.9338 (1.0025) model_time 0.9336 (1.0001) loss 1.8826 (1.6446) grad_norm 15.6354 (12.2158/5.7236) mem 68106MB [2022-12-19 05:29:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][210/1519] eta 0:21:52 lr 0.000034 time 0.9287 (1.0024) model_time 0.9285 (1.0002) loss 1.7061 (1.6424) grad_norm 7.3285 (12.0461/5.6409) mem 68106MB [2022-12-19 05:29:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][220/1519] eta 0:21:42 lr 0.000034 time 0.9432 (1.0023) model_time 0.9430 (1.0002) loss 1.4092 (1.6394) grad_norm 8.9713 (11.9268/5.5567) mem 68106MB [2022-12-19 05:29:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][230/1519] eta 0:21:32 lr 0.000034 time 0.9965 (1.0024) model_time 0.9963 (1.0004) loss 1.7767 (1.6418) grad_norm 10.5528 (11.8404/5.4837) mem 68106MB [2022-12-19 05:29:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][240/1519] eta 0:21:22 lr 0.000034 time 0.9337 (1.0028) model_time 0.9336 (1.0008) loss 1.5799 (1.6406) grad_norm 11.2994 (11.9646/5.6529) mem 68106MB [2022-12-19 05:29:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][250/1519] eta 0:21:12 lr 0.000034 time 0.9280 (1.0027) model_time 0.9278 (1.0007) loss 1.4976 (1.6350) grad_norm 13.0142 (11.9106/5.6230) mem 68106MB [2022-12-19 05:30:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][260/1519] eta 0:21:02 lr 0.000034 time 0.9339 (1.0026) model_time 0.9336 (1.0007) loss 1.9519 (1.6417) grad_norm 29.6723 (12.0424/5.7362) mem 68106MB [2022-12-19 05:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][270/1519] eta 0:20:52 lr 0.000034 time 0.9337 (1.0027) model_time 0.9335 (1.0008) loss 1.7827 (1.6430) grad_norm 26.1911 (12.0727/5.7909) mem 68106MB [2022-12-19 05:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][280/1519] eta 0:20:42 lr 0.000034 time 0.9331 (1.0025) model_time 0.9330 (1.0007) loss 1.3265 (1.6364) grad_norm 15.4824 (12.1567/5.8826) mem 68106MB [2022-12-19 05:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][290/1519] eta 0:20:31 lr 0.000034 time 0.9263 (1.0024) model_time 0.9262 (1.0006) loss 1.6016 (1.6366) grad_norm 9.0254 (12.0968/5.8202) mem 68106MB [2022-12-19 05:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][300/1519] eta 0:20:21 lr 0.000034 time 0.9304 (1.0023) model_time 0.9302 (1.0006) loss 1.1110 (1.6313) grad_norm 6.7978 (12.0255/5.7687) mem 68106MB [2022-12-19 05:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][310/1519] eta 0:20:11 lr 0.000034 time 0.9312 (1.0022) model_time 0.9310 (1.0006) loss 1.8489 (1.6339) grad_norm 5.5375 (12.0009/5.7273) mem 68106MB [2022-12-19 05:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][320/1519] eta 0:20:01 lr 0.000034 time 0.9465 (1.0022) model_time 0.9464 (1.0006) loss 1.4958 (1.6331) grad_norm 12.8887 (11.9712/5.6454) mem 68106MB [2022-12-19 05:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][330/1519] eta 0:19:51 lr 0.000034 time 0.9359 (1.0023) model_time 0.9357 (1.0007) loss 1.4985 (1.6330) grad_norm 9.7743 (11.9624/5.5793) mem 68106MB [2022-12-19 05:31:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][340/1519] eta 0:19:41 lr 0.000034 time 0.9335 (1.0022) model_time 0.9333 (1.0006) loss 2.2318 (1.6337) grad_norm 12.7025 (11.9557/5.5077) mem 68106MB [2022-12-19 05:31:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][350/1519] eta 0:19:31 lr 0.000034 time 0.9171 (1.0021) model_time 0.9169 (1.0006) loss 2.1377 (1.6309) grad_norm 10.4413 (11.9748/5.4757) mem 68106MB [2022-12-19 05:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][360/1519] eta 0:19:21 lr 0.000034 time 0.9279 (1.0020) model_time 0.9278 (1.0006) loss 1.3805 (1.6290) grad_norm 8.5275 (11.8604/5.4434) mem 68106MB [2022-12-19 05:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][370/1519] eta 0:19:11 lr 0.000034 time 0.9313 (1.0022) model_time 0.9311 (1.0008) loss 2.0054 (1.6336) grad_norm 9.5025 (11.7663/5.4018) mem 68106MB [2022-12-19 05:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][380/1519] eta 0:19:01 lr 0.000034 time 0.9369 (1.0022) model_time 0.9367 (1.0008) loss 1.7001 (1.6308) grad_norm 9.5315 (11.7466/5.3676) mem 68106MB [2022-12-19 05:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][390/1519] eta 0:18:51 lr 0.000034 time 0.9356 (1.0022) model_time 0.9355 (1.0008) loss 1.8033 (1.6313) grad_norm 10.0136 (11.6592/5.3284) mem 68106MB [2022-12-19 05:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][400/1519] eta 0:18:41 lr 0.000034 time 0.9307 (1.0021) model_time 0.9306 (1.0007) loss 1.0936 (1.6319) grad_norm 9.5206 (11.6140/5.2935) mem 68106MB [2022-12-19 05:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][410/1519] eta 0:18:31 lr 0.000034 time 1.0019 (1.0023) model_time 1.0017 (1.0010) loss 2.0620 (1.6320) grad_norm 8.9682 (11.5455/5.2473) mem 68106MB [2022-12-19 05:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][420/1519] eta 0:18:22 lr 0.000034 time 0.9328 (1.0028) model_time 0.9327 (1.0015) loss 1.5791 (1.6295) grad_norm 9.8178 (11.4645/5.2183) mem 68106MB [2022-12-19 05:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][430/1519] eta 0:18:11 lr 0.000034 time 0.9388 (1.0027) model_time 0.9387 (1.0015) loss 1.7468 (1.6283) grad_norm 7.8310 (11.3945/5.1883) mem 68106MB [2022-12-19 05:33:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][440/1519] eta 0:18:01 lr 0.000034 time 0.9464 (1.0027) model_time 0.9463 (1.0015) loss 1.4958 (1.6252) grad_norm 15.6922 (11.4038/5.1417) mem 68106MB [2022-12-19 05:33:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][450/1519] eta 0:17:51 lr 0.000034 time 0.9370 (1.0026) model_time 0.9368 (1.0014) loss 1.4879 (1.6242) grad_norm 7.6950 (11.3565/5.1064) mem 68106MB [2022-12-19 05:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][460/1519] eta 0:17:41 lr 0.000034 time 0.9317 (1.0026) model_time 0.9316 (1.0014) loss 1.4459 (1.6233) grad_norm 9.1362 (11.3227/5.0688) mem 68106MB [2022-12-19 05:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][470/1519] eta 0:17:31 lr 0.000034 time 0.9317 (1.0026) model_time 0.9316 (1.0014) loss 1.7122 (1.6238) grad_norm 8.0583 (11.3187/5.0307) mem 68106MB [2022-12-19 05:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][480/1519] eta 0:17:21 lr 0.000034 time 0.9361 (1.0026) model_time 0.9360 (1.0014) loss 1.4295 (1.6207) grad_norm 6.8354 (11.2388/5.0141) mem 68106MB [2022-12-19 05:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][490/1519] eta 0:17:11 lr 0.000034 time 0.9329 (1.0026) model_time 0.9327 (1.0014) loss 1.3215 (1.6181) grad_norm 6.3217 (11.1926/4.9803) mem 68106MB [2022-12-19 05:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][500/1519] eta 0:17:01 lr 0.000034 time 0.9307 (1.0025) model_time 0.9306 (1.0014) loss 1.4911 (1.6169) grad_norm 9.1257 (11.2214/4.9776) mem 68106MB [2022-12-19 05:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][510/1519] eta 0:16:51 lr 0.000034 time 0.9332 (1.0024) model_time 0.9331 (1.0013) loss 1.3381 (1.6184) grad_norm 11.9064 (11.2305/4.9319) mem 68106MB [2022-12-19 05:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][520/1519] eta 0:16:41 lr 0.000034 time 0.9363 (1.0024) model_time 0.9361 (1.0013) loss 1.4691 (1.6198) grad_norm 8.4460 (11.2509/4.9069) mem 68106MB [2022-12-19 05:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][530/1519] eta 0:16:31 lr 0.000034 time 0.9379 (1.0024) model_time 0.9378 (1.0013) loss 1.3524 (1.6222) grad_norm 7.6387 (11.2441/4.8898) mem 68106MB [2022-12-19 05:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][540/1519] eta 0:16:21 lr 0.000034 time 0.9301 (1.0023) model_time 0.9300 (1.0012) loss 1.4440 (1.6224) grad_norm 5.9374 (11.1773/4.8736) mem 68106MB [2022-12-19 05:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][550/1519] eta 0:16:11 lr 0.000034 time 0.9422 (1.0023) model_time 0.9421 (1.0012) loss 0.8940 (1.6207) grad_norm 9.6399 (11.1395/4.8384) mem 68106MB [2022-12-19 05:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][560/1519] eta 0:16:01 lr 0.000034 time 0.9336 (1.0022) model_time 0.9335 (1.0012) loss 1.6227 (1.6197) grad_norm 5.9903 (11.1031/4.8104) mem 68106MB [2022-12-19 05:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][570/1519] eta 0:15:51 lr 0.000034 time 0.9341 (1.0022) model_time 0.9340 (1.0011) loss 1.0555 (1.6183) grad_norm 12.4515 (11.1076/4.7848) mem 68106MB [2022-12-19 05:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][580/1519] eta 0:15:41 lr 0.000034 time 0.9331 (1.0022) model_time 0.9330 (1.0011) loss 2.1178 (1.6176) grad_norm 7.1769 (11.0643/4.7563) mem 68106MB [2022-12-19 05:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][590/1519] eta 0:15:31 lr 0.000034 time 0.9396 (1.0022) model_time 0.9395 (1.0012) loss 1.1522 (1.6147) grad_norm 14.7344 (11.0707/4.7236) mem 68106MB [2022-12-19 05:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][600/1519] eta 0:15:21 lr 0.000034 time 0.9315 (1.0022) model_time 0.9314 (1.0012) loss 1.3844 (1.6145) grad_norm 14.0195 (11.0892/4.7073) mem 68106MB [2022-12-19 05:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][610/1519] eta 0:15:11 lr 0.000034 time 0.9357 (1.0022) model_time 0.9356 (1.0012) loss 1.1353 (1.6127) grad_norm 8.6645 (11.0762/4.7090) mem 68106MB [2022-12-19 05:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][620/1519] eta 0:15:00 lr 0.000034 time 0.9375 (1.0022) model_time 0.9374 (1.0012) loss 1.7274 (1.6131) grad_norm 9.3770 (11.0046/4.5876) mem 68106MB [2022-12-19 05:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][630/1519] eta 0:14:50 lr 0.000034 time 0.9457 (1.0022) model_time 0.9456 (1.0013) loss 1.4044 (1.6105) grad_norm 8.4733 (11.0624/4.5864) mem 68106MB [2022-12-19 05:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][640/1519] eta 0:14:40 lr 0.000034 time 0.9308 (1.0023) model_time 0.9307 (1.0013) loss 1.0791 (1.6089) grad_norm 11.9708 (11.1213/4.6879) mem 68106MB [2022-12-19 05:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][650/1519] eta 0:14:30 lr 0.000034 time 0.9524 (1.0022) model_time 0.9523 (1.0013) loss 1.3189 (1.6055) grad_norm 6.6665 (11.1146/4.6929) mem 68106MB [2022-12-19 05:36:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][660/1519] eta 0:14:20 lr 0.000034 time 0.9473 (1.0023) model_time 0.9472 (1.0013) loss 1.7887 (1.6046) grad_norm 13.5899 (10.9982/4.3685) mem 68106MB [2022-12-19 05:36:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][670/1519] eta 0:14:10 lr 0.000034 time 0.9397 (1.0023) model_time 0.9396 (1.0014) loss 1.5121 (1.6026) grad_norm 20.1846 (11.0694/4.3984) mem 68106MB [2022-12-19 05:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][680/1519] eta 0:14:01 lr 0.000034 time 1.0086 (1.0025) model_time 1.0085 (1.0015) loss 1.5316 (1.6015) grad_norm 8.0045 (11.0805/4.4030) mem 68106MB [2022-12-19 05:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][690/1519] eta 0:13:51 lr 0.000034 time 0.9348 (1.0025) model_time 0.9346 (1.0016) loss 1.6709 (1.6049) grad_norm 9.7368 (10.9971/4.3456) mem 68106MB [2022-12-19 05:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][700/1519] eta 0:13:40 lr 0.000034 time 0.9407 (1.0024) model_time 0.9405 (1.0015) loss 1.4450 (1.6028) grad_norm 9.3698 (10.9869/4.3436) mem 68106MB [2022-12-19 05:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][710/1519] eta 0:13:31 lr 0.000034 time 0.9447 (1.0025) model_time 0.9445 (1.0016) loss 1.2915 (1.6014) grad_norm 7.3114 (10.9347/4.3009) mem 68106MB [2022-12-19 05:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][720/1519] eta 0:13:20 lr 0.000034 time 0.9231 (1.0025) model_time 0.9229 (1.0016) loss 1.8084 (1.6016) grad_norm 12.3891 (10.9550/4.2982) mem 68106MB [2022-12-19 05:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][730/1519] eta 0:13:11 lr 0.000034 time 0.9176 (1.0029) model_time 0.9175 (1.0020) loss 1.6567 (1.6029) grad_norm 12.0542 (10.8624/4.1334) mem 68106MB [2022-12-19 05:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][740/1519] eta 0:13:01 lr 0.000034 time 0.9467 (1.0030) model_time 0.9466 (1.0021) loss 1.3479 (1.6036) grad_norm 8.2776 (10.8230/4.1134) mem 68106MB [2022-12-19 05:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][750/1519] eta 0:12:51 lr 0.000034 time 0.9374 (1.0029) model_time 0.9373 (1.0020) loss 1.3285 (1.6042) grad_norm 11.4138 (10.8846/4.1003) mem 68106MB [2022-12-19 05:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][760/1519] eta 0:12:41 lr 0.000034 time 0.9380 (1.0029) model_time 0.9379 (1.0020) loss 1.6341 (1.6045) grad_norm 12.0156 (10.8905/4.0855) mem 68106MB [2022-12-19 05:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][770/1519] eta 0:12:31 lr 0.000034 time 0.9348 (1.0028) model_time 0.9347 (1.0020) loss 1.5179 (1.6016) grad_norm 10.3083 (10.8282/4.0383) mem 68106MB [2022-12-19 05:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][780/1519] eta 0:12:21 lr 0.000034 time 0.9343 (1.0028) model_time 0.9342 (1.0020) loss 1.7036 (1.6027) grad_norm 15.1319 (10.8201/4.0441) mem 68106MB [2022-12-19 05:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][790/1519] eta 0:12:11 lr 0.000034 time 0.9371 (1.0028) model_time 0.9369 (1.0019) loss 1.5943 (1.6018) grad_norm 7.2109 (10.7930/4.0521) mem 68106MB [2022-12-19 05:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][800/1519] eta 0:12:00 lr 0.000034 time 0.9320 (1.0028) model_time 0.9319 (1.0019) loss 1.3112 (1.6004) grad_norm 11.3263 (10.6769/3.9154) mem 68106MB [2022-12-19 05:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][810/1519] eta 0:11:50 lr 0.000034 time 0.9354 (1.0028) model_time 0.9353 (1.0019) loss 1.3122 (1.5985) grad_norm 9.5361 (10.6964/3.9133) mem 68106MB [2022-12-19 05:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][820/1519] eta 0:11:40 lr 0.000034 time 0.9425 (1.0028) model_time 0.9424 (1.0020) loss 1.6175 (1.5979) grad_norm 18.3324 (10.7077/3.9404) mem 68106MB [2022-12-19 05:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][830/1519] eta 0:11:30 lr 0.000034 time 0.9373 (1.0028) model_time 0.9371 (1.0020) loss 1.6479 (1.5974) grad_norm 11.5607 (10.7797/3.9609) mem 68106MB [2022-12-19 05:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][840/1519] eta 0:11:20 lr 0.000034 time 0.9287 (1.0028) model_time 0.9285 (1.0020) loss 1.5237 (1.5979) grad_norm 8.6030 (10.7228/3.7981) mem 68106MB [2022-12-19 05:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][850/1519] eta 0:11:10 lr 0.000034 time 0.9355 (1.0028) model_time 0.9353 (1.0020) loss 1.4824 (1.5952) grad_norm 8.7810 (10.7130/3.7703) mem 68106MB [2022-12-19 05:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][860/1519] eta 0:11:00 lr 0.000034 time 1.0132 (1.0029) model_time 1.0131 (1.0021) loss 2.2068 (1.5955) grad_norm 8.1679 (10.5756/3.6293) mem 68106MB [2022-12-19 05:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][870/1519] eta 0:10:50 lr 0.000034 time 0.9374 (1.0029) model_time 0.9372 (1.0021) loss 2.0080 (1.5962) grad_norm 16.1986 (10.5823/3.5295) mem 68106MB [2022-12-19 05:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][880/1519] eta 0:10:40 lr 0.000034 time 0.9367 (1.0029) model_time 0.9366 (1.0021) loss 1.2470 (1.5953) grad_norm 19.1927 (10.5832/3.4616) mem 68106MB [2022-12-19 05:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][890/1519] eta 0:10:30 lr 0.000034 time 0.9316 (1.0028) model_time 0.9315 (1.0020) loss 1.8336 (1.5961) grad_norm 10.8144 (10.5550/3.4522) mem 68106MB [2022-12-19 05:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][900/1519] eta 0:10:20 lr 0.000034 time 0.9330 (1.0029) model_time 0.9329 (1.0021) loss 1.3836 (1.5956) grad_norm 9.2084 (10.5239/3.4394) mem 68106MB [2022-12-19 05:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][910/1519] eta 0:10:10 lr 0.000034 time 0.9335 (1.0028) model_time 0.9334 (1.0021) loss 1.3380 (1.5950) grad_norm 15.4904 (10.5409/3.4313) mem 68106MB [2022-12-19 05:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][920/1519] eta 0:10:00 lr 0.000034 time 0.9396 (1.0028) model_time 0.9395 (1.0021) loss 1.4271 (1.5940) grad_norm 13.8142 (10.5330/3.4381) mem 68106MB [2022-12-19 05:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][930/1519] eta 0:09:50 lr 0.000034 time 0.9424 (1.0028) model_time 0.9423 (1.0021) loss 1.7043 (1.5928) grad_norm 17.7555 (10.5340/3.4640) mem 68106MB [2022-12-19 05:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][940/1519] eta 0:09:40 lr 0.000034 time 0.9319 (1.0028) model_time 0.9318 (1.0020) loss 1.1799 (1.5913) grad_norm 9.3106 (10.4695/3.4744) mem 68106MB [2022-12-19 05:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][950/1519] eta 0:09:30 lr 0.000034 time 0.9414 (1.0028) model_time 0.9413 (1.0020) loss 1.3552 (1.5911) grad_norm 10.4423 (10.4640/3.4520) mem 68106MB [2022-12-19 05:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][960/1519] eta 0:09:20 lr 0.000034 time 0.9299 (1.0027) model_time 0.9298 (1.0020) loss 1.8520 (1.5900) grad_norm 14.7936 (10.5079/3.4604) mem 68106MB [2022-12-19 05:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][970/1519] eta 0:09:10 lr 0.000034 time 0.9324 (1.0027) model_time 0.9323 (1.0020) loss 1.7193 (1.5903) grad_norm 10.5176 (10.5073/3.4657) mem 68106MB [2022-12-19 05:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][980/1519] eta 0:09:00 lr 0.000034 time 0.9337 (1.0027) model_time 0.9335 (1.0020) loss 1.1352 (1.5891) grad_norm 10.6587 (10.5066/3.4988) mem 68106MB [2022-12-19 05:42:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][990/1519] eta 0:08:50 lr 0.000034 time 0.9342 (1.0027) model_time 0.9341 (1.0020) loss 1.0231 (1.5876) grad_norm 10.6808 (10.5195/3.4921) mem 68106MB [2022-12-19 05:42:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1000/1519] eta 0:08:40 lr 0.000034 time 0.9355 (1.0027) model_time 0.9354 (1.0020) loss 1.0781 (1.5862) grad_norm 12.0322 (10.5206/3.4767) mem 68106MB [2022-12-19 05:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1010/1519] eta 0:08:30 lr 0.000034 time 0.9322 (1.0027) model_time 0.9320 (1.0019) loss 1.5164 (1.5853) grad_norm 17.4285 (10.5503/3.5212) mem 68106MB [2022-12-19 05:42:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1020/1519] eta 0:08:20 lr 0.000034 time 0.9289 (1.0026) model_time 0.9287 (1.0019) loss 1.6226 (1.5831) grad_norm 6.1010 (10.5738/3.5194) mem 68106MB [2022-12-19 05:42:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1030/1519] eta 0:08:10 lr 0.000034 time 0.9623 (1.0026) model_time 0.9621 (1.0019) loss 1.0597 (1.5821) grad_norm 11.6202 (10.6087/3.5245) mem 68106MB [2022-12-19 05:43:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1040/1519] eta 0:08:00 lr 0.000034 time 0.9489 (1.0027) model_time 0.9487 (1.0020) loss 1.5970 (1.5823) grad_norm 12.2618 (10.5811/3.5142) mem 68106MB [2022-12-19 05:43:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1050/1519] eta 0:07:50 lr 0.000034 time 0.9380 (1.0028) model_time 0.9379 (1.0021) loss 2.0810 (1.5827) grad_norm 11.2202 (10.5828/3.5211) mem 68106MB [2022-12-19 05:43:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1060/1519] eta 0:07:40 lr 0.000034 time 0.9230 (1.0028) model_time 0.9228 (1.0021) loss 1.6249 (1.5817) grad_norm 13.1883 (10.5946/3.5169) mem 68106MB [2022-12-19 05:43:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1070/1519] eta 0:07:30 lr 0.000034 time 0.9409 (1.0028) model_time 0.9408 (1.0021) loss 1.8183 (1.5818) grad_norm 8.3056 (10.5736/3.5436) mem 68106MB [2022-12-19 05:43:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1080/1519] eta 0:07:20 lr 0.000034 time 0.9308 (1.0028) model_time 0.9307 (1.0020) loss 1.4708 (1.5815) grad_norm 7.6679 (10.6082/3.5229) mem 68106MB [2022-12-19 05:43:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1090/1519] eta 0:07:10 lr 0.000034 time 0.9288 (1.0027) model_time 0.9287 (1.0020) loss 1.3717 (1.5822) grad_norm 12.9736 (10.6618/3.5260) mem 68106MB [2022-12-19 05:44:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1100/1519] eta 0:07:00 lr 0.000034 time 0.9332 (1.0027) model_time 0.9331 (1.0020) loss 1.3795 (1.5811) grad_norm 13.2889 (10.6695/3.5324) mem 68106MB [2022-12-19 05:44:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1110/1519] eta 0:06:50 lr 0.000034 time 0.9430 (1.0027) model_time 0.9429 (1.0020) loss 1.5244 (1.5804) grad_norm 7.6777 (10.6203/3.5391) mem 68106MB [2022-12-19 05:44:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1120/1519] eta 0:06:40 lr 0.000034 time 0.9398 (1.0027) model_time 0.9397 (1.0020) loss 1.6876 (1.5794) grad_norm 11.1803 (10.6135/3.5284) mem 68106MB [2022-12-19 05:44:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1130/1519] eta 0:06:30 lr 0.000034 time 0.9344 (1.0027) model_time 0.9343 (1.0020) loss 1.8619 (1.5799) grad_norm 11.6703 (10.6565/3.6046) mem 68106MB [2022-12-19 05:44:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1140/1519] eta 0:06:20 lr 0.000034 time 0.9341 (1.0027) model_time 0.9340 (1.0020) loss 1.8279 (1.5790) grad_norm 7.3226 (10.6814/3.6258) mem 68106MB [2022-12-19 05:44:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1150/1519] eta 0:06:09 lr 0.000034 time 0.9386 (1.0027) model_time 0.9385 (1.0020) loss 1.5573 (1.5780) grad_norm 5.0110 (10.6578/3.6453) mem 68106MB [2022-12-19 05:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1160/1519] eta 0:05:59 lr 0.000034 time 0.9364 (1.0027) model_time 0.9362 (1.0020) loss 1.1488 (1.5767) grad_norm 9.3405 (10.6629/3.6345) mem 68106MB [2022-12-19 05:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1170/1519] eta 0:05:49 lr 0.000034 time 0.9297 (1.0027) model_time 0.9296 (1.0020) loss 1.9348 (1.5765) grad_norm 6.9655 (10.6012/3.6360) mem 68106MB [2022-12-19 05:45:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1180/1519] eta 0:05:39 lr 0.000034 time 0.9360 (1.0027) model_time 0.9359 (1.0020) loss 1.5015 (1.5750) grad_norm 17.2830 (10.7152/3.7209) mem 68106MB [2022-12-19 05:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1190/1519] eta 0:05:29 lr 0.000034 time 0.9323 (1.0027) model_time 0.9321 (1.0021) loss 0.9825 (1.5740) grad_norm 14.4791 (10.6976/3.7263) mem 68106MB [2022-12-19 05:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1200/1519] eta 0:05:19 lr 0.000034 time 0.9325 (1.0027) model_time 0.9323 (1.0021) loss 1.5107 (1.5735) grad_norm 10.4590 (10.7339/3.8248) mem 68106MB [2022-12-19 05:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1210/1519] eta 0:05:09 lr 0.000034 time 0.9309 (1.0027) model_time 0.9308 (1.0021) loss 1.4807 (1.5722) grad_norm 15.2638 (10.8121/3.8342) mem 68106MB [2022-12-19 05:46:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1220/1519] eta 0:04:59 lr 0.000034 time 0.9343 (1.0027) model_time 0.9342 (1.0021) loss 1.1026 (1.5710) grad_norm 7.6109 (10.7837/3.8286) mem 68106MB [2022-12-19 05:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1230/1519] eta 0:04:49 lr 0.000034 time 0.9338 (1.0027) model_time 0.9337 (1.0021) loss 0.9716 (1.5702) grad_norm 6.8807 (10.7512/3.8143) mem 68106MB [2022-12-19 05:46:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1240/1519] eta 0:04:39 lr 0.000034 time 0.9370 (1.0027) model_time 0.9369 (1.0020) loss 1.8750 (1.5704) grad_norm 16.4953 (10.6531/3.6981) mem 68106MB [2022-12-19 05:46:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1250/1519] eta 0:04:29 lr 0.000034 time 0.9408 (1.0027) model_time 0.9406 (1.0020) loss 1.3966 (1.5683) grad_norm 9.8917 (10.7331/3.7539) mem 68106MB [2022-12-19 05:46:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1260/1519] eta 0:04:19 lr 0.000034 time 0.9310 (1.0027) model_time 0.9309 (1.0020) loss 1.4446 (1.5671) grad_norm 10.6747 (10.6897/3.7286) mem 68106MB [2022-12-19 05:46:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1270/1519] eta 0:04:09 lr 0.000034 time 0.9353 (1.0027) model_time 0.9352 (1.0020) loss 1.7121 (1.5657) grad_norm 25.3571 (10.6834/3.7852) mem 68106MB [2022-12-19 05:47:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1280/1519] eta 0:03:59 lr 0.000034 time 0.9345 (1.0027) model_time 0.9344 (1.0020) loss 1.7418 (1.5649) grad_norm 8.8008 (10.7139/3.8253) mem 68106MB [2022-12-19 05:47:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1290/1519] eta 0:03:49 lr 0.000034 time 0.9360 (1.0027) model_time 0.9359 (1.0020) loss 1.9760 (1.5649) grad_norm 16.8154 (10.7806/3.8471) mem 68106MB [2022-12-19 05:47:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1300/1519] eta 0:03:39 lr 0.000034 time 0.9314 (1.0027) model_time 0.9312 (1.0020) loss 1.5722 (1.5651) grad_norm 5.7409 (10.7693/3.8563) mem 68106MB [2022-12-19 05:47:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1310/1519] eta 0:03:29 lr 0.000034 time 0.9470 (1.0027) model_time 0.9469 (1.0020) loss 1.8704 (1.5645) grad_norm 13.1590 (10.8046/3.8592) mem 68106MB [2022-12-19 05:47:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1320/1519] eta 0:03:19 lr 0.000034 time 0.9290 (1.0027) model_time 0.9288 (1.0020) loss 1.4507 (1.5629) grad_norm 5.8688 (10.7346/3.8583) mem 68106MB [2022-12-19 05:47:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1330/1519] eta 0:03:09 lr 0.000034 time 0.9376 (1.0027) model_time 0.9375 (1.0020) loss 1.3963 (1.5630) grad_norm 9.9448 (10.7979/4.0692) mem 68106MB [2022-12-19 05:48:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1340/1519] eta 0:02:59 lr 0.000034 time 0.9334 (1.0027) model_time 0.9333 (1.0020) loss 1.4620 (1.5628) grad_norm 10.9459 (10.7863/4.0331) mem 68106MB [2022-12-19 05:48:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1350/1519] eta 0:02:49 lr 0.000034 time 0.9371 (1.0027) model_time 0.9370 (1.0021) loss 1.2253 (1.5622) grad_norm 7.6392 (10.7576/4.0223) mem 68106MB [2022-12-19 05:48:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1360/1519] eta 0:02:39 lr 0.000034 time 0.9427 (1.0029) model_time 0.9426 (1.0023) loss 1.5284 (1.5614) grad_norm 9.9290 (10.7624/4.0085) mem 68106MB [2022-12-19 05:48:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1370/1519] eta 0:02:29 lr 0.000034 time 0.9367 (1.0029) model_time 0.9366 (1.0023) loss 1.3043 (1.5602) grad_norm 21.1360 (10.7781/4.0311) mem 68106MB [2022-12-19 05:48:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1380/1519] eta 0:02:19 lr 0.000034 time 0.9323 (1.0029) model_time 0.9322 (1.0023) loss 1.5147 (1.5603) grad_norm 9.6358 (10.7746/4.0238) mem 68106MB [2022-12-19 05:48:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1390/1519] eta 0:02:09 lr 0.000034 time 0.9285 (1.0029) model_time 0.9283 (1.0023) loss 1.3223 (1.5600) grad_norm 8.1689 (10.8444/4.1865) mem 68106MB [2022-12-19 05:49:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1400/1519] eta 0:01:59 lr 0.000034 time 0.9389 (1.0029) model_time 0.9387 (1.0022) loss 2.1974 (1.5594) grad_norm 7.8638 (10.8291/4.1935) mem 68106MB [2022-12-19 05:49:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1410/1519] eta 0:01:49 lr 0.000034 time 0.9322 (1.0029) model_time 0.9321 (1.0022) loss 1.8463 (1.5598) grad_norm 11.7303 (10.8226/4.2070) mem 68106MB [2022-12-19 05:49:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1420/1519] eta 0:01:39 lr 0.000034 time 0.9333 (1.0028) model_time 0.9332 (1.0022) loss 1.0601 (1.5596) grad_norm 10.8007 (10.8053/4.1843) mem 68106MB [2022-12-19 05:49:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1430/1519] eta 0:01:29 lr 0.000034 time 0.9412 (1.0028) model_time 0.9411 (1.0022) loss 1.6246 (1.5594) grad_norm 9.1718 (10.7095/4.1637) mem 68106MB [2022-12-19 05:49:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1440/1519] eta 0:01:19 lr 0.000034 time 0.9407 (1.0028) model_time 0.9405 (1.0022) loss 1.5944 (1.5583) grad_norm 8.5774 (10.7086/4.1840) mem 68106MB [2022-12-19 05:49:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1450/1519] eta 0:01:09 lr 0.000034 time 0.9376 (1.0028) model_time 0.9375 (1.0022) loss 1.2442 (1.5574) grad_norm 12.3614 (10.7890/4.2105) mem 68106MB [2022-12-19 05:50:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1460/1519] eta 0:00:59 lr 0.000034 time 0.9387 (1.0028) model_time 0.9385 (1.0022) loss 1.1359 (1.5567) grad_norm 15.2190 (10.8461/4.2005) mem 68106MB [2022-12-19 05:50:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1470/1519] eta 0:00:49 lr 0.000034 time 0.9440 (1.0028) model_time 0.9439 (1.0022) loss 1.8927 (1.5565) grad_norm 10.3291 (10.7989/4.2142) mem 68106MB [2022-12-19 05:50:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1480/1519] eta 0:00:39 lr 0.000034 time 0.9337 (1.0028) model_time 0.9335 (1.0021) loss 1.9248 (1.5555) grad_norm 7.0301 (10.7382/4.1794) mem 68106MB [2022-12-19 05:50:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1490/1519] eta 0:00:29 lr 0.000034 time 0.9401 (1.0028) model_time 0.9400 (1.0022) loss 1.2712 (1.5541) grad_norm 7.1685 (10.7404/4.1753) mem 68106MB [2022-12-19 05:50:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1500/1519] eta 0:00:19 lr 0.000034 time 0.9401 (1.0028) model_time 0.9400 (1.0022) loss 2.0221 (1.5549) grad_norm 20.6163 (10.8262/4.2358) mem 68106MB [2022-12-19 05:50:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [3/100][1510/1519] eta 0:00:09 lr 0.000034 time 0.9248 (1.0027) model_time 0.9248 (1.0021) loss 1.7265 (1.5543) grad_norm 7.7930 (10.8347/4.2789) mem 68106MB [2022-12-19 05:51:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 3 training takes 0:25:23 [2022-12-19 05:51:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_3.pth saving...... [2022-12-19 05:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_3.pth saved !!! [2022-12-19 05:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.692 (0.692) Loss 2.6690 (2.6690) Acc@1 48.958 (48.958) Acc@5 78.472 (78.472) Mem 68106MB [2022-12-19 05:51:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.334) Loss 2.9091 (2.7979) Acc@1 41.667 (44.255) Acc@5 73.611 (74.874) Mem 68106MB [2022-12-19 05:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.317) Loss 2.6450 (2.7872) Acc@1 43.750 (44.263) Acc@5 78.125 (74.934) Mem 68106MB [2022-12-19 05:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.310) Loss 2.9088 (2.7932) Acc@1 42.361 (43.716) Acc@5 74.653 (74.888) Mem 68106MB [2022-12-19 05:51:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.308) Loss 2.7303 (2.7751) Acc@1 47.569 (43.877) Acc@5 76.736 (75.169) Mem 68106MB [2022-12-19 05:51:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.306) Loss 2.8783 (2.7708) Acc@1 39.583 (43.907) Acc@5 72.569 (75.252) Mem 68106MB [2022-12-19 05:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 2.9040 (2.7776) Acc@1 40.278 (43.699) Acc@5 75.000 (75.211) Mem 68106MB [2022-12-19 05:51:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.306 (0.305) Loss 2.9496 (2.7853) Acc@1 42.014 (43.594) Acc@5 73.264 (75.108) Mem 68106MB [2022-12-19 05:51:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.301 (0.304) Loss 2.7393 (2.7847) Acc@1 40.625 (43.720) Acc@5 79.167 (75.266) Mem 68106MB [2022-12-19 05:51:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:3] * Acc@1 43.676 Acc@5 75.213 [2022-12-19 05:51:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 43.7% [2022-12-19 05:51:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 05:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 05:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 43.68% [2022-12-19 05:52:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][0/1519] eta 0:35:34 lr 0.000034 time 1.4052 (1.4052) model_time 1.0054 (1.0054) loss 1.7747 (1.7747) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 05:52:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][10/1519] eta 0:26:05 lr 0.000034 time 0.9401 (1.0375) model_time 0.9400 (1.0008) loss 1.2156 (1.5554) grad_norm 6.2976 (8.9055/1.5734) mem 68106MB [2022-12-19 05:52:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][20/1519] eta 0:25:31 lr 0.000034 time 0.9308 (1.0216) model_time 0.9306 (1.0022) loss 1.8588 (1.5946) grad_norm 8.7209 (9.4324/2.7412) mem 68106MB [2022-12-19 05:52:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][30/1519] eta 0:25:19 lr 0.000034 time 0.9260 (1.0206) model_time 0.9259 (1.0073) loss 1.0259 (1.4732) grad_norm 14.9213 (9.3696/2.7853) mem 68106MB [2022-12-19 05:53:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][40/1519] eta 0:25:01 lr 0.000034 time 0.9233 (1.0154) model_time 0.9232 (1.0052) loss 2.2026 (1.4693) grad_norm 10.3433 (8.9794/2.6453) mem 68106MB [2022-12-19 05:53:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][50/1519] eta 0:24:47 lr 0.000034 time 0.9321 (1.0127) model_time 0.9318 (1.0045) loss 1.2583 (1.4603) grad_norm 14.0089 (10.3885/5.0739) mem 68106MB [2022-12-19 05:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][60/1519] eta 0:24:40 lr 0.000034 time 0.9410 (1.0148) model_time 0.9408 (1.0079) loss 1.4893 (1.4936) grad_norm 16.6219 (10.6765/4.8366) mem 68106MB [2022-12-19 05:53:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][70/1519] eta 0:24:27 lr 0.000034 time 0.9356 (1.0130) model_time 0.9355 (1.0069) loss 1.4747 (1.4898) grad_norm 15.9552 (11.4094/5.0281) mem 68106MB [2022-12-19 05:53:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][80/1519] eta 0:24:15 lr 0.000034 time 0.9368 (1.0114) model_time 0.9367 (1.0060) loss 1.5394 (1.5169) grad_norm 8.0067 (11.0571/4.8372) mem 68106MB [2022-12-19 05:53:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][90/1519] eta 0:24:03 lr 0.000034 time 0.9345 (1.0101) model_time 0.9343 (1.0053) loss 1.2562 (1.5067) grad_norm 9.2342 (11.2488/4.8602) mem 68106MB [2022-12-19 05:54:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][100/1519] eta 0:23:51 lr 0.000034 time 0.9310 (1.0090) model_time 0.9309 (1.0046) loss 1.7520 (1.5151) grad_norm 8.7781 (11.0035/4.6818) mem 68106MB [2022-12-19 05:54:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][110/1519] eta 0:23:40 lr 0.000034 time 0.9307 (1.0081) model_time 0.9306 (1.0041) loss 1.6091 (1.5216) grad_norm 10.0191 (10.9080/4.6511) mem 68106MB [2022-12-19 05:54:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][120/1519] eta 0:23:29 lr 0.000034 time 0.9368 (1.0074) model_time 0.9367 (1.0037) loss 1.5126 (1.5170) grad_norm 16.4673 (11.2169/4.6805) mem 68106MB [2022-12-19 05:54:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][130/1519] eta 0:23:18 lr 0.000034 time 0.9359 (1.0067) model_time 0.9357 (1.0033) loss 1.3627 (1.5123) grad_norm 5.1627 (11.2166/4.6177) mem 68106MB [2022-12-19 05:54:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][140/1519] eta 0:23:07 lr 0.000034 time 0.9282 (1.0062) model_time 0.9280 (1.0029) loss 1.3228 (1.5054) grad_norm 5.4728 (11.0855/4.6106) mem 68106MB [2022-12-19 05:54:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][150/1519] eta 0:22:56 lr 0.000034 time 0.9328 (1.0058) model_time 0.9327 (1.0027) loss 0.9867 (1.4988) grad_norm 6.1288 (10.8863/4.5320) mem 68106MB [2022-12-19 05:55:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][160/1519] eta 0:22:46 lr 0.000034 time 0.9059 (1.0054) model_time 0.9057 (1.0025) loss 1.3745 (1.4974) grad_norm 8.6178 (10.8246/4.4839) mem 68106MB [2022-12-19 05:55:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][170/1519] eta 0:22:35 lr 0.000034 time 0.9272 (1.0051) model_time 0.9271 (1.0023) loss 1.5601 (1.4904) grad_norm 12.6740 (10.9975/4.5706) mem 68106MB [2022-12-19 05:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][180/1519] eta 0:22:25 lr 0.000034 time 0.9284 (1.0047) model_time 0.9282 (1.0021) loss 1.8012 (1.4880) grad_norm 10.6757 (10.9326/4.4686) mem 68106MB [2022-12-19 05:55:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][190/1519] eta 0:22:14 lr 0.000034 time 0.9324 (1.0045) model_time 0.9323 (1.0020) loss 1.6269 (1.4891) grad_norm 7.0751 (10.8742/4.4027) mem 68106MB [2022-12-19 05:55:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][200/1519] eta 0:22:04 lr 0.000034 time 0.9472 (1.0044) model_time 0.9471 (1.0020) loss 1.1426 (1.4890) grad_norm 13.6365 (10.8695/4.3204) mem 68106MB [2022-12-19 05:55:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][210/1519] eta 0:21:54 lr 0.000034 time 0.9331 (1.0043) model_time 0.9330 (1.0020) loss 1.4419 (1.4891) grad_norm 12.3733 (10.8067/4.2851) mem 68106MB [2022-12-19 05:56:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][220/1519] eta 0:21:44 lr 0.000034 time 0.9328 (1.0042) model_time 0.9325 (1.0020) loss 1.3016 (1.4870) grad_norm 12.9213 (10.8758/4.3353) mem 68106MB [2022-12-19 05:56:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][230/1519] eta 0:21:34 lr 0.000034 time 0.9430 (1.0041) model_time 0.9429 (1.0020) loss 1.5433 (1.4789) grad_norm 13.2145 (10.9260/4.3388) mem 68106MB [2022-12-19 05:56:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][240/1519] eta 0:21:24 lr 0.000034 time 0.9142 (1.0039) model_time 0.9139 (1.0019) loss 1.4061 (1.4741) grad_norm 6.8150 (10.9188/4.3099) mem 68106MB [2022-12-19 05:56:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][250/1519] eta 0:21:13 lr 0.000034 time 0.9277 (1.0038) model_time 0.9275 (1.0018) loss 0.9884 (1.4701) grad_norm 9.5626 (10.9030/4.3023) mem 68106MB [2022-12-19 05:56:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][260/1519] eta 0:21:03 lr 0.000034 time 0.9246 (1.0039) model_time 0.9244 (1.0020) loss 1.0831 (1.4642) grad_norm 13.4468 (11.0148/4.3775) mem 68106MB [2022-12-19 05:56:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][270/1519] eta 0:20:53 lr 0.000034 time 0.9289 (1.0037) model_time 0.9287 (1.0018) loss 1.3116 (1.4621) grad_norm 8.9660 (11.1707/4.7162) mem 68106MB [2022-12-19 05:57:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][280/1519] eta 0:20:43 lr 0.000034 time 0.9128 (1.0037) model_time 0.9127 (1.0019) loss 1.4462 (1.4608) grad_norm 15.1786 (11.3393/4.7555) mem 68106MB [2022-12-19 05:57:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][290/1519] eta 0:20:33 lr 0.000034 time 0.9291 (1.0036) model_time 0.9289 (1.0018) loss 1.3167 (1.4581) grad_norm 11.4409 (11.3275/4.6818) mem 68106MB [2022-12-19 05:57:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][300/1519] eta 0:20:23 lr 0.000034 time 0.9327 (1.0035) model_time 0.9325 (1.0018) loss 1.1705 (1.4542) grad_norm 10.3958 (11.3334/4.6092) mem 68106MB [2022-12-19 05:57:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][310/1519] eta 0:20:13 lr 0.000034 time 0.9295 (1.0034) model_time 0.9294 (1.0018) loss 1.6646 (1.4539) grad_norm 7.5658 (11.3562/4.6235) mem 68106MB [2022-12-19 05:57:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][320/1519] eta 0:20:03 lr 0.000034 time 0.9307 (1.0034) model_time 0.9305 (1.0018) loss 1.8412 (1.4512) grad_norm 9.1681 (11.3528/4.6006) mem 68106MB [2022-12-19 05:57:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][330/1519] eta 0:19:53 lr 0.000034 time 0.9317 (1.0036) model_time 0.9316 (1.0020) loss 1.6034 (1.4551) grad_norm 8.2469 (11.2764/4.5570) mem 68106MB [2022-12-19 05:58:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][340/1519] eta 0:19:44 lr 0.000034 time 1.1807 (1.0042) model_time 1.1805 (1.0027) loss 1.3476 (1.4515) grad_norm 7.4251 (11.2461/4.5443) mem 68106MB [2022-12-19 05:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][350/1519] eta 0:19:33 lr 0.000034 time 0.9298 (1.0042) model_time 0.9296 (1.0026) loss 1.3699 (1.4467) grad_norm 21.6336 (11.2685/4.5594) mem 68106MB [2022-12-19 05:58:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][360/1519] eta 0:19:23 lr 0.000034 time 0.9315 (1.0040) model_time 0.9313 (1.0025) loss 1.4006 (1.4464) grad_norm 9.7772 (11.2668/4.5153) mem 68106MB [2022-12-19 05:58:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][370/1519] eta 0:19:13 lr 0.000034 time 0.9239 (1.0041) model_time 0.9238 (1.0026) loss 1.5702 (1.4493) grad_norm 13.1607 (11.3083/4.5445) mem 68106MB [2022-12-19 05:58:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][380/1519] eta 0:19:03 lr 0.000034 time 0.9314 (1.0039) model_time 0.9313 (1.0025) loss 1.9717 (1.4513) grad_norm 20.9505 (11.3622/4.5812) mem 68106MB [2022-12-19 05:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][390/1519] eta 0:18:53 lr 0.000034 time 0.9419 (1.0039) model_time 0.9417 (1.0024) loss 1.5704 (1.4489) grad_norm 9.7003 (11.3338/4.5330) mem 68106MB [2022-12-19 05:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][400/1519] eta 0:18:43 lr 0.000034 time 0.9300 (1.0038) model_time 0.9298 (1.0024) loss 1.2705 (1.4472) grad_norm 12.7953 (11.3502/4.4919) mem 68106MB [2022-12-19 05:59:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][410/1519] eta 0:18:33 lr 0.000034 time 0.9332 (1.0037) model_time 0.9331 (1.0023) loss 1.6983 (1.4455) grad_norm 14.9547 (11.3308/4.4872) mem 68106MB [2022-12-19 05:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][420/1519] eta 0:18:22 lr 0.000034 time 0.9297 (1.0035) model_time 0.9296 (1.0022) loss 1.6075 (1.4475) grad_norm 7.0877 (11.3490/4.4870) mem 68106MB [2022-12-19 05:59:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][430/1519] eta 0:18:12 lr 0.000034 time 0.9266 (1.0035) model_time 0.9265 (1.0021) loss 1.2618 (1.4463) grad_norm 11.2347 (11.3836/4.4490) mem 68106MB [2022-12-19 05:59:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][440/1519] eta 0:18:02 lr 0.000034 time 0.9353 (1.0034) model_time 0.9352 (1.0021) loss 1.4019 (1.4471) grad_norm 8.9800 (11.3542/4.4107) mem 68106MB [2022-12-19 05:59:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][450/1519] eta 0:17:52 lr 0.000034 time 0.9200 (1.0033) model_time 0.9199 (1.0021) loss 0.9826 (1.4463) grad_norm 10.7293 (11.3352/4.3840) mem 68106MB [2022-12-19 06:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][460/1519] eta 0:17:42 lr 0.000034 time 0.9305 (1.0033) model_time 0.9303 (1.0021) loss 1.3045 (1.4464) grad_norm 6.3171 (11.3000/4.3616) mem 68106MB [2022-12-19 06:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][470/1519] eta 0:17:32 lr 0.000034 time 0.9269 (1.0032) model_time 0.9267 (1.0020) loss 1.5526 (1.4453) grad_norm 6.8088 (11.2224/4.3509) mem 68106MB [2022-12-19 06:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][480/1519] eta 0:17:22 lr 0.000034 time 0.9388 (1.0032) model_time 0.9387 (1.0020) loss 1.5455 (1.4460) grad_norm 12.0556 (11.1973/4.3167) mem 68106MB [2022-12-19 06:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][490/1519] eta 0:17:12 lr 0.000034 time 0.9245 (1.0031) model_time 0.9243 (1.0019) loss 2.0176 (1.4443) grad_norm 7.1175 (11.1534/4.2867) mem 68106MB [2022-12-19 06:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][500/1519] eta 0:17:02 lr 0.000034 time 0.9267 (1.0032) model_time 0.9266 (1.0020) loss 1.7263 (1.4427) grad_norm 8.6808 (11.1468/4.2664) mem 68106MB [2022-12-19 06:00:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][510/1519] eta 0:16:52 lr 0.000034 time 0.9318 (1.0033) model_time 0.9316 (1.0021) loss 1.6114 (1.4421) grad_norm 17.7030 (11.1946/4.2520) mem 68106MB [2022-12-19 06:01:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][520/1519] eta 0:16:42 lr 0.000034 time 1.2189 (1.0038) model_time 1.2188 (1.0026) loss 1.6179 (1.4413) grad_norm 10.8554 (11.1814/4.2317) mem 68106MB [2022-12-19 06:01:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][530/1519] eta 0:16:32 lr 0.000034 time 0.9236 (1.0037) model_time 0.9234 (1.0025) loss 1.2872 (1.4421) grad_norm 10.8040 (11.2028/4.1991) mem 68106MB [2022-12-19 06:01:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][540/1519] eta 0:16:22 lr 0.000034 time 0.9295 (1.0036) model_time 0.9294 (1.0024) loss 1.7042 (1.4409) grad_norm 11.1647 (11.2312/4.1736) mem 68106MB [2022-12-19 06:01:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][550/1519] eta 0:16:12 lr 0.000034 time 0.9241 (1.0035) model_time 0.9240 (1.0024) loss 1.2760 (1.4397) grad_norm 10.4945 (11.2079/4.1617) mem 68106MB [2022-12-19 06:01:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][560/1519] eta 0:16:02 lr 0.000034 time 0.9377 (1.0035) model_time 0.9376 (1.0024) loss 1.3666 (1.4391) grad_norm 6.1411 (11.1727/4.1570) mem 68106MB [2022-12-19 06:01:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][570/1519] eta 0:15:52 lr 0.000034 time 0.9335 (1.0035) model_time 0.9333 (1.0024) loss 1.5146 (1.4399) grad_norm 7.0077 (11.1622/4.1433) mem 68106MB [2022-12-19 06:02:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][580/1519] eta 0:15:42 lr 0.000034 time 0.9345 (1.0034) model_time 0.9343 (1.0024) loss 0.9290 (1.4380) grad_norm 17.6413 (11.1602/4.1501) mem 68106MB [2022-12-19 06:02:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][590/1519] eta 0:15:32 lr 0.000034 time 0.9274 (1.0034) model_time 0.9270 (1.0023) loss 1.3198 (1.4377) grad_norm 14.2907 (11.2107/4.1509) mem 68106MB [2022-12-19 06:02:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][600/1519] eta 0:15:22 lr 0.000034 time 0.9360 (1.0035) model_time 0.9359 (1.0024) loss 1.8949 (1.4392) grad_norm 14.9062 (11.2205/4.1330) mem 68106MB [2022-12-19 06:02:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][610/1519] eta 0:15:12 lr 0.000034 time 0.9190 (1.0034) model_time 0.9188 (1.0024) loss 0.8905 (1.4385) grad_norm 28.2135 (11.3408/4.2794) mem 68106MB [2022-12-19 06:02:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][620/1519] eta 0:15:01 lr 0.000034 time 0.9230 (1.0033) model_time 0.9228 (1.0023) loss 1.3657 (1.4355) grad_norm 5.7160 (11.3286/4.2753) mem 68106MB [2022-12-19 06:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][630/1519] eta 0:14:51 lr 0.000034 time 0.9340 (1.0033) model_time 0.9338 (1.0023) loss 1.4424 (1.4352) grad_norm 8.5434 (11.3944/4.2885) mem 68106MB [2022-12-19 06:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][640/1519] eta 0:14:41 lr 0.000034 time 0.9281 (1.0033) model_time 0.9279 (1.0023) loss 2.1404 (1.4360) grad_norm 11.8666 (11.4298/4.2676) mem 68106MB [2022-12-19 06:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][650/1519] eta 0:14:31 lr 0.000034 time 0.9203 (1.0032) model_time 0.9201 (1.0022) loss 1.4984 (1.4345) grad_norm 11.2028 (11.3757/4.1234) mem 68106MB [2022-12-19 06:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][660/1519] eta 0:14:21 lr 0.000034 time 0.9327 (1.0032) model_time 0.9326 (1.0022) loss 1.6121 (1.4365) grad_norm 7.4335 (11.3069/4.1303) mem 68106MB [2022-12-19 06:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][670/1519] eta 0:14:11 lr 0.000034 time 0.9485 (1.0031) model_time 0.9484 (1.0022) loss 1.4964 (1.4350) grad_norm 20.6144 (11.1977/4.1350) mem 68106MB [2022-12-19 06:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][680/1519] eta 0:14:01 lr 0.000034 time 0.9346 (1.0035) model_time 0.9343 (1.0025) loss 1.2871 (1.4334) grad_norm 8.1742 (11.3333/4.2187) mem 68106MB [2022-12-19 06:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][690/1519] eta 0:13:51 lr 0.000034 time 0.9294 (1.0034) model_time 0.9293 (1.0025) loss 1.4207 (1.4337) grad_norm 13.3682 (11.3236/4.1837) mem 68106MB [2022-12-19 06:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][700/1519] eta 0:13:41 lr 0.000034 time 0.9277 (1.0035) model_time 0.9275 (1.0025) loss 1.3283 (1.4333) grad_norm 21.2876 (11.3867/4.2175) mem 68106MB [2022-12-19 06:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][710/1519] eta 0:13:31 lr 0.000034 time 0.9171 (1.0034) model_time 0.9170 (1.0025) loss 1.4980 (1.4323) grad_norm 12.9528 (11.4009/4.1889) mem 68106MB [2022-12-19 06:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][720/1519] eta 0:13:21 lr 0.000034 time 0.9330 (1.0035) model_time 0.9329 (1.0026) loss 1.5584 (1.4312) grad_norm 4.6036 (11.3233/4.1826) mem 68106MB [2022-12-19 06:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][730/1519] eta 0:13:11 lr 0.000034 time 0.9228 (1.0035) model_time 0.9227 (1.0025) loss 1.5906 (1.4295) grad_norm 7.7192 (11.2997/4.1740) mem 68106MB [2022-12-19 06:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][740/1519] eta 0:13:01 lr 0.000034 time 0.9238 (1.0035) model_time 0.9236 (1.0026) loss 1.0784 (1.4283) grad_norm 9.5282 (11.2947/4.1501) mem 68106MB [2022-12-19 06:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][750/1519] eta 0:12:51 lr 0.000034 time 0.9402 (1.0034) model_time 0.9401 (1.0025) loss 1.4124 (1.4276) grad_norm 12.5903 (11.3295/4.1345) mem 68106MB [2022-12-19 06:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][760/1519] eta 0:12:41 lr 0.000034 time 0.9336 (1.0034) model_time 0.9334 (1.0025) loss 1.4138 (1.4282) grad_norm 17.3272 (11.3593/4.1248) mem 68106MB [2022-12-19 06:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][770/1519] eta 0:12:31 lr 0.000034 time 0.9366 (1.0034) model_time 0.9364 (1.0025) loss 1.2768 (1.4268) grad_norm 23.0710 (11.3689/4.1451) mem 68106MB [2022-12-19 06:05:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][780/1519] eta 0:12:21 lr 0.000034 time 0.9399 (1.0033) model_time 0.9397 (1.0025) loss 1.4407 (1.4263) grad_norm 11.4116 (11.3987/4.1700) mem 68106MB [2022-12-19 06:05:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][790/1519] eta 0:12:11 lr 0.000034 time 0.9451 (1.0034) model_time 0.9449 (1.0025) loss 1.9452 (1.4262) grad_norm 15.7617 (11.3837/4.1889) mem 68106MB [2022-12-19 06:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][800/1519] eta 0:12:01 lr 0.000034 time 0.9319 (1.0033) model_time 0.9318 (1.0024) loss 1.5479 (1.4263) grad_norm 7.1384 (11.3549/4.2019) mem 68106MB [2022-12-19 06:05:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][810/1519] eta 0:11:51 lr 0.000034 time 0.9146 (1.0033) model_time 0.9143 (1.0025) loss 0.8718 (1.4261) grad_norm 10.3623 (11.3577/4.1919) mem 68106MB [2022-12-19 06:06:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][820/1519] eta 0:11:41 lr 0.000034 time 0.9317 (1.0033) model_time 0.9315 (1.0024) loss 1.1946 (1.4239) grad_norm 8.8617 (11.2804/4.1725) mem 68106MB [2022-12-19 06:06:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][830/1519] eta 0:11:31 lr 0.000034 time 0.9340 (1.0033) model_time 0.9338 (1.0025) loss 1.4732 (1.4239) grad_norm 13.3518 (11.3229/4.1634) mem 68106MB [2022-12-19 06:06:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][840/1519] eta 0:11:21 lr 0.000034 time 0.9252 (1.0037) model_time 0.9250 (1.0029) loss 1.0070 (1.4245) grad_norm 5.6280 (11.3282/4.2614) mem 68106MB [2022-12-19 06:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][850/1519] eta 0:11:11 lr 0.000034 time 0.9176 (1.0037) model_time 0.9174 (1.0029) loss 1.0972 (1.4226) grad_norm 13.7586 (11.3284/4.2391) mem 68106MB [2022-12-19 06:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][860/1519] eta 0:11:01 lr 0.000034 time 0.9268 (1.0038) model_time 0.9267 (1.0029) loss 1.5847 (1.4224) grad_norm 9.5724 (11.3184/4.2416) mem 68106MB [2022-12-19 06:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][870/1519] eta 0:10:51 lr 0.000034 time 0.9350 (1.0037) model_time 0.9349 (1.0029) loss 1.4510 (1.4216) grad_norm 7.3433 (11.2341/4.0582) mem 68106MB [2022-12-19 06:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][880/1519] eta 0:10:41 lr 0.000034 time 0.9356 (1.0037) model_time 0.9355 (1.0029) loss 1.3367 (1.4222) grad_norm 9.4821 (11.1476/4.0031) mem 68106MB [2022-12-19 06:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][890/1519] eta 0:10:31 lr 0.000034 time 0.9361 (1.0037) model_time 0.9360 (1.0029) loss 1.3068 (1.4211) grad_norm 11.2452 (11.1350/4.0106) mem 68106MB [2022-12-19 06:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][900/1519] eta 0:10:21 lr 0.000034 time 0.9284 (1.0037) model_time 0.9283 (1.0028) loss 1.4824 (1.4213) grad_norm 7.8502 (11.1072/4.0140) mem 68106MB [2022-12-19 06:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][910/1519] eta 0:10:11 lr 0.000034 time 0.9339 (1.0037) model_time 0.9337 (1.0029) loss 1.3595 (1.4202) grad_norm 5.8251 (11.1155/4.0203) mem 68106MB [2022-12-19 06:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][920/1519] eta 0:10:01 lr 0.000034 time 0.9716 (1.0037) model_time 0.9715 (1.0029) loss 0.9287 (1.4197) grad_norm 9.4240 (11.0967/4.0130) mem 68106MB [2022-12-19 06:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][930/1519] eta 0:09:51 lr 0.000034 time 0.9313 (1.0038) model_time 0.9312 (1.0029) loss 1.1481 (1.4186) grad_norm 10.2541 (11.1084/4.0141) mem 68106MB [2022-12-19 06:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][940/1519] eta 0:09:41 lr 0.000034 time 0.9323 (1.0037) model_time 0.9322 (1.0029) loss 1.2715 (1.4187) grad_norm 12.3254 (11.1283/3.9914) mem 68106MB [2022-12-19 06:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][950/1519] eta 0:09:31 lr 0.000034 time 0.9273 (1.0037) model_time 0.9270 (1.0029) loss 1.6910 (1.4177) grad_norm 11.8252 (11.0734/3.9555) mem 68106MB [2022-12-19 06:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][960/1519] eta 0:09:21 lr 0.000034 time 0.9544 (1.0037) model_time 0.9542 (1.0029) loss 0.8621 (1.4174) grad_norm 13.9333 (11.0703/3.9480) mem 68106MB [2022-12-19 06:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][970/1519] eta 0:09:11 lr 0.000034 time 0.9252 (1.0037) model_time 0.9250 (1.0029) loss 1.0054 (1.4159) grad_norm 9.8236 (11.0208/3.9046) mem 68106MB [2022-12-19 06:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][980/1519] eta 0:09:00 lr 0.000034 time 0.9315 (1.0037) model_time 0.9313 (1.0029) loss 1.1322 (1.4157) grad_norm 12.1195 (10.9781/3.8522) mem 68106MB [2022-12-19 06:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][990/1519] eta 0:08:50 lr 0.000034 time 0.9313 (1.0037) model_time 0.9312 (1.0029) loss 1.8269 (1.4150) grad_norm 20.4827 (10.9825/3.9021) mem 68106MB [2022-12-19 06:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1000/1519] eta 0:08:40 lr 0.000034 time 0.9412 (1.0036) model_time 0.9411 (1.0029) loss 1.1204 (1.4150) grad_norm 9.2959 (11.0037/4.0293) mem 68106MB [2022-12-19 06:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1010/1519] eta 0:08:30 lr 0.000034 time 0.9295 (1.0036) model_time 0.9294 (1.0028) loss 1.2366 (1.4141) grad_norm 8.8132 (10.9764/4.0126) mem 68106MB [2022-12-19 06:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1020/1519] eta 0:08:20 lr 0.000034 time 0.9264 (1.0037) model_time 0.9259 (1.0029) loss 1.5476 (1.4128) grad_norm 9.0232 (10.9167/3.9831) mem 68106MB [2022-12-19 06:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1030/1519] eta 0:08:10 lr 0.000034 time 0.9325 (1.0037) model_time 0.9323 (1.0029) loss 1.1862 (1.4124) grad_norm 5.8141 (10.8246/3.9994) mem 68106MB [2022-12-19 06:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1040/1519] eta 0:08:00 lr 0.000034 time 0.9344 (1.0037) model_time 0.9343 (1.0029) loss 1.8125 (1.4133) grad_norm 13.9942 (10.8742/4.0604) mem 68106MB [2022-12-19 06:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1050/1519] eta 0:07:50 lr 0.000034 time 0.9334 (1.0036) model_time 0.9332 (1.0029) loss 1.3398 (1.4130) grad_norm 10.7986 (10.8770/4.0498) mem 68106MB [2022-12-19 06:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1060/1519] eta 0:07:40 lr 0.000034 time 0.9321 (1.0036) model_time 0.9319 (1.0028) loss 1.3284 (1.4127) grad_norm 7.6214 (10.8385/4.0643) mem 68106MB [2022-12-19 06:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1070/1519] eta 0:07:30 lr 0.000034 time 0.9339 (1.0035) model_time 0.9338 (1.0028) loss 1.1465 (1.4142) grad_norm 6.2494 (10.8732/4.0693) mem 68106MB [2022-12-19 06:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1080/1519] eta 0:07:20 lr 0.000034 time 0.9331 (1.0035) model_time 0.9329 (1.0028) loss 1.5004 (1.4138) grad_norm 9.2418 (10.8828/4.0919) mem 68106MB [2022-12-19 06:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1090/1519] eta 0:07:10 lr 0.000034 time 0.9337 (1.0035) model_time 0.9335 (1.0028) loss 1.5825 (1.4125) grad_norm 8.2862 (10.8900/4.0994) mem 68106MB [2022-12-19 06:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1100/1519] eta 0:07:00 lr 0.000034 time 0.9368 (1.0035) model_time 0.9367 (1.0028) loss 1.2285 (1.4129) grad_norm 11.1929 (10.8603/4.0904) mem 68106MB [2022-12-19 06:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1110/1519] eta 0:06:50 lr 0.000034 time 0.9356 (1.0035) model_time 0.9355 (1.0027) loss 1.6128 (1.4120) grad_norm 10.3568 (10.7984/4.0789) mem 68106MB [2022-12-19 06:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1120/1519] eta 0:06:40 lr 0.000034 time 0.9310 (1.0034) model_time 0.9308 (1.0027) loss 1.4485 (1.4115) grad_norm 6.5047 (10.7727/4.0918) mem 68106MB [2022-12-19 06:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1130/1519] eta 0:06:30 lr 0.000034 time 0.9346 (1.0035) model_time 0.9345 (1.0027) loss 1.3865 (1.4109) grad_norm 8.9360 (10.7072/4.1003) mem 68106MB [2022-12-19 06:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1140/1519] eta 0:06:20 lr 0.000034 time 0.9316 (1.0034) model_time 0.9315 (1.0027) loss 1.3551 (1.4107) grad_norm 12.8413 (10.6605/4.1084) mem 68106MB [2022-12-19 06:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1150/1519] eta 0:06:10 lr 0.000034 time 0.9308 (1.0036) model_time 0.9306 (1.0029) loss 1.2938 (1.4096) grad_norm 6.8332 (10.6543/4.1054) mem 68106MB [2022-12-19 06:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1160/1519] eta 0:06:00 lr 0.000034 time 0.9342 (1.0035) model_time 0.9340 (1.0028) loss 1.6015 (1.4110) grad_norm 11.2124 (10.6866/4.0956) mem 68106MB [2022-12-19 06:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1170/1519] eta 0:05:50 lr 0.000034 time 0.8969 (1.0036) model_time 0.8967 (1.0029) loss 1.3741 (1.4107) grad_norm 18.2185 (10.6942/4.1076) mem 68106MB [2022-12-19 06:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1180/1519] eta 0:05:40 lr 0.000034 time 0.9557 (1.0036) model_time 0.9555 (1.0029) loss 1.1721 (1.4099) grad_norm 10.2694 (10.7572/4.1724) mem 68106MB [2022-12-19 06:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1190/1519] eta 0:05:30 lr 0.000034 time 0.9335 (1.0036) model_time 0.9333 (1.0028) loss 1.5629 (1.4107) grad_norm 9.6522 (10.7181/4.1553) mem 68106MB [2022-12-19 06:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1200/1519] eta 0:05:20 lr 0.000034 time 0.9296 (1.0035) model_time 0.9294 (1.0028) loss 1.6928 (1.4105) grad_norm 13.7845 (10.6927/4.1804) mem 68106MB [2022-12-19 06:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1210/1519] eta 0:05:10 lr 0.000034 time 0.9370 (1.0035) model_time 0.9368 (1.0028) loss 1.3560 (1.4111) grad_norm 14.2753 (10.7089/4.1052) mem 68106MB [2022-12-19 06:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1220/1519] eta 0:05:00 lr 0.000034 time 0.9431 (1.0035) model_time 0.9429 (1.0028) loss 1.6843 (1.4121) grad_norm 13.5017 (10.7284/4.0939) mem 68106MB [2022-12-19 06:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1230/1519] eta 0:04:50 lr 0.000034 time 0.9462 (1.0035) model_time 0.9460 (1.0028) loss 1.4722 (1.4122) grad_norm 11.9078 (10.6867/4.0579) mem 68106MB [2022-12-19 06:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1240/1519] eta 0:04:39 lr 0.000034 time 0.9425 (1.0035) model_time 0.9423 (1.0028) loss 1.8345 (1.4133) grad_norm 9.1708 (10.6677/4.0813) mem 68106MB [2022-12-19 06:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1250/1519] eta 0:04:29 lr 0.000034 time 0.9281 (1.0035) model_time 0.9279 (1.0028) loss 1.2328 (1.4125) grad_norm 12.0214 (10.6411/4.0735) mem 68106MB [2022-12-19 06:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1260/1519] eta 0:04:19 lr 0.000034 time 0.9316 (1.0035) model_time 0.9314 (1.0028) loss 1.1512 (1.4115) grad_norm 10.1299 (10.6494/4.0705) mem 68106MB [2022-12-19 06:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1270/1519] eta 0:04:09 lr 0.000034 time 0.9362 (1.0035) model_time 0.9360 (1.0028) loss 1.3196 (1.4114) grad_norm 14.1469 (10.6823/4.0156) mem 68106MB [2022-12-19 06:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1280/1519] eta 0:03:59 lr 0.000034 time 0.9379 (1.0034) model_time 0.9377 (1.0027) loss 1.5394 (1.4113) grad_norm 6.5840 (10.5379/3.9072) mem 68106MB [2022-12-19 06:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1290/1519] eta 0:03:49 lr 0.000034 time 0.9428 (1.0034) model_time 0.9427 (1.0027) loss 1.5291 (1.4118) grad_norm 6.0894 (10.5101/3.9217) mem 68106MB [2022-12-19 06:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1300/1519] eta 0:03:39 lr 0.000034 time 0.9395 (1.0035) model_time 0.9392 (1.0028) loss 1.9823 (1.4117) grad_norm 22.4489 (10.4976/3.9344) mem 68106MB [2022-12-19 06:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1310/1519] eta 0:03:29 lr 0.000034 time 0.9306 (1.0034) model_time 0.9305 (1.0028) loss 1.0232 (1.4102) grad_norm 14.7882 (10.5454/3.9673) mem 68106MB [2022-12-19 06:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1320/1519] eta 0:03:19 lr 0.000034 time 0.9284 (1.0034) model_time 0.9282 (1.0027) loss 1.6029 (1.4105) grad_norm 15.5838 (10.5775/3.9438) mem 68106MB [2022-12-19 06:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1330/1519] eta 0:03:09 lr 0.000034 time 0.9320 (1.0034) model_time 0.9318 (1.0027) loss 0.8776 (1.4095) grad_norm 9.9365 (10.6199/3.9501) mem 68106MB [2022-12-19 06:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1340/1519] eta 0:02:59 lr 0.000034 time 0.9430 (1.0034) model_time 0.9429 (1.0027) loss 1.6169 (1.4096) grad_norm 13.3583 (10.6662/3.9883) mem 68106MB [2022-12-19 06:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1350/1519] eta 0:02:49 lr 0.000034 time 0.9397 (1.0034) model_time 0.9395 (1.0027) loss 1.3358 (1.4086) grad_norm 31.4178 (10.7433/4.1657) mem 68106MB [2022-12-19 06:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1360/1519] eta 0:02:39 lr 0.000034 time 0.9297 (1.0034) model_time 0.9295 (1.0027) loss 1.2707 (1.4078) grad_norm 6.0384 (10.6931/4.1673) mem 68106MB [2022-12-19 06:15:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1370/1519] eta 0:02:29 lr 0.000034 time 0.9422 (1.0034) model_time 0.9420 (1.0027) loss 1.2570 (1.4073) grad_norm 7.3849 (10.6403/4.1015) mem 68106MB [2022-12-19 06:15:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1380/1519] eta 0:02:19 lr 0.000034 time 0.9454 (1.0034) model_time 0.9452 (1.0027) loss 1.4824 (1.4076) grad_norm 7.0512 (10.5999/4.0791) mem 68106MB [2022-12-19 06:15:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1390/1519] eta 0:02:09 lr 0.000034 time 0.9340 (1.0034) model_time 0.9338 (1.0027) loss 1.5058 (1.4063) grad_norm 11.2944 (10.6389/4.0624) mem 68106MB [2022-12-19 06:15:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1400/1519] eta 0:01:59 lr 0.000034 time 0.9357 (1.0034) model_time 0.9355 (1.0027) loss 1.2683 (1.4061) grad_norm 9.4956 (10.6449/4.0580) mem 68106MB [2022-12-19 06:16:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1410/1519] eta 0:01:49 lr 0.000034 time 0.9426 (1.0034) model_time 0.9424 (1.0027) loss 1.1777 (1.4058) grad_norm 10.7739 (10.6919/4.0672) mem 68106MB [2022-12-19 06:16:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1420/1519] eta 0:01:39 lr 0.000034 time 0.9332 (1.0034) model_time 0.9330 (1.0027) loss 1.5266 (1.4045) grad_norm 8.1634 (10.7632/4.0696) mem 68106MB [2022-12-19 06:16:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1430/1519] eta 0:01:29 lr 0.000034 time 0.9314 (1.0033) model_time 0.9312 (1.0027) loss 1.6672 (1.4054) grad_norm 14.8078 (10.6821/4.0526) mem 68106MB [2022-12-19 06:16:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1440/1519] eta 0:01:19 lr 0.000034 time 0.9967 (1.0034) model_time 0.9965 (1.0027) loss 1.3235 (1.4055) grad_norm 11.5666 (10.7055/3.9489) mem 68106MB [2022-12-19 06:16:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1450/1519] eta 0:01:09 lr 0.000034 time 0.9411 (1.0034) model_time 0.9410 (1.0027) loss 1.5526 (1.4051) grad_norm 14.1879 (10.7131/3.9497) mem 68106MB [2022-12-19 06:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1460/1519] eta 0:00:59 lr 0.000034 time 0.9353 (1.0035) model_time 0.9351 (1.0028) loss 1.4711 (1.4052) grad_norm 12.5104 (10.7195/3.9407) mem 68106MB [2022-12-19 06:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1470/1519] eta 0:00:49 lr 0.000034 time 0.9285 (1.0035) model_time 0.9283 (1.0028) loss 1.7872 (1.4048) grad_norm 9.1146 (10.6987/3.9354) mem 68106MB [2022-12-19 06:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1480/1519] eta 0:00:39 lr 0.000034 time 0.9392 (1.0035) model_time 0.9390 (1.0028) loss 1.0505 (1.4044) grad_norm 5.1813 (10.6864/3.9397) mem 68106MB [2022-12-19 06:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1490/1519] eta 0:00:29 lr 0.000034 time 0.9429 (1.0036) model_time 0.9428 (1.0029) loss 0.9346 (1.4042) grad_norm 10.1360 (10.6493/3.9465) mem 68106MB [2022-12-19 06:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1500/1519] eta 0:00:19 lr 0.000034 time 0.9496 (1.0036) model_time 0.9494 (1.0029) loss 1.3667 (1.4044) grad_norm 13.7636 (10.6465/3.9527) mem 68106MB [2022-12-19 06:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [4/100][1510/1519] eta 0:00:09 lr 0.000034 time 0.9302 (1.0035) model_time 0.9301 (1.0029) loss 1.2869 (1.4042) grad_norm 18.3418 (10.6209/3.9186) mem 68106MB [2022-12-19 06:17:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 4 training takes 0:25:24 [2022-12-19 06:17:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_4.pth saving...... [2022-12-19 06:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_4.pth saved !!! [2022-12-19 06:18:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.661 (0.661) Loss 2.0746 (2.0746) Acc@1 59.722 (59.722) Acc@5 86.111 (86.111) Mem 68106MB [2022-12-19 06:18:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.331) Loss 2.2168 (2.1588) Acc@1 54.167 (56.439) Acc@5 86.111 (84.028) Mem 68106MB [2022-12-19 06:18:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.315) Loss 2.0016 (2.1472) Acc@1 59.375 (56.052) Acc@5 87.847 (84.392) Mem 68106MB [2022-12-19 06:18:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.310) Loss 2.2476 (2.1496) Acc@1 55.556 (55.701) Acc@5 85.069 (84.353) Mem 68106MB [2022-12-19 06:18:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.309 (0.307) Loss 2.1438 (2.1329) Acc@1 57.639 (55.784) Acc@5 82.986 (84.443) Mem 68106MB [2022-12-19 06:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.306) Loss 2.1929 (2.1252) Acc@1 56.597 (55.842) Acc@5 81.597 (84.579) Mem 68106MB [2022-12-19 06:18:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 2.2410 (2.1275) Acc@1 53.819 (55.760) Acc@5 84.375 (84.677) Mem 68106MB [2022-12-19 06:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.303 (0.304) Loss 2.3050 (2.1339) Acc@1 55.208 (55.825) Acc@5 79.514 (84.585) Mem 68106MB [2022-12-19 06:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.300 (0.303) Loss 2.1017 (2.1341) Acc@1 54.514 (55.838) Acc@5 85.764 (84.624) Mem 68106MB [2022-12-19 06:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:4] * Acc@1 55.812 Acc@5 84.569 [2022-12-19 06:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 55.8% [2022-12-19 06:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 06:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 06:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 55.81% [2022-12-19 06:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][0/1519] eta 0:34:33 lr 0.000034 time 1.3651 (1.3651) model_time 0.9698 (0.9698) loss 1.2325 (1.2325) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 06:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][10/1519] eta 0:25:58 lr 0.000034 time 0.9396 (1.0328) model_time 0.9395 (0.9966) loss 1.3043 (1.3356) grad_norm 9.9233 (9.3698/1.1965) mem 68106MB [2022-12-19 06:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][20/1519] eta 0:25:23 lr 0.000034 time 0.9257 (1.0161) model_time 0.9253 (0.9969) loss 1.2503 (1.3617) grad_norm 11.3981 (9.1595/2.0899) mem 68106MB [2022-12-19 06:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][30/1519] eta 0:25:04 lr 0.000034 time 0.9253 (1.0101) model_time 0.9247 (0.9970) loss 1.6356 (1.3528) grad_norm 9.2943 (9.0771/1.8171) mem 68106MB [2022-12-19 06:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][40/1519] eta 0:24:49 lr 0.000034 time 0.9217 (1.0071) model_time 0.9215 (0.9971) loss 1.6085 (1.3455) grad_norm 8.7992 (8.6579/1.9632) mem 68106MB [2022-12-19 06:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][50/1519] eta 0:24:37 lr 0.000034 time 0.9325 (1.0058) model_time 0.9324 (0.9976) loss 1.1637 (1.3596) grad_norm 8.3404 (8.7719/2.2439) mem 68106MB [2022-12-19 06:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][60/1519] eta 0:24:25 lr 0.000034 time 0.9037 (1.0045) model_time 0.9031 (0.9976) loss 1.3976 (1.3748) grad_norm 12.1989 (9.0843/2.5154) mem 68106MB [2022-12-19 06:20:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][70/1519] eta 0:24:14 lr 0.000034 time 0.9325 (1.0037) model_time 0.9324 (0.9977) loss 1.5208 (1.3764) grad_norm 7.4539 (8.9918/2.4868) mem 68106MB [2022-12-19 06:20:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][80/1519] eta 0:24:03 lr 0.000034 time 0.9331 (1.0032) model_time 0.9329 (0.9979) loss 1.1201 (1.3707) grad_norm 6.0248 (9.4066/3.1255) mem 68106MB [2022-12-19 06:20:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][90/1519] eta 0:23:52 lr 0.000034 time 0.9322 (1.0026) model_time 0.9320 (0.9979) loss 0.8723 (1.3650) grad_norm 8.7340 (9.4989/3.1126) mem 68106MB [2022-12-19 06:20:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][100/1519] eta 0:23:42 lr 0.000034 time 0.9213 (1.0024) model_time 0.9212 (0.9980) loss 1.0935 (1.3624) grad_norm 6.5617 (9.3709/2.9980) mem 68106MB [2022-12-19 06:20:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][110/1519] eta 0:23:32 lr 0.000034 time 0.9282 (1.0025) model_time 0.9280 (0.9985) loss 1.1197 (1.3444) grad_norm 6.4608 (9.4610/3.2998) mem 68106MB [2022-12-19 06:21:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][120/1519] eta 0:23:22 lr 0.000034 time 0.9272 (1.0022) model_time 0.9271 (0.9985) loss 1.2550 (1.3322) grad_norm 8.2772 (9.3617/3.2333) mem 68106MB [2022-12-19 06:21:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][130/1519] eta 0:23:11 lr 0.000034 time 0.9316 (1.0021) model_time 0.9314 (0.9987) loss 1.5348 (1.3311) grad_norm 13.6422 (9.5108/3.3339) mem 68106MB [2022-12-19 06:21:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][140/1519] eta 0:23:01 lr 0.000034 time 0.9249 (1.0021) model_time 0.9225 (0.9988) loss 1.1878 (1.3198) grad_norm 8.0041 (9.5028/3.2689) mem 68106MB [2022-12-19 06:21:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][150/1519] eta 0:22:53 lr 0.000034 time 1.1716 (1.0035) model_time 1.1715 (1.0005) loss 1.2891 (1.3222) grad_norm 9.9192 (9.6460/3.2869) mem 68106MB [2022-12-19 06:21:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][160/1519] eta 0:22:43 lr 0.000034 time 0.9342 (1.0032) model_time 0.9340 (1.0003) loss 0.8634 (1.3221) grad_norm 7.9842 (9.7898/3.5372) mem 68106MB [2022-12-19 06:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][170/1519] eta 0:22:33 lr 0.000034 time 0.9312 (1.0033) model_time 0.9311 (1.0005) loss 1.1847 (1.3238) grad_norm 10.7507 (9.8079/3.4766) mem 68106MB [2022-12-19 06:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][180/1519] eta 0:22:23 lr 0.000034 time 0.9281 (1.0036) model_time 0.9279 (1.0010) loss 1.2084 (1.3207) grad_norm 7.0152 (9.7149/3.4181) mem 68106MB [2022-12-19 06:22:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][190/1519] eta 0:22:13 lr 0.000034 time 0.9937 (1.0037) model_time 0.9935 (1.0012) loss 1.4586 (1.3233) grad_norm 17.5109 (9.8397/3.5182) mem 68106MB [2022-12-19 06:22:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][200/1519] eta 0:22:03 lr 0.000034 time 0.9300 (1.0037) model_time 0.9298 (1.0014) loss 1.0641 (1.3260) grad_norm 11.6213 (9.8639/3.4503) mem 68106MB [2022-12-19 06:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][210/1519] eta 0:21:55 lr 0.000034 time 0.9338 (1.0048) model_time 0.9337 (1.0026) loss 1.7764 (1.3261) grad_norm 12.2255 (9.8507/3.3858) mem 68106MB [2022-12-19 06:22:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][220/1519] eta 0:21:45 lr 0.000034 time 0.9367 (1.0048) model_time 0.9365 (1.0026) loss 1.5782 (1.3258) grad_norm 12.7683 (9.8982/3.3303) mem 68106MB [2022-12-19 06:22:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][230/1519] eta 0:21:34 lr 0.000034 time 0.9287 (1.0046) model_time 0.9285 (1.0025) loss 1.5315 (1.3279) grad_norm 8.7463 (9.8406/3.3043) mem 68106MB [2022-12-19 06:23:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][240/1519] eta 0:21:24 lr 0.000034 time 0.9335 (1.0045) model_time 0.9333 (1.0024) loss 1.3754 (1.3311) grad_norm 10.2777 (9.8216/3.2507) mem 68106MB [2022-12-19 06:23:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][250/1519] eta 0:21:14 lr 0.000034 time 0.9216 (1.0043) model_time 0.9215 (1.0023) loss 1.0922 (1.3342) grad_norm 19.5189 (9.9987/3.5124) mem 68106MB [2022-12-19 06:23:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][260/1519] eta 0:21:04 lr 0.000034 time 0.9349 (1.0042) model_time 0.9347 (1.0022) loss 1.2465 (1.3298) grad_norm 11.9052 (10.0890/3.5820) mem 68106MB [2022-12-19 06:23:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][270/1519] eta 0:20:54 lr 0.000034 time 0.9316 (1.0041) model_time 0.9315 (1.0022) loss 1.2070 (1.3266) grad_norm 25.9094 (10.1964/3.7984) mem 68106MB [2022-12-19 06:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][280/1519] eta 0:20:43 lr 0.000034 time 0.9302 (1.0039) model_time 0.9301 (1.0021) loss 0.9604 (1.3270) grad_norm 6.7090 (10.1245/3.7631) mem 68106MB [2022-12-19 06:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][290/1519] eta 0:20:33 lr 0.000034 time 0.9226 (1.0038) model_time 0.9225 (1.0020) loss 1.2116 (1.3271) grad_norm 13.3894 (10.1179/3.7331) mem 68106MB [2022-12-19 06:24:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][300/1519] eta 0:20:23 lr 0.000034 time 0.9297 (1.0038) model_time 0.9295 (1.0021) loss 1.6263 (1.3307) grad_norm 14.7057 (10.1898/3.7469) mem 68106MB [2022-12-19 06:24:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][310/1519] eta 0:20:13 lr 0.000034 time 0.9265 (1.0037) model_time 0.9264 (1.0020) loss 1.0203 (1.3300) grad_norm 9.6267 (10.3158/3.7965) mem 68106MB [2022-12-19 06:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][320/1519] eta 0:20:03 lr 0.000034 time 0.9441 (1.0038) model_time 0.9439 (1.0022) loss 1.1654 (1.3273) grad_norm 5.2603 (10.2828/3.7726) mem 68106MB [2022-12-19 06:24:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][330/1519] eta 0:19:53 lr 0.000034 time 0.9243 (1.0037) model_time 0.9241 (1.0021) loss 1.2805 (1.3259) grad_norm 7.9794 (10.2598/3.7271) mem 68106MB [2022-12-19 06:24:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][340/1519] eta 0:19:43 lr 0.000034 time 0.9441 (1.0037) model_time 0.9439 (1.0021) loss 1.3935 (1.3306) grad_norm 9.2079 (10.2656/3.7111) mem 68106MB [2022-12-19 06:24:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][350/1519] eta 0:19:33 lr 0.000034 time 0.9282 (1.0037) model_time 0.9280 (1.0021) loss 1.6503 (1.3291) grad_norm 6.1701 (10.2078/3.6831) mem 68106MB [2022-12-19 06:25:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][360/1519] eta 0:19:23 lr 0.000034 time 0.9253 (1.0036) model_time 0.9251 (1.0021) loss 1.1226 (1.3292) grad_norm 15.7125 (10.3170/3.7636) mem 68106MB [2022-12-19 06:25:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][370/1519] eta 0:19:13 lr 0.000034 time 1.0128 (1.0037) model_time 1.0127 (1.0022) loss 1.2504 (1.3297) grad_norm 6.7621 (10.3439/3.7874) mem 68106MB [2022-12-19 06:25:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][380/1519] eta 0:19:03 lr 0.000034 time 0.9462 (1.0037) model_time 0.9460 (1.0022) loss 1.3174 (1.3277) grad_norm 7.5750 (10.4024/3.8565) mem 68106MB [2022-12-19 06:25:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][390/1519] eta 0:18:53 lr 0.000034 time 0.9278 (1.0035) model_time 0.9275 (1.0021) loss 1.3657 (1.3239) grad_norm 6.0465 (10.4408/3.8723) mem 68106MB [2022-12-19 06:25:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][400/1519] eta 0:18:42 lr 0.000034 time 0.9180 (1.0034) model_time 0.9178 (1.0021) loss 1.4751 (1.3221) grad_norm 10.0294 (10.3827/3.8448) mem 68106MB [2022-12-19 06:25:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][410/1519] eta 0:18:32 lr 0.000034 time 0.9279 (1.0033) model_time 0.9278 (1.0020) loss 1.1168 (1.3205) grad_norm 16.9568 (10.3769/3.8455) mem 68106MB [2022-12-19 06:26:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][420/1519] eta 0:18:22 lr 0.000034 time 0.9243 (1.0033) model_time 0.9241 (1.0019) loss 1.0388 (1.3178) grad_norm 6.4692 (10.3645/3.8150) mem 68106MB [2022-12-19 06:26:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][430/1519] eta 0:18:12 lr 0.000034 time 0.9266 (1.0032) model_time 0.9264 (1.0019) loss 1.4092 (1.3156) grad_norm 7.5605 (10.3162/3.7972) mem 68106MB [2022-12-19 06:26:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][440/1519] eta 0:18:02 lr 0.000034 time 0.9286 (1.0031) model_time 0.9284 (1.0018) loss 1.6488 (1.3150) grad_norm 6.9513 (10.3455/3.7992) mem 68106MB [2022-12-19 06:26:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][450/1519] eta 0:17:52 lr 0.000034 time 0.9278 (1.0030) model_time 0.9276 (1.0017) loss 1.3410 (1.3150) grad_norm 7.1031 (10.2669/3.7957) mem 68106MB [2022-12-19 06:26:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][460/1519] eta 0:17:42 lr 0.000034 time 0.9297 (1.0029) model_time 0.9296 (1.0016) loss 1.4599 (1.3147) grad_norm 10.8688 (10.2150/3.7796) mem 68106MB [2022-12-19 06:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][470/1519] eta 0:17:32 lr 0.000034 time 0.9199 (1.0029) model_time 0.9198 (1.0017) loss 1.6315 (1.3168) grad_norm 8.2557 (10.2523/3.7667) mem 68106MB [2022-12-19 06:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][480/1519] eta 0:17:22 lr 0.000034 time 0.9384 (1.0029) model_time 0.9382 (1.0017) loss 1.0457 (1.3182) grad_norm 12.4797 (10.2219/3.7447) mem 68106MB [2022-12-19 06:27:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][490/1519] eta 0:17:12 lr 0.000034 time 0.9361 (1.0030) model_time 0.9355 (1.0018) loss 1.1866 (1.3154) grad_norm 9.3086 (10.2713/3.7887) mem 68106MB [2022-12-19 06:27:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][500/1519] eta 0:17:01 lr 0.000034 time 0.9291 (1.0029) model_time 0.9289 (1.0017) loss 0.9884 (1.3180) grad_norm 6.5046 (10.2677/3.7800) mem 68106MB [2022-12-19 06:27:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][510/1519] eta 0:16:51 lr 0.000034 time 0.9424 (1.0029) model_time 0.9422 (1.0017) loss 1.2315 (1.3214) grad_norm 6.4110 (10.2076/3.7777) mem 68106MB [2022-12-19 06:27:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][520/1519] eta 0:16:42 lr 0.000034 time 0.9314 (1.0033) model_time 0.9312 (1.0021) loss 1.3887 (1.3240) grad_norm 9.7136 (10.2031/3.7545) mem 68106MB [2022-12-19 06:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][530/1519] eta 0:16:32 lr 0.000034 time 0.9304 (1.0033) model_time 0.9302 (1.0022) loss 1.2459 (1.3236) grad_norm 9.0945 (10.1977/3.7245) mem 68106MB [2022-12-19 06:28:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][540/1519] eta 0:16:22 lr 0.000034 time 0.9401 (1.0033) model_time 0.9399 (1.0022) loss 1.2290 (1.3241) grad_norm 13.1775 (10.2577/3.7419) mem 68106MB [2022-12-19 06:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][550/1519] eta 0:16:12 lr 0.000034 time 0.9275 (1.0033) model_time 0.9273 (1.0022) loss 1.0082 (1.3233) grad_norm 11.4778 (10.2692/3.7116) mem 68106MB [2022-12-19 06:28:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][560/1519] eta 0:16:02 lr 0.000034 time 0.9322 (1.0033) model_time 0.9317 (1.0022) loss 1.2520 (1.3222) grad_norm 12.6241 (10.2614/3.7050) mem 68106MB [2022-12-19 06:28:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][570/1519] eta 0:15:52 lr 0.000034 time 0.9322 (1.0033) model_time 0.9321 (1.0022) loss 1.3280 (1.3213) grad_norm 8.8025 (10.2396/3.6898) mem 68106MB [2022-12-19 06:28:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][580/1519] eta 0:15:42 lr 0.000034 time 0.9383 (1.0033) model_time 0.9382 (1.0022) loss 1.6600 (1.3228) grad_norm 9.4832 (10.2309/3.6708) mem 68106MB [2022-12-19 06:28:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][590/1519] eta 0:15:32 lr 0.000034 time 0.9348 (1.0033) model_time 0.9347 (1.0022) loss 1.3646 (1.3232) grad_norm 7.2203 (10.2603/3.6943) mem 68106MB [2022-12-19 06:29:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][600/1519] eta 0:15:21 lr 0.000034 time 0.9332 (1.0033) model_time 0.9330 (1.0022) loss 1.0073 (1.3224) grad_norm 5.3165 (10.3111/3.8911) mem 68106MB [2022-12-19 06:29:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][610/1519] eta 0:15:11 lr 0.000034 time 0.9367 (1.0032) model_time 0.9365 (1.0022) loss 1.5211 (1.3208) grad_norm 5.8760 (10.2864/3.9051) mem 68106MB [2022-12-19 06:29:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][620/1519] eta 0:15:01 lr 0.000034 time 0.9396 (1.0033) model_time 0.9394 (1.0022) loss 1.4486 (1.3195) grad_norm 7.3628 (10.2782/3.8953) mem 68106MB [2022-12-19 06:29:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][630/1519] eta 0:14:51 lr 0.000034 time 0.9695 (1.0033) model_time 0.9694 (1.0023) loss 1.9862 (1.3207) grad_norm 10.6896 (10.3351/3.9047) mem 68106MB [2022-12-19 06:29:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][640/1519] eta 0:14:41 lr 0.000034 time 0.9337 (1.0033) model_time 0.9336 (1.0023) loss 1.3582 (1.3220) grad_norm 13.0466 (10.3715/3.9008) mem 68106MB [2022-12-19 06:29:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][650/1519] eta 0:14:31 lr 0.000034 time 0.9308 (1.0032) model_time 0.9306 (1.0022) loss 1.1350 (1.3221) grad_norm 9.5082 (10.3714/3.8844) mem 68106MB [2022-12-19 06:30:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][660/1519] eta 0:14:21 lr 0.000034 time 0.9307 (1.0032) model_time 0.9306 (1.0022) loss 1.2699 (1.3212) grad_norm 11.5330 (10.4352/3.9899) mem 68106MB [2022-12-19 06:30:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][670/1519] eta 0:14:11 lr 0.000034 time 0.9515 (1.0032) model_time 0.9512 (1.0022) loss 0.8920 (1.3219) grad_norm 7.0988 (10.4815/3.9894) mem 68106MB [2022-12-19 06:30:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][680/1519] eta 0:14:01 lr 0.000034 time 0.9312 (1.0031) model_time 0.9310 (1.0021) loss 1.3513 (1.3241) grad_norm 12.2707 (10.4517/3.9354) mem 68106MB [2022-12-19 06:30:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][690/1519] eta 0:13:51 lr 0.000034 time 0.9380 (1.0032) model_time 0.9378 (1.0022) loss 1.2644 (1.3240) grad_norm 7.8282 (10.4297/3.9339) mem 68106MB [2022-12-19 06:30:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][700/1519] eta 0:13:41 lr 0.000034 time 0.9354 (1.0030) model_time 0.9352 (1.0020) loss 1.3174 (1.3258) grad_norm 8.3729 (10.4775/3.9568) mem 68106MB [2022-12-19 06:30:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][710/1519] eta 0:13:31 lr 0.000034 time 0.9385 (1.0029) model_time 0.9383 (1.0020) loss 1.8335 (1.3247) grad_norm 8.0458 (10.4725/3.9097) mem 68106MB [2022-12-19 06:31:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][720/1519] eta 0:13:21 lr 0.000034 time 0.9376 (1.0029) model_time 0.9375 (1.0020) loss 1.2844 (1.3248) grad_norm 13.8185 (10.5971/3.9741) mem 68106MB [2022-12-19 06:31:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][730/1519] eta 0:13:11 lr 0.000034 time 0.9213 (1.0029) model_time 0.9211 (1.0019) loss 1.2595 (1.3253) grad_norm 6.7393 (10.5848/3.9622) mem 68106MB [2022-12-19 06:31:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][740/1519] eta 0:13:01 lr 0.000034 time 0.9311 (1.0028) model_time 0.9308 (1.0019) loss 1.1804 (1.3238) grad_norm 14.9580 (10.5833/3.9638) mem 68106MB [2022-12-19 06:31:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][750/1519] eta 0:12:51 lr 0.000034 time 0.9300 (1.0028) model_time 0.9298 (1.0018) loss 1.4724 (1.3231) grad_norm 7.3071 (10.5633/3.9658) mem 68106MB [2022-12-19 06:31:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][760/1519] eta 0:12:41 lr 0.000034 time 0.9343 (1.0027) model_time 0.9340 (1.0018) loss 1.2191 (1.3235) grad_norm 14.5756 (10.5373/3.9057) mem 68106MB [2022-12-19 06:31:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][770/1519] eta 0:12:30 lr 0.000034 time 0.9302 (1.0027) model_time 0.9301 (1.0018) loss 1.2011 (1.3242) grad_norm 10.3176 (10.5645/3.9062) mem 68106MB [2022-12-19 06:32:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][780/1519] eta 0:12:21 lr 0.000034 time 0.9290 (1.0029) model_time 0.9289 (1.0020) loss 1.4585 (1.3242) grad_norm 7.4866 (10.5654/3.9127) mem 68106MB [2022-12-19 06:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][790/1519] eta 0:12:11 lr 0.000033 time 0.9293 (1.0030) model_time 0.9292 (1.0021) loss 1.7295 (1.3246) grad_norm 6.0434 (10.5553/3.9096) mem 68106MB [2022-12-19 06:32:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][800/1519] eta 0:12:01 lr 0.000033 time 0.9342 (1.0030) model_time 0.9340 (1.0021) loss 0.9545 (1.3250) grad_norm 17.5552 (10.5362/3.9157) mem 68106MB [2022-12-19 06:32:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][810/1519] eta 0:11:51 lr 0.000033 time 0.9297 (1.0030) model_time 0.9296 (1.0021) loss 1.1000 (1.3247) grad_norm 5.4449 (10.5143/3.9349) mem 68106MB [2022-12-19 06:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][820/1519] eta 0:11:41 lr 0.000033 time 0.9378 (1.0029) model_time 0.9376 (1.0021) loss 1.6915 (1.3245) grad_norm 11.0715 (10.5043/3.9354) mem 68106MB [2022-12-19 06:32:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][830/1519] eta 0:11:31 lr 0.000033 time 0.9309 (1.0032) model_time 0.9308 (1.0024) loss 1.6225 (1.3240) grad_norm 6.3989 (10.4751/3.9432) mem 68106MB [2022-12-19 06:33:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][840/1519] eta 0:11:21 lr 0.000033 time 0.9374 (1.0032) model_time 0.9373 (1.0023) loss 1.1422 (1.3222) grad_norm 7.0991 (10.5161/4.0223) mem 68106MB [2022-12-19 06:33:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][850/1519] eta 0:11:11 lr 0.000033 time 0.9343 (1.0031) model_time 0.9341 (1.0023) loss 1.0803 (1.3205) grad_norm 10.3507 (10.4752/3.9878) mem 68106MB [2022-12-19 06:33:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][860/1519] eta 0:11:01 lr 0.000033 time 0.9360 (1.0031) model_time 0.9359 (1.0023) loss 1.4228 (1.3192) grad_norm 6.4763 (10.3797/3.9211) mem 68106MB [2022-12-19 06:33:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][870/1519] eta 0:10:50 lr 0.000033 time 0.9275 (1.0031) model_time 0.9273 (1.0022) loss 1.1830 (1.3190) grad_norm 5.5806 (10.3316/3.9371) mem 68106MB [2022-12-19 06:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][880/1519] eta 0:10:40 lr 0.000033 time 0.9305 (1.0030) model_time 0.9303 (1.0022) loss 1.1763 (1.3191) grad_norm 9.3154 (10.2908/3.8338) mem 68106MB [2022-12-19 06:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][890/1519] eta 0:10:30 lr 0.000033 time 0.9338 (1.0030) model_time 0.9337 (1.0022) loss 1.1304 (1.3194) grad_norm 7.6254 (10.2805/3.8291) mem 68106MB [2022-12-19 06:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][900/1519] eta 0:10:20 lr 0.000033 time 0.9437 (1.0030) model_time 0.9436 (1.0022) loss 1.6056 (1.3190) grad_norm 8.5423 (10.2601/3.8253) mem 68106MB [2022-12-19 06:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][910/1519] eta 0:10:10 lr 0.000033 time 0.9360 (1.0029) model_time 0.9359 (1.0021) loss 1.2170 (1.3187) grad_norm 8.2098 (10.1708/3.7891) mem 68106MB [2022-12-19 06:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][920/1519] eta 0:10:00 lr 0.000033 time 0.9333 (1.0029) model_time 0.9332 (1.0021) loss 1.2068 (1.3171) grad_norm 7.1000 (10.2166/3.8948) mem 68106MB [2022-12-19 06:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][930/1519] eta 0:09:50 lr 0.000033 time 0.9335 (1.0029) model_time 0.9334 (1.0021) loss 1.5321 (1.3166) grad_norm 9.7169 (10.2392/3.9011) mem 68106MB [2022-12-19 06:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][940/1519] eta 0:09:40 lr 0.000033 time 0.9305 (1.0028) model_time 0.9303 (1.0020) loss 1.3652 (1.3175) grad_norm 7.4028 (10.2811/3.9602) mem 68106MB [2022-12-19 06:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][950/1519] eta 0:09:30 lr 0.000033 time 0.9390 (1.0029) model_time 0.9388 (1.0021) loss 1.0535 (1.3159) grad_norm 9.1025 (10.2853/3.9708) mem 68106MB [2022-12-19 06:35:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][960/1519] eta 0:09:20 lr 0.000033 time 0.9340 (1.0028) model_time 0.9339 (1.0020) loss 1.1584 (1.3162) grad_norm 12.3939 (10.2425/3.9174) mem 68106MB [2022-12-19 06:35:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][970/1519] eta 0:09:10 lr 0.000033 time 0.9327 (1.0029) model_time 0.9326 (1.0021) loss 1.6021 (1.3164) grad_norm 8.9359 (10.2406/3.9502) mem 68106MB [2022-12-19 06:35:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][980/1519] eta 0:09:00 lr 0.000033 time 0.9305 (1.0029) model_time 0.9304 (1.0021) loss 1.0143 (1.3159) grad_norm 11.7095 (10.2161/3.8768) mem 68106MB [2022-12-19 06:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][990/1519] eta 0:08:50 lr 0.000033 time 0.9295 (1.0029) model_time 0.9294 (1.0021) loss 0.9982 (1.3170) grad_norm 20.9648 (10.3114/4.1087) mem 68106MB [2022-12-19 06:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1000/1519] eta 0:08:40 lr 0.000033 time 0.9346 (1.0029) model_time 0.9345 (1.0021) loss 1.2627 (1.3172) grad_norm 10.8977 (10.3749/4.1074) mem 68106MB [2022-12-19 06:35:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1010/1519] eta 0:08:30 lr 0.000033 time 0.9322 (1.0030) model_time 0.9321 (1.0022) loss 1.4676 (1.3175) grad_norm 8.4083 (10.3690/4.1095) mem 68106MB [2022-12-19 06:36:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1020/1519] eta 0:08:20 lr 0.000033 time 0.9363 (1.0030) model_time 0.9362 (1.0022) loss 1.1795 (1.3169) grad_norm 6.3151 (10.3690/4.1258) mem 68106MB [2022-12-19 06:36:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1030/1519] eta 0:08:10 lr 0.000033 time 0.9367 (1.0030) model_time 0.9366 (1.0022) loss 1.1586 (1.3162) grad_norm 10.6590 (10.3867/4.1198) mem 68106MB [2022-12-19 06:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1040/1519] eta 0:08:00 lr 0.000033 time 0.9319 (1.0029) model_time 0.9318 (1.0022) loss 1.3748 (1.3155) grad_norm 15.4094 (10.3970/4.1051) mem 68106MB [2022-12-19 06:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1050/1519] eta 0:07:50 lr 0.000033 time 0.9345 (1.0029) model_time 0.9343 (1.0021) loss 2.0357 (1.3148) grad_norm 11.5970 (10.4624/4.0795) mem 68106MB [2022-12-19 06:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1060/1519] eta 0:07:40 lr 0.000033 time 0.9296 (1.0029) model_time 0.9295 (1.0021) loss 1.3011 (1.3157) grad_norm 7.8094 (10.4843/4.0892) mem 68106MB [2022-12-19 06:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1070/1519] eta 0:07:30 lr 0.000033 time 0.9321 (1.0028) model_time 0.9320 (1.0021) loss 1.3721 (1.3159) grad_norm 9.2452 (10.4127/4.0905) mem 68106MB [2022-12-19 06:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1080/1519] eta 0:07:20 lr 0.000033 time 0.9315 (1.0028) model_time 0.9313 (1.0020) loss 1.1909 (1.3166) grad_norm 10.5441 (10.4448/4.0826) mem 68106MB [2022-12-19 06:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1090/1519] eta 0:07:10 lr 0.000033 time 0.9335 (1.0028) model_time 0.9334 (1.0021) loss 1.4865 (1.3171) grad_norm 9.8039 (10.3929/4.0462) mem 68106MB [2022-12-19 06:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1100/1519] eta 0:07:00 lr 0.000033 time 0.9313 (1.0028) model_time 0.9311 (1.0020) loss 1.2715 (1.3168) grad_norm 11.5911 (10.3949/4.0339) mem 68106MB [2022-12-19 06:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1110/1519] eta 0:06:50 lr 0.000033 time 0.9342 (1.0028) model_time 0.9340 (1.0020) loss 1.1948 (1.3158) grad_norm 18.2718 (10.4450/4.0456) mem 68106MB [2022-12-19 06:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1120/1519] eta 0:06:40 lr 0.000033 time 0.9314 (1.0027) model_time 0.9313 (1.0020) loss 0.9643 (1.3147) grad_norm 11.7024 (10.4196/4.0551) mem 68106MB [2022-12-19 06:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1130/1519] eta 0:06:30 lr 0.000033 time 0.9403 (1.0027) model_time 0.9401 (1.0020) loss 1.0221 (1.3138) grad_norm 15.2398 (10.4068/4.0740) mem 68106MB [2022-12-19 06:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1140/1519] eta 0:06:20 lr 0.000033 time 0.9454 (1.0027) model_time 0.9453 (1.0020) loss 1.0673 (1.3129) grad_norm 13.9244 (10.3659/4.0471) mem 68106MB [2022-12-19 06:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1150/1519] eta 0:06:09 lr 0.000033 time 0.9388 (1.0027) model_time 0.9387 (1.0020) loss 1.5551 (1.3119) grad_norm 16.9798 (10.3701/4.0598) mem 68106MB [2022-12-19 06:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1160/1519] eta 0:05:59 lr 0.000033 time 0.9328 (1.0027) model_time 0.9326 (1.0020) loss 1.1297 (1.3105) grad_norm 8.9031 (10.3975/4.1169) mem 68106MB [2022-12-19 06:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1170/1519] eta 0:05:49 lr 0.000033 time 0.9310 (1.0027) model_time 0.9308 (1.0020) loss 1.2683 (1.3096) grad_norm 6.1072 (10.3766/4.1121) mem 68106MB [2022-12-19 06:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1180/1519] eta 0:05:39 lr 0.000033 time 0.9397 (1.0027) model_time 0.9396 (1.0020) loss 0.9554 (1.3091) grad_norm 6.8753 (10.3375/4.1231) mem 68106MB [2022-12-19 06:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1190/1519] eta 0:05:29 lr 0.000033 time 0.9422 (1.0027) model_time 0.9420 (1.0020) loss 1.3557 (1.3088) grad_norm 5.3943 (10.2631/4.0989) mem 68106MB [2022-12-19 06:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1200/1519] eta 0:05:19 lr 0.000033 time 0.9282 (1.0027) model_time 0.9280 (1.0020) loss 1.0510 (1.3088) grad_norm 9.0178 (10.1715/3.9083) mem 68106MB [2022-12-19 06:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1210/1519] eta 0:05:09 lr 0.000033 time 0.9341 (1.0027) model_time 0.9339 (1.0020) loss 1.1292 (1.3087) grad_norm 5.6548 (10.1811/3.9089) mem 68106MB [2022-12-19 06:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1220/1519] eta 0:04:59 lr 0.000033 time 0.9309 (1.0026) model_time 0.9307 (1.0019) loss 1.2680 (1.3095) grad_norm 11.3923 (10.2077/3.9155) mem 68106MB [2022-12-19 06:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1230/1519] eta 0:04:49 lr 0.000033 time 0.9348 (1.0026) model_time 0.9346 (1.0019) loss 1.3333 (1.3112) grad_norm 16.1887 (10.2307/3.9390) mem 68106MB [2022-12-19 06:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1240/1519] eta 0:04:39 lr 0.000033 time 0.9344 (1.0026) model_time 0.9342 (1.0019) loss 1.1507 (1.3105) grad_norm 15.7100 (10.2462/3.9467) mem 68106MB [2022-12-19 06:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1250/1519] eta 0:04:29 lr 0.000033 time 0.9377 (1.0026) model_time 0.9376 (1.0019) loss 1.8233 (1.3103) grad_norm 7.4447 (10.2232/3.9475) mem 68106MB [2022-12-19 06:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1260/1519] eta 0:04:19 lr 0.000033 time 0.9310 (1.0026) model_time 0.9308 (1.0019) loss 1.2313 (1.3090) grad_norm 13.7149 (10.1923/3.8418) mem 68106MB [2022-12-19 06:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9409 (1.0026) model_time 0.9408 (1.0019) loss 0.9307 (1.3085) grad_norm 9.1209 (10.1178/3.8432) mem 68106MB [2022-12-19 06:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9207 (1.0027) model_time 0.9205 (1.0020) loss 1.3029 (1.3082) grad_norm 11.8637 (10.1460/3.8471) mem 68106MB [2022-12-19 06:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9342 (1.0026) model_time 0.9341 (1.0020) loss 1.1376 (1.3080) grad_norm 7.2979 (10.1148/3.8465) mem 68106MB [2022-12-19 06:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9260 (1.0027) model_time 0.9258 (1.0020) loss 1.2117 (1.3074) grad_norm 8.9817 (10.0467/3.8315) mem 68106MB [2022-12-19 06:40:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9312 (1.0027) model_time 0.9310 (1.0020) loss 1.1564 (1.3072) grad_norm 9.6438 (10.0452/3.8413) mem 68106MB [2022-12-19 06:41:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9334 (1.0027) model_time 0.9333 (1.0020) loss 1.7010 (1.3077) grad_norm 10.5321 (9.9879/3.7676) mem 68106MB [2022-12-19 06:41:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9346 (1.0028) model_time 0.9344 (1.0022) loss 0.8563 (1.3075) grad_norm 23.0579 (10.0100/3.8395) mem 68106MB [2022-12-19 06:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9376 (1.0028) model_time 0.9374 (1.0021) loss 1.0340 (1.3077) grad_norm 8.6705 (10.0142/3.8605) mem 68106MB [2022-12-19 06:41:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9373 (1.0028) model_time 0.9372 (1.0021) loss 1.3128 (1.3071) grad_norm 10.0588 (10.0113/3.8420) mem 68106MB [2022-12-19 06:41:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9346 (1.0028) model_time 0.9344 (1.0021) loss 1.0434 (1.3061) grad_norm 8.7338 (9.9680/3.8375) mem 68106MB [2022-12-19 06:41:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9411 (1.0028) model_time 0.9410 (1.0021) loss 1.1049 (1.3051) grad_norm 33.6210 (10.0452/4.0927) mem 68106MB [2022-12-19 06:42:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9374 (1.0028) model_time 0.9372 (1.0021) loss 1.1079 (1.3051) grad_norm 11.7454 (10.0690/4.0813) mem 68106MB [2022-12-19 06:42:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9348 (1.0027) model_time 0.9346 (1.0021) loss 1.3218 (1.3053) grad_norm 22.6735 (10.0988/4.1312) mem 68106MB [2022-12-19 06:42:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9312 (1.0027) model_time 0.9311 (1.0021) loss 1.5402 (1.3057) grad_norm 10.4646 (10.0630/4.1185) mem 68106MB [2022-12-19 06:42:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9307 (1.0027) model_time 0.9306 (1.0020) loss 1.6928 (1.3059) grad_norm 6.9659 (10.1035/4.1162) mem 68106MB [2022-12-19 06:42:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9316 (1.0027) model_time 0.9314 (1.0020) loss 0.9287 (1.3061) grad_norm 11.1587 (10.1016/4.1348) mem 68106MB [2022-12-19 06:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9284 (1.0026) model_time 0.9283 (1.0020) loss 1.3858 (1.3057) grad_norm 12.6639 (10.1492/4.1295) mem 68106MB [2022-12-19 06:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9279 (1.0026) model_time 0.9277 (1.0020) loss 1.1071 (1.3064) grad_norm 14.2785 (10.1511/4.0874) mem 68106MB [2022-12-19 06:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9326 (1.0026) model_time 0.9324 (1.0020) loss 1.7842 (1.3064) grad_norm 17.4033 (10.1910/4.1410) mem 68106MB [2022-12-19 06:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9472 (1.0026) model_time 0.9471 (1.0019) loss 1.4202 (1.3067) grad_norm 9.2298 (10.2204/4.1486) mem 68106MB [2022-12-19 06:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9380 (1.0026) model_time 0.9379 (1.0019) loss 1.2402 (1.3067) grad_norm 10.7321 (10.2608/4.1283) mem 68106MB [2022-12-19 06:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9328 (1.0025) model_time 0.9326 (1.0019) loss 1.2580 (1.3061) grad_norm 11.0101 (10.2605/4.1277) mem 68106MB [2022-12-19 06:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9350 (1.0026) model_time 0.9349 (1.0019) loss 1.2248 (1.3061) grad_norm 7.4195 (10.3239/4.1473) mem 68106MB [2022-12-19 06:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9286 (1.0026) model_time 0.9285 (1.0019) loss 1.1700 (1.3063) grad_norm 10.2679 (10.2959/4.1322) mem 68106MB [2022-12-19 06:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [5/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9274 (1.0025) model_time 0.9273 (1.0019) loss 1.7359 (1.3060) grad_norm 10.9136 (10.3117/4.1189) mem 68106MB [2022-12-19 06:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 5 training takes 0:25:22 [2022-12-19 06:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_5.pth saving...... [2022-12-19 06:44:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_5.pth saved !!! [2022-12-19 06:44:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.611 (0.611) Loss 1.6027 (1.6027) Acc@1 68.403 (68.403) Acc@5 90.972 (90.972) Mem 68106MB [2022-12-19 06:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.326) Loss 1.7380 (1.6754) Acc@1 67.708 (65.972) Acc@5 89.931 (90.025) Mem 68106MB [2022-12-19 06:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.312) Loss 1.5948 (1.6701) Acc@1 66.319 (65.261) Acc@5 92.708 (90.146) Mem 68106MB [2022-12-19 06:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.300 (0.308) Loss 1.7315 (1.6723) Acc@1 63.194 (65.031) Acc@5 90.278 (90.121) Mem 68106MB [2022-12-19 06:45:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.306) Loss 1.5966 (1.6552) Acc@1 68.403 (65.354) Acc@5 89.236 (90.354) Mem 68106MB [2022-12-19 06:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.304) Loss 1.7041 (1.6496) Acc@1 64.236 (65.598) Acc@5 89.583 (90.441) Mem 68106MB [2022-12-19 06:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 1.7438 (1.6533) Acc@1 64.931 (65.505) Acc@5 89.583 (90.454) Mem 68106MB [2022-12-19 06:45:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 1.8243 (1.6595) Acc@1 64.583 (65.561) Acc@5 87.847 (90.390) Mem 68106MB [2022-12-19 06:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.299 (0.302) Loss 1.6302 (1.6604) Acc@1 65.625 (65.582) Acc@5 92.708 (90.346) Mem 68106MB [2022-12-19 06:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:5] * Acc@1 65.537 Acc@5 90.336 [2022-12-19 06:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 65.5% [2022-12-19 06:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 06:45:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 06:45:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 65.54% [2022-12-19 06:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][0/1519] eta 0:36:31 lr 0.000033 time 1.4426 (1.4426) model_time 0.9767 (0.9767) loss 1.2230 (1.2230) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 06:45:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][10/1519] eta 0:26:19 lr 0.000033 time 0.9378 (1.0465) model_time 0.9377 (1.0037) loss 1.4763 (1.2588) grad_norm 5.9906 (7.2960/0.7523) mem 68106MB [2022-12-19 06:46:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][20/1519] eta 0:25:38 lr 0.000033 time 0.9309 (1.0264) model_time 0.9308 (1.0038) loss 1.4713 (1.2070) grad_norm 10.0638 (9.3977/3.5129) mem 68106MB [2022-12-19 06:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][30/1519] eta 0:25:15 lr 0.000033 time 0.9265 (1.0175) model_time 0.9264 (1.0022) loss 1.1479 (1.2331) grad_norm 9.2722 (9.5375/3.1305) mem 68106MB [2022-12-19 06:46:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][40/1519] eta 0:24:58 lr 0.000033 time 0.9276 (1.0131) model_time 0.9275 (1.0013) loss 1.0916 (1.2156) grad_norm 7.9251 (9.6982/3.3712) mem 68106MB [2022-12-19 06:46:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][50/1519] eta 0:24:44 lr 0.000033 time 0.9354 (1.0105) model_time 0.9352 (1.0010) loss 1.2070 (1.2152) grad_norm 9.2239 (9.6898/3.2202) mem 68106MB [2022-12-19 06:46:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][60/1519] eta 0:24:31 lr 0.000033 time 0.9306 (1.0085) model_time 0.9304 (1.0005) loss 1.1615 (1.2176) grad_norm 5.7661 (9.5922/3.3504) mem 68106MB [2022-12-19 06:46:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][70/1519] eta 0:24:24 lr 0.000033 time 0.9278 (1.0105) model_time 0.9277 (1.0036) loss 1.6009 (1.2224) grad_norm 8.0743 (9.7972/3.9724) mem 68106MB [2022-12-19 06:47:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][80/1519] eta 0:24:13 lr 0.000033 time 0.9372 (1.0100) model_time 0.9370 (1.0039) loss 1.0427 (1.2359) grad_norm 12.1400 (9.8428/3.8206) mem 68106MB [2022-12-19 06:47:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][90/1519] eta 0:24:02 lr 0.000033 time 0.9270 (1.0095) model_time 0.9269 (1.0040) loss 1.5676 (1.2310) grad_norm 8.3973 (9.6403/3.7324) mem 68106MB [2022-12-19 06:47:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][100/1519] eta 0:23:52 lr 0.000033 time 0.9292 (1.0094) model_time 0.9291 (1.0044) loss 1.1169 (1.2325) grad_norm 16.6371 (9.7964/3.7910) mem 68106MB [2022-12-19 06:47:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][110/1519] eta 0:23:41 lr 0.000033 time 0.9322 (1.0085) model_time 0.9321 (1.0040) loss 1.5595 (1.2354) grad_norm 24.8901 (9.9893/4.1903) mem 68106MB [2022-12-19 06:47:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][120/1519] eta 0:23:32 lr 0.000033 time 0.9345 (1.0099) model_time 0.9343 (1.0057) loss 0.9938 (1.2316) grad_norm 9.5749 (10.0256/4.1381) mem 68106MB [2022-12-19 06:47:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][130/1519] eta 0:23:21 lr 0.000033 time 0.9209 (1.0093) model_time 0.9207 (1.0054) loss 0.9407 (1.2288) grad_norm 7.0190 (9.9012/4.0268) mem 68106MB [2022-12-19 06:48:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][140/1519] eta 0:23:10 lr 0.000033 time 0.9350 (1.0086) model_time 0.9346 (1.0050) loss 1.0478 (1.2346) grad_norm 13.7996 (10.0378/3.9540) mem 68106MB [2022-12-19 06:48:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][150/1519] eta 0:23:00 lr 0.000033 time 0.9263 (1.0082) model_time 0.9261 (1.0048) loss 1.2910 (1.2421) grad_norm 8.6911 (10.1796/3.9493) mem 68106MB [2022-12-19 06:48:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][160/1519] eta 0:22:49 lr 0.000033 time 0.9268 (1.0077) model_time 0.9266 (1.0044) loss 1.6554 (1.2353) grad_norm 13.6462 (10.1015/3.8975) mem 68106MB [2022-12-19 06:48:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][170/1519] eta 0:22:38 lr 0.000033 time 0.9282 (1.0073) model_time 0.9281 (1.0042) loss 1.4351 (1.2409) grad_norm 8.2914 (10.0865/3.8653) mem 68106MB [2022-12-19 06:48:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][180/1519] eta 0:22:28 lr 0.000033 time 0.9323 (1.0070) model_time 0.9321 (1.0040) loss 1.1931 (1.2410) grad_norm 6.7829 (10.2206/4.1770) mem 68106MB [2022-12-19 06:48:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][190/1519] eta 0:22:17 lr 0.000033 time 0.9325 (1.0066) model_time 0.9323 (1.0038) loss 1.2469 (1.2404) grad_norm 8.2416 (10.0979/4.1218) mem 68106MB [2022-12-19 06:49:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][200/1519] eta 0:22:07 lr 0.000033 time 0.9362 (1.0064) model_time 0.9360 (1.0037) loss 1.2529 (1.2429) grad_norm 8.7970 (10.1164/4.1099) mem 68106MB [2022-12-19 06:49:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][210/1519] eta 0:21:56 lr 0.000033 time 0.9326 (1.0061) model_time 0.9324 (1.0035) loss 1.0110 (1.2490) grad_norm 14.1931 (10.1430/4.0370) mem 68106MB [2022-12-19 06:49:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][220/1519] eta 0:21:46 lr 0.000033 time 0.9351 (1.0058) model_time 0.9349 (1.0033) loss 1.1619 (1.2490) grad_norm 6.5683 (10.0777/3.9789) mem 68106MB [2022-12-19 06:49:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][230/1519] eta 0:21:36 lr 0.000033 time 0.9255 (1.0055) model_time 0.9254 (1.0031) loss 1.0846 (1.2484) grad_norm 11.6337 (10.0012/3.9306) mem 68106MB [2022-12-19 06:49:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][240/1519] eta 0:21:25 lr 0.000033 time 0.9361 (1.0054) model_time 0.9359 (1.0031) loss 1.0103 (1.2461) grad_norm 10.2335 (10.0725/3.8997) mem 68106MB [2022-12-19 06:49:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][250/1519] eta 0:21:15 lr 0.000033 time 0.9241 (1.0052) model_time 0.9239 (1.0030) loss 1.1642 (1.2451) grad_norm 13.3834 (10.0762/3.8539) mem 68106MB [2022-12-19 06:50:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][260/1519] eta 0:21:05 lr 0.000033 time 0.9371 (1.0054) model_time 0.9370 (1.0032) loss 1.6599 (1.2432) grad_norm 8.4395 (10.0171/3.8079) mem 68106MB [2022-12-19 06:50:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][270/1519] eta 0:20:55 lr 0.000033 time 0.9329 (1.0054) model_time 0.9327 (1.0034) loss 1.1926 (1.2425) grad_norm 12.2256 (9.9852/3.7578) mem 68106MB [2022-12-19 06:50:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][280/1519] eta 0:20:45 lr 0.000033 time 0.9408 (1.0054) model_time 0.9406 (1.0034) loss 1.1721 (1.2392) grad_norm 9.7204 (9.9204/3.7087) mem 68106MB [2022-12-19 06:50:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][290/1519] eta 0:20:36 lr 0.000033 time 0.9382 (1.0059) model_time 0.9380 (1.0039) loss 1.4458 (1.2399) grad_norm 9.1130 (9.8737/3.6550) mem 68106MB [2022-12-19 06:50:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][300/1519] eta 0:20:26 lr 0.000033 time 0.9246 (1.0062) model_time 0.9244 (1.0043) loss 1.1159 (1.2401) grad_norm 11.3274 (9.9262/3.6101) mem 68106MB [2022-12-19 06:50:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][310/1519] eta 0:20:16 lr 0.000033 time 0.9251 (1.0062) model_time 0.9249 (1.0043) loss 1.5279 (1.2413) grad_norm 8.0574 (9.8715/3.5683) mem 68106MB [2022-12-19 06:51:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][320/1519] eta 0:20:06 lr 0.000033 time 0.9445 (1.0061) model_time 0.9443 (1.0043) loss 1.0478 (1.2395) grad_norm 6.8992 (9.8009/3.5410) mem 68106MB [2022-12-19 06:51:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][330/1519] eta 0:19:56 lr 0.000033 time 0.9278 (1.0060) model_time 0.9277 (1.0042) loss 1.0271 (1.2378) grad_norm 9.3126 (9.7857/3.4938) mem 68106MB [2022-12-19 06:51:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][340/1519] eta 0:19:45 lr 0.000033 time 0.9238 (1.0058) model_time 0.9236 (1.0041) loss 1.4065 (1.2375) grad_norm 10.2297 (9.9114/3.6676) mem 68106MB [2022-12-19 06:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][350/1519] eta 0:19:35 lr 0.000033 time 0.9267 (1.0056) model_time 0.9265 (1.0040) loss 1.0158 (1.2396) grad_norm 13.7270 (10.0429/3.7349) mem 68106MB [2022-12-19 06:51:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][360/1519] eta 0:19:25 lr 0.000033 time 0.9484 (1.0056) model_time 0.9482 (1.0040) loss 1.1034 (1.2440) grad_norm 7.8131 (10.0488/3.7331) mem 68106MB [2022-12-19 06:51:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][370/1519] eta 0:19:15 lr 0.000033 time 0.9390 (1.0055) model_time 0.9388 (1.0039) loss 1.5983 (1.2425) grad_norm 13.0630 (10.0346/3.7037) mem 68106MB [2022-12-19 06:52:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][380/1519] eta 0:19:05 lr 0.000033 time 0.9132 (1.0056) model_time 0.9131 (1.0040) loss 1.0857 (1.2391) grad_norm 11.0192 (10.0280/3.6754) mem 68106MB [2022-12-19 06:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][390/1519] eta 0:18:55 lr 0.000033 time 0.9245 (1.0054) model_time 0.9243 (1.0039) loss 1.3776 (1.2382) grad_norm 8.8224 (10.0118/3.6376) mem 68106MB [2022-12-19 06:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][400/1519] eta 0:18:45 lr 0.000033 time 0.9275 (1.0054) model_time 0.9274 (1.0039) loss 0.9776 (1.2370) grad_norm 6.2549 (9.9648/3.6177) mem 68106MB [2022-12-19 06:52:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][410/1519] eta 0:18:35 lr 0.000033 time 0.9313 (1.0059) model_time 0.9312 (1.0044) loss 1.2246 (1.2336) grad_norm 10.9736 (10.1667/3.8570) mem 68106MB [2022-12-19 06:52:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][420/1519] eta 0:18:25 lr 0.000033 time 0.9244 (1.0057) model_time 0.9243 (1.0043) loss 1.8927 (1.2351) grad_norm 14.6142 (10.1960/3.8407) mem 68106MB [2022-12-19 06:52:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][430/1519] eta 0:18:15 lr 0.000033 time 0.9315 (1.0057) model_time 0.9313 (1.0042) loss 1.5330 (1.2373) grad_norm 11.4843 (10.1729/3.8103) mem 68106MB [2022-12-19 06:53:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][440/1519] eta 0:18:04 lr 0.000033 time 0.9273 (1.0055) model_time 0.9271 (1.0041) loss 0.7263 (1.2369) grad_norm 9.2658 (10.2045/3.7985) mem 68106MB [2022-12-19 06:53:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][450/1519] eta 0:17:54 lr 0.000033 time 0.9264 (1.0054) model_time 0.9263 (1.0040) loss 1.0681 (1.2348) grad_norm 12.3620 (10.2691/3.7870) mem 68106MB [2022-12-19 06:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][460/1519] eta 0:17:44 lr 0.000033 time 0.9217 (1.0054) model_time 0.9215 (1.0040) loss 1.0599 (1.2373) grad_norm 9.8318 (10.2370/3.7574) mem 68106MB [2022-12-19 06:53:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][470/1519] eta 0:17:34 lr 0.000033 time 0.9175 (1.0054) model_time 0.9173 (1.0041) loss 1.2007 (1.2363) grad_norm 10.0908 (10.2415/3.7364) mem 68106MB [2022-12-19 06:53:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][480/1519] eta 0:17:24 lr 0.000033 time 0.9244 (1.0053) model_time 0.9242 (1.0040) loss 1.3329 (1.2348) grad_norm 13.9099 (10.2255/3.7148) mem 68106MB [2022-12-19 06:53:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][490/1519] eta 0:17:14 lr 0.000033 time 0.9181 (1.0052) model_time 0.9180 (1.0039) loss 1.3342 (1.2356) grad_norm 5.2077 (10.1970/3.6923) mem 68106MB [2022-12-19 06:54:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][500/1519] eta 0:17:04 lr 0.000033 time 0.9299 (1.0052) model_time 0.9297 (1.0039) loss 1.3749 (1.2355) grad_norm 7.6079 (10.2534/3.7128) mem 68106MB [2022-12-19 06:54:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][510/1519] eta 0:16:54 lr 0.000033 time 0.9224 (1.0051) model_time 0.9223 (1.0038) loss 1.2161 (1.2349) grad_norm 8.8946 (10.2157/3.6876) mem 68106MB [2022-12-19 06:54:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][520/1519] eta 0:16:44 lr 0.000033 time 0.9274 (1.0050) model_time 0.9272 (1.0038) loss 1.1687 (1.2328) grad_norm 6.2589 (10.1932/3.6691) mem 68106MB [2022-12-19 06:54:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][530/1519] eta 0:16:33 lr 0.000033 time 0.9285 (1.0050) model_time 0.9284 (1.0037) loss 1.2622 (1.2318) grad_norm 8.0676 (10.2174/3.6565) mem 68106MB [2022-12-19 06:54:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][540/1519] eta 0:16:23 lr 0.000033 time 0.9317 (1.0049) model_time 0.9317 (1.0037) loss 1.0214 (1.2317) grad_norm 7.7026 (10.1790/3.6370) mem 68106MB [2022-12-19 06:54:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][550/1519] eta 0:16:13 lr 0.000033 time 0.9509 (1.0048) model_time 0.9507 (1.0036) loss 1.0140 (1.2308) grad_norm 23.5963 (10.2304/3.6955) mem 68106MB [2022-12-19 06:55:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][560/1519] eta 0:16:03 lr 0.000033 time 0.9202 (1.0047) model_time 0.9201 (1.0036) loss 1.6838 (1.2336) grad_norm 12.3208 (10.2363/3.6760) mem 68106MB [2022-12-19 06:55:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][570/1519] eta 0:15:53 lr 0.000033 time 0.9230 (1.0048) model_time 0.9229 (1.0036) loss 1.0764 (1.2352) grad_norm 13.2479 (10.2363/3.6514) mem 68106MB [2022-12-19 06:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][580/1519] eta 0:15:43 lr 0.000033 time 0.9403 (1.0048) model_time 0.9401 (1.0037) loss 0.9846 (1.2333) grad_norm 8.9255 (10.2121/3.6280) mem 68106MB [2022-12-19 06:55:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][590/1519] eta 0:15:33 lr 0.000033 time 0.9267 (1.0049) model_time 0.9264 (1.0037) loss 1.4664 (1.2328) grad_norm 12.1596 (10.2264/3.6051) mem 68106MB [2022-12-19 06:55:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][600/1519] eta 0:15:23 lr 0.000033 time 0.9222 (1.0053) model_time 0.9221 (1.0041) loss 1.2585 (1.2312) grad_norm 6.4724 (10.2203/3.6132) mem 68106MB [2022-12-19 06:55:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][610/1519] eta 0:15:14 lr 0.000033 time 0.9283 (1.0056) model_time 0.9282 (1.0045) loss 1.6785 (1.2320) grad_norm 7.0328 (10.2188/3.6221) mem 68106MB [2022-12-19 06:56:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][620/1519] eta 0:15:03 lr 0.000033 time 0.9222 (1.0055) model_time 0.9220 (1.0044) loss 1.1982 (1.2315) grad_norm 7.7819 (10.2242/3.6936) mem 68106MB [2022-12-19 06:56:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][630/1519] eta 0:14:53 lr 0.000033 time 0.9237 (1.0054) model_time 0.9234 (1.0043) loss 1.2900 (1.2320) grad_norm 11.4135 (10.2630/3.7428) mem 68106MB [2022-12-19 06:56:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][640/1519] eta 0:14:43 lr 0.000033 time 0.9317 (1.0054) model_time 0.9314 (1.0043) loss 1.1864 (1.2317) grad_norm 15.6120 (10.3060/3.7657) mem 68106MB [2022-12-19 06:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][650/1519] eta 0:14:33 lr 0.000033 time 0.9279 (1.0053) model_time 0.9276 (1.0042) loss 1.5155 (1.2313) grad_norm 11.9440 (10.3223/3.7955) mem 68106MB [2022-12-19 06:56:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][660/1519] eta 0:14:23 lr 0.000033 time 0.9346 (1.0052) model_time 0.9344 (1.0041) loss 1.5078 (1.2323) grad_norm 7.9643 (10.3708/3.7897) mem 68106MB [2022-12-19 06:56:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][670/1519] eta 0:14:13 lr 0.000033 time 0.9348 (1.0051) model_time 0.9346 (1.0040) loss 0.9864 (1.2318) grad_norm 6.6573 (10.3436/3.7083) mem 68106MB [2022-12-19 06:57:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][680/1519] eta 0:14:03 lr 0.000033 time 0.9303 (1.0050) model_time 0.9302 (1.0040) loss 1.0707 (1.2318) grad_norm 8.2977 (10.3304/3.7185) mem 68106MB [2022-12-19 06:57:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][690/1519] eta 0:13:53 lr 0.000033 time 0.9277 (1.0050) model_time 0.9275 (1.0039) loss 1.0219 (1.2299) grad_norm 7.4197 (10.3459/3.7041) mem 68106MB [2022-12-19 06:57:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][700/1519] eta 0:13:43 lr 0.000033 time 0.9277 (1.0049) model_time 0.9276 (1.0039) loss 1.3405 (1.2301) grad_norm 8.3246 (10.3122/3.6914) mem 68106MB [2022-12-19 06:57:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][710/1519] eta 0:13:32 lr 0.000033 time 0.9254 (1.0048) model_time 0.9253 (1.0038) loss 1.6553 (1.2318) grad_norm 10.8124 (10.2555/3.5962) mem 68106MB [2022-12-19 06:57:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][720/1519] eta 0:13:22 lr 0.000033 time 0.9225 (1.0047) model_time 0.9223 (1.0037) loss 1.6922 (1.2318) grad_norm 6.7828 (10.2914/3.7024) mem 68106MB [2022-12-19 06:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][730/1519] eta 0:13:12 lr 0.000033 time 0.9248 (1.0047) model_time 0.9247 (1.0037) loss 1.4345 (1.2337) grad_norm 7.3640 (10.2861/3.7001) mem 68106MB [2022-12-19 06:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][740/1519] eta 0:13:02 lr 0.000033 time 0.9256 (1.0046) model_time 0.9255 (1.0036) loss 1.2863 (1.2335) grad_norm 7.2385 (10.2104/3.7061) mem 68106MB [2022-12-19 06:58:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][750/1519] eta 0:12:52 lr 0.000033 time 0.9179 (1.0046) model_time 0.9177 (1.0036) loss 1.6037 (1.2335) grad_norm 8.9584 (10.2124/3.7153) mem 68106MB [2022-12-19 06:58:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][760/1519] eta 0:12:42 lr 0.000033 time 0.9283 (1.0046) model_time 0.9282 (1.0036) loss 0.9476 (1.2342) grad_norm 17.7704 (10.3200/3.8229) mem 68106MB [2022-12-19 06:58:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][770/1519] eta 0:12:32 lr 0.000033 time 0.9194 (1.0045) model_time 0.9193 (1.0035) loss 1.4984 (1.2337) grad_norm 7.9419 (10.3922/3.8999) mem 68106MB [2022-12-19 06:58:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][780/1519] eta 0:12:22 lr 0.000033 time 0.9297 (1.0046) model_time 0.9295 (1.0036) loss 1.3409 (1.2333) grad_norm 7.9705 (10.4376/4.0106) mem 68106MB [2022-12-19 06:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][790/1519] eta 0:12:12 lr 0.000033 time 0.9314 (1.0045) model_time 0.9313 (1.0035) loss 0.9684 (1.2318) grad_norm 8.7844 (10.4777/4.0035) mem 68106MB [2022-12-19 06:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][800/1519] eta 0:12:02 lr 0.000033 time 0.9255 (1.0044) model_time 0.9254 (1.0035) loss 1.0918 (1.2329) grad_norm 19.9669 (10.5364/4.0220) mem 68106MB [2022-12-19 06:59:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][810/1519] eta 0:11:52 lr 0.000033 time 0.9307 (1.0043) model_time 0.9305 (1.0034) loss 0.9747 (1.2330) grad_norm 12.1683 (10.5627/4.0438) mem 68106MB [2022-12-19 06:59:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][820/1519] eta 0:11:41 lr 0.000033 time 0.9259 (1.0043) model_time 0.9257 (1.0033) loss 1.0490 (1.2310) grad_norm 14.1755 (10.6435/4.0629) mem 68106MB [2022-12-19 06:59:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][830/1519] eta 0:11:31 lr 0.000033 time 0.9298 (1.0043) model_time 0.9296 (1.0034) loss 1.2197 (1.2304) grad_norm 15.3177 (10.7388/4.0918) mem 68106MB [2022-12-19 06:59:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][840/1519] eta 0:11:21 lr 0.000033 time 0.9298 (1.0043) model_time 0.9296 (1.0034) loss 1.4309 (1.2309) grad_norm 14.7045 (10.7395/4.0847) mem 68106MB [2022-12-19 06:59:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][850/1519] eta 0:11:11 lr 0.000033 time 0.9396 (1.0042) model_time 0.9395 (1.0033) loss 1.1878 (1.2305) grad_norm 10.1075 (10.7514/4.0759) mem 68106MB [2022-12-19 07:00:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][860/1519] eta 0:11:01 lr 0.000033 time 0.9167 (1.0042) model_time 0.9165 (1.0033) loss 1.1114 (1.2303) grad_norm 11.2648 (10.7966/4.0853) mem 68106MB [2022-12-19 07:00:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][870/1519] eta 0:10:51 lr 0.000033 time 0.9264 (1.0042) model_time 0.9262 (1.0033) loss 1.4016 (1.2304) grad_norm 5.8688 (10.8025/4.0841) mem 68106MB [2022-12-19 07:00:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][880/1519] eta 0:10:41 lr 0.000033 time 0.9360 (1.0042) model_time 0.9359 (1.0033) loss 1.2613 (1.2295) grad_norm 13.6732 (10.8792/4.0893) mem 68106MB [2022-12-19 07:00:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][890/1519] eta 0:10:31 lr 0.000033 time 1.0065 (1.0042) model_time 1.0063 (1.0033) loss 1.3878 (1.2302) grad_norm 6.2807 (10.8768/4.0979) mem 68106MB [2022-12-19 07:00:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][900/1519] eta 0:10:21 lr 0.000033 time 0.8884 (1.0043) model_time 0.8883 (1.0034) loss 1.3526 (1.2291) grad_norm 12.7694 (10.8749/4.0995) mem 68106MB [2022-12-19 07:00:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][910/1519] eta 0:10:11 lr 0.000033 time 0.9271 (1.0042) model_time 0.9269 (1.0033) loss 1.3297 (1.2291) grad_norm 5.5041 (10.9228/4.2079) mem 68106MB [2022-12-19 07:01:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][920/1519] eta 0:10:01 lr 0.000033 time 1.2401 (1.0045) model_time 1.2398 (1.0037) loss 1.3419 (1.2286) grad_norm 7.2676 (10.9970/4.2154) mem 68106MB [2022-12-19 07:01:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][930/1519] eta 0:09:51 lr 0.000033 time 0.9195 (1.0045) model_time 0.9193 (1.0036) loss 0.9179 (1.2281) grad_norm 15.6688 (11.0062/4.2318) mem 68106MB [2022-12-19 07:01:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][940/1519] eta 0:09:41 lr 0.000033 time 0.9234 (1.0044) model_time 0.9233 (1.0036) loss 1.0214 (1.2298) grad_norm 10.1738 (10.9331/4.1502) mem 68106MB [2022-12-19 07:01:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][950/1519] eta 0:09:31 lr 0.000033 time 0.9224 (1.0044) model_time 0.9219 (1.0035) loss 1.3647 (1.2300) grad_norm 13.7288 (10.8824/4.1129) mem 68106MB [2022-12-19 07:01:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][960/1519] eta 0:09:21 lr 0.000033 time 0.9254 (1.0043) model_time 0.9252 (1.0035) loss 1.2137 (1.2298) grad_norm 11.2476 (10.8892/4.1026) mem 68106MB [2022-12-19 07:01:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][970/1519] eta 0:09:11 lr 0.000033 time 0.9273 (1.0043) model_time 0.9270 (1.0034) loss 1.5144 (1.2296) grad_norm 5.9540 (10.8778/4.1100) mem 68106MB [2022-12-19 07:02:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][980/1519] eta 0:09:01 lr 0.000033 time 0.9289 (1.0042) model_time 0.9286 (1.0034) loss 0.9975 (1.2284) grad_norm 11.3688 (10.9015/4.0978) mem 68106MB [2022-12-19 07:02:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][990/1519] eta 0:08:51 lr 0.000033 time 0.9207 (1.0042) model_time 0.9206 (1.0034) loss 1.4259 (1.2279) grad_norm 11.0549 (10.9474/4.1540) mem 68106MB [2022-12-19 07:02:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1000/1519] eta 0:08:41 lr 0.000033 time 0.9225 (1.0041) model_time 0.9223 (1.0033) loss 1.1614 (1.2276) grad_norm 9.9858 (10.9762/4.1401) mem 68106MB [2022-12-19 07:02:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1010/1519] eta 0:08:31 lr 0.000033 time 0.9263 (1.0042) model_time 0.9262 (1.0033) loss 1.4002 (1.2283) grad_norm 11.0347 (10.8704/4.0083) mem 68106MB [2022-12-19 07:02:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1020/1519] eta 0:08:21 lr 0.000033 time 0.9244 (1.0041) model_time 0.9242 (1.0033) loss 1.2839 (1.2284) grad_norm 6.2755 (10.8368/4.0044) mem 68106MB [2022-12-19 07:02:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1030/1519] eta 0:08:10 lr 0.000033 time 0.9371 (1.0041) model_time 0.9369 (1.0032) loss 0.9644 (1.2282) grad_norm 11.8346 (10.8416/4.0018) mem 68106MB [2022-12-19 07:03:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1040/1519] eta 0:08:00 lr 0.000033 time 0.9229 (1.0040) model_time 0.9227 (1.0032) loss 1.1497 (1.2261) grad_norm 6.3397 (10.8080/4.0175) mem 68106MB [2022-12-19 07:03:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1050/1519] eta 0:07:50 lr 0.000033 time 0.9274 (1.0040) model_time 0.9272 (1.0032) loss 1.1142 (1.2255) grad_norm 9.9908 (10.7573/4.0038) mem 68106MB [2022-12-19 07:03:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1060/1519] eta 0:07:40 lr 0.000033 time 0.9275 (1.0040) model_time 0.9273 (1.0032) loss 0.9850 (1.2245) grad_norm 8.1041 (10.7709/4.0060) mem 68106MB [2022-12-19 07:03:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1070/1519] eta 0:07:30 lr 0.000033 time 0.9223 (1.0040) model_time 0.9222 (1.0032) loss 1.1601 (1.2241) grad_norm 7.2876 (10.7246/4.0195) mem 68106MB [2022-12-19 07:03:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1080/1519] eta 0:07:20 lr 0.000033 time 0.9158 (1.0039) model_time 0.9156 (1.0031) loss 1.6160 (1.2249) grad_norm 7.6777 (10.7227/4.0144) mem 68106MB [2022-12-19 07:03:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1090/1519] eta 0:07:10 lr 0.000033 time 0.9294 (1.0041) model_time 0.9292 (1.0033) loss 1.3631 (1.2249) grad_norm 10.2953 (10.7722/4.0571) mem 68106MB [2022-12-19 07:04:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1100/1519] eta 0:07:00 lr 0.000033 time 0.9213 (1.0040) model_time 0.9212 (1.0032) loss 1.2356 (1.2251) grad_norm 14.4459 (10.7280/4.0392) mem 68106MB [2022-12-19 07:04:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1110/1519] eta 0:06:50 lr 0.000033 time 0.9240 (1.0040) model_time 0.9239 (1.0032) loss 1.1301 (1.2249) grad_norm 12.5570 (10.7720/4.0462) mem 68106MB [2022-12-19 07:04:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1120/1519] eta 0:06:40 lr 0.000033 time 0.9319 (1.0040) model_time 0.9316 (1.0032) loss 1.3319 (1.2250) grad_norm 6.5940 (10.8566/4.1997) mem 68106MB [2022-12-19 07:04:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1130/1519] eta 0:06:30 lr 0.000033 time 0.9275 (1.0039) model_time 0.9273 (1.0032) loss 1.3218 (1.2260) grad_norm 20.1069 (10.8394/4.2354) mem 68106MB [2022-12-19 07:04:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1140/1519] eta 0:06:20 lr 0.000033 time 0.9373 (1.0040) model_time 0.9372 (1.0032) loss 1.0609 (1.2263) grad_norm 5.9040 (10.8516/4.2370) mem 68106MB [2022-12-19 07:04:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1150/1519] eta 0:06:10 lr 0.000033 time 0.9284 (1.0039) model_time 0.9282 (1.0031) loss 1.1800 (1.2270) grad_norm 10.3980 (10.8097/4.1929) mem 68106MB [2022-12-19 07:05:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1160/1519] eta 0:06:00 lr 0.000033 time 0.9266 (1.0039) model_time 0.9265 (1.0031) loss 1.1636 (1.2273) grad_norm 11.1846 (10.8411/4.2175) mem 68106MB [2022-12-19 07:05:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1170/1519] eta 0:05:50 lr 0.000033 time 0.9309 (1.0038) model_time 0.9306 (1.0031) loss 1.0744 (1.2271) grad_norm 9.2045 (10.7953/4.2347) mem 68106MB [2022-12-19 07:05:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1180/1519] eta 0:05:40 lr 0.000033 time 0.9216 (1.0038) model_time 0.9215 (1.0030) loss 0.8863 (1.2274) grad_norm 10.4332 (10.8242/4.2273) mem 68106MB [2022-12-19 07:05:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1190/1519] eta 0:05:30 lr 0.000033 time 0.9685 (1.0038) model_time 0.9684 (1.0031) loss 1.0687 (1.2259) grad_norm 15.1784 (10.8082/4.2453) mem 68106MB [2022-12-19 07:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1200/1519] eta 0:05:20 lr 0.000033 time 0.9308 (1.0038) model_time 0.9307 (1.0030) loss 1.3852 (1.2269) grad_norm 14.6166 (10.8172/4.2233) mem 68106MB [2022-12-19 07:05:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1210/1519] eta 0:05:10 lr 0.000033 time 0.9292 (1.0039) model_time 0.9290 (1.0031) loss 1.0334 (1.2264) grad_norm 9.7429 (10.8926/4.2244) mem 68106MB [2022-12-19 07:06:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1220/1519] eta 0:05:00 lr 0.000033 time 0.9263 (1.0039) model_time 0.9262 (1.0032) loss 1.0061 (1.2259) grad_norm 7.9022 (10.8313/4.1506) mem 68106MB [2022-12-19 07:06:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1230/1519] eta 0:04:50 lr 0.000033 time 0.9335 (1.0039) model_time 0.9334 (1.0031) loss 0.7910 (1.2257) grad_norm 8.1649 (10.8100/4.1324) mem 68106MB [2022-12-19 07:06:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1240/1519] eta 0:04:40 lr 0.000033 time 0.9303 (1.0040) model_time 0.9301 (1.0032) loss 0.8712 (1.2260) grad_norm 6.3500 (10.7864/4.1424) mem 68106MB [2022-12-19 07:06:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1250/1519] eta 0:04:30 lr 0.000033 time 0.9306 (1.0039) model_time 0.9305 (1.0032) loss 1.4848 (1.2269) grad_norm 8.1358 (10.7716/4.1256) mem 68106MB [2022-12-19 07:06:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1260/1519] eta 0:04:20 lr 0.000033 time 0.9306 (1.0039) model_time 0.9305 (1.0032) loss 1.4672 (1.2273) grad_norm 15.2970 (10.7255/4.1241) mem 68106MB [2022-12-19 07:06:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9302 (1.0039) model_time 0.9300 (1.0031) loss 1.1369 (1.2261) grad_norm 23.0950 (10.7818/4.2104) mem 68106MB [2022-12-19 07:07:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9351 (1.0039) model_time 0.9349 (1.0031) loss 1.2618 (1.2260) grad_norm 11.6612 (10.8021/4.1905) mem 68106MB [2022-12-19 07:07:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9280 (1.0038) model_time 0.9279 (1.0031) loss 1.2901 (1.2267) grad_norm 10.0179 (10.9026/4.3146) mem 68106MB [2022-12-19 07:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9393 (1.0038) model_time 0.9392 (1.0031) loss 1.4208 (1.2265) grad_norm 10.7780 (10.9230/4.3019) mem 68106MB [2022-12-19 07:07:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9300 (1.0038) model_time 0.9298 (1.0030) loss 1.4172 (1.2276) grad_norm 10.5492 (10.9403/4.2923) mem 68106MB [2022-12-19 07:07:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9337 (1.0037) model_time 0.9336 (1.0030) loss 1.3512 (1.2275) grad_norm 15.1342 (10.9042/4.2056) mem 68106MB [2022-12-19 07:07:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9323 (1.0037) model_time 0.9322 (1.0030) loss 1.3669 (1.2270) grad_norm 13.3466 (10.9179/4.2101) mem 68106MB [2022-12-19 07:08:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9314 (1.0037) model_time 0.9313 (1.0030) loss 1.1762 (1.2276) grad_norm 8.8414 (11.0453/4.2958) mem 68106MB [2022-12-19 07:08:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9221 (1.0037) model_time 0.9220 (1.0030) loss 1.2396 (1.2267) grad_norm 13.1458 (10.9694/4.3002) mem 68106MB [2022-12-19 07:08:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9268 (1.0037) model_time 0.9267 (1.0029) loss 1.1392 (1.2261) grad_norm 7.1671 (10.8574/4.2176) mem 68106MB [2022-12-19 07:08:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9729 (1.0037) model_time 0.9727 (1.0030) loss 1.1728 (1.2251) grad_norm 14.0839 (10.7638/4.1574) mem 68106MB [2022-12-19 07:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9210 (1.0036) model_time 0.9209 (1.0029) loss 0.7993 (1.2252) grad_norm 5.3626 (10.6627/3.9519) mem 68106MB [2022-12-19 07:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9226 (1.0036) model_time 0.9224 (1.0029) loss 1.1992 (1.2255) grad_norm 9.1677 (10.6481/3.9505) mem 68106MB [2022-12-19 07:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1400/1519] eta 0:01:59 lr 0.000033 time 1.1765 (1.0038) model_time 1.1763 (1.0031) loss 1.0845 (1.2254) grad_norm 9.0195 (10.5722/3.9483) mem 68106MB [2022-12-19 07:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9326 (1.0038) model_time 0.9324 (1.0031) loss 1.1104 (1.2258) grad_norm 9.6448 (10.4998/3.9338) mem 68106MB [2022-12-19 07:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9267 (1.0039) model_time 0.9266 (1.0032) loss 1.0918 (1.2256) grad_norm 6.7861 (10.4291/3.9047) mem 68106MB [2022-12-19 07:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9209 (1.0039) model_time 0.9207 (1.0032) loss 1.3239 (1.2255) grad_norm 5.5446 (10.3090/3.8808) mem 68106MB [2022-12-19 07:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9307 (1.0039) model_time 0.9306 (1.0032) loss 0.8908 (1.2252) grad_norm 12.9829 (10.2889/3.8995) mem 68106MB [2022-12-19 07:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9317 (1.0038) model_time 0.9315 (1.0031) loss 0.9988 (1.2243) grad_norm 6.7725 (10.2339/3.9115) mem 68106MB [2022-12-19 07:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9333 (1.0038) model_time 0.9331 (1.0031) loss 1.4844 (1.2244) grad_norm 12.0674 (10.2457/3.8908) mem 68106MB [2022-12-19 07:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9304 (1.0038) model_time 0.9302 (1.0031) loss 1.1828 (1.2237) grad_norm 8.3505 (10.2214/3.9011) mem 68106MB [2022-12-19 07:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9327 (1.0038) model_time 0.9325 (1.0031) loss 1.4037 (1.2244) grad_norm 16.2303 (10.2008/3.9028) mem 68106MB [2022-12-19 07:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9280 (1.0037) model_time 0.9279 (1.0030) loss 0.9170 (1.2238) grad_norm 12.6299 (10.2128/3.8999) mem 68106MB [2022-12-19 07:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9202 (1.0037) model_time 0.9193 (1.0031) loss 0.9289 (1.2243) grad_norm 9.5598 (10.2210/3.9034) mem 68106MB [2022-12-19 07:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [6/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9233 (1.0037) model_time 0.9232 (1.0030) loss 1.3595 (1.2242) grad_norm 9.5117 (10.1691/3.7787) mem 68106MB [2022-12-19 07:11:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 6 training takes 0:25:24 [2022-12-19 07:11:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_6.pth saving...... [2022-12-19 07:11:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_6.pth saved !!! [2022-12-19 07:11:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.642 (0.642) Loss 1.3380 (1.3380) Acc@1 72.222 (72.222) Acc@5 92.361 (92.361) Mem 68106MB [2022-12-19 07:11:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.328) Loss 1.3994 (1.3775) Acc@1 74.306 (71.843) Acc@5 92.014 (92.330) Mem 68106MB [2022-12-19 07:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.313) Loss 1.3009 (1.3667) Acc@1 73.611 (71.759) Acc@5 94.792 (92.791) Mem 68106MB [2022-12-19 07:11:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.308) Loss 1.4463 (1.3681) Acc@1 69.792 (71.528) Acc@5 90.972 (92.843) Mem 68106MB [2022-12-19 07:11:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.306) Loss 1.2876 (1.3520) Acc@1 74.306 (71.875) Acc@5 93.056 (93.030) Mem 68106MB [2022-12-19 07:11:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.304) Loss 1.4042 (1.3452) Acc@1 69.097 (72.066) Acc@5 92.361 (93.151) Mem 68106MB [2022-12-19 07:11:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.295 (0.303) Loss 1.4197 (1.3484) Acc@1 70.139 (71.812) Acc@5 91.319 (93.135) Mem 68106MB [2022-12-19 07:11:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 1.5031 (1.3546) Acc@1 71.528 (71.821) Acc@5 90.972 (93.134) Mem 68106MB [2022-12-19 07:11:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 1.3256 (1.3556) Acc@1 68.403 (71.806) Acc@5 94.444 (93.103) Mem 68106MB [2022-12-19 07:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:6] * Acc@1 71.807 Acc@5 93.066 [2022-12-19 07:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 71.8% [2022-12-19 07:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 07:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 07:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 71.81% [2022-12-19 07:12:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][0/1519] eta 0:32:58 lr 0.000033 time 1.3027 (1.3027) model_time 0.9203 (0.9203) loss 1.4543 (1.4543) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 07:12:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][10/1519] eta 0:25:58 lr 0.000033 time 0.9268 (1.0325) model_time 0.9266 (0.9975) loss 0.9961 (1.3228) grad_norm 7.6565 (10.3343/3.8111) mem 68106MB [2022-12-19 07:12:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][20/1519] eta 0:25:32 lr 0.000033 time 0.9280 (1.0226) model_time 0.9279 (1.0041) loss 1.7950 (1.2278) grad_norm 10.7966 (9.4348/3.1776) mem 68106MB [2022-12-19 07:12:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][30/1519] eta 0:25:13 lr 0.000033 time 0.9289 (1.0163) model_time 0.9288 (1.0037) loss 1.2232 (1.2132) grad_norm 5.5459 (8.4686/2.9471) mem 68106MB [2022-12-19 07:13:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][40/1519] eta 0:25:01 lr 0.000033 time 0.9460 (1.0151) model_time 0.9459 (1.0055) loss 1.2142 (1.2202) grad_norm 7.8151 (9.0462/3.0956) mem 68106MB [2022-12-19 07:13:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][50/1519] eta 0:24:47 lr 0.000033 time 0.9320 (1.0127) model_time 0.9318 (1.0049) loss 1.4254 (1.2178) grad_norm 6.2697 (8.9890/2.8809) mem 68106MB [2022-12-19 07:13:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][60/1519] eta 0:24:34 lr 0.000033 time 0.9211 (1.0108) model_time 0.9209 (1.0042) loss 1.3105 (1.1949) grad_norm 5.4892 (8.9833/2.8132) mem 68106MB [2022-12-19 07:13:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][70/1519] eta 0:24:22 lr 0.000033 time 0.9159 (1.0091) model_time 0.9157 (1.0034) loss 1.4981 (1.1915) grad_norm 10.6828 (9.0996/2.7142) mem 68106MB [2022-12-19 07:13:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][80/1519] eta 0:24:10 lr 0.000033 time 0.9283 (1.0081) model_time 0.9282 (1.0031) loss 1.0349 (1.1799) grad_norm 8.9897 (9.1325/2.6713) mem 68106MB [2022-12-19 07:13:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][90/1519] eta 0:23:58 lr 0.000033 time 0.9288 (1.0069) model_time 0.9286 (1.0023) loss 1.0476 (1.1816) grad_norm 6.5040 (9.2856/2.7931) mem 68106MB [2022-12-19 07:14:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][100/1519] eta 0:23:47 lr 0.000033 time 0.9333 (1.0063) model_time 0.9331 (1.0022) loss 1.3067 (1.1873) grad_norm 8.8874 (9.3939/2.7866) mem 68106MB [2022-12-19 07:14:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][110/1519] eta 0:23:37 lr 0.000033 time 0.9214 (1.0058) model_time 0.9213 (1.0020) loss 1.2602 (1.1850) grad_norm 10.1226 (9.6191/3.5599) mem 68106MB [2022-12-19 07:14:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][120/1519] eta 0:23:26 lr 0.000033 time 0.9349 (1.0057) model_time 0.9347 (1.0022) loss 1.2993 (1.1918) grad_norm 8.6093 (9.5492/3.4260) mem 68106MB [2022-12-19 07:14:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][130/1519] eta 0:23:16 lr 0.000033 time 0.9293 (1.0052) model_time 0.9292 (1.0020) loss 1.2509 (1.1970) grad_norm 16.7208 (9.7076/3.5029) mem 68106MB [2022-12-19 07:14:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][140/1519] eta 0:23:05 lr 0.000033 time 0.9332 (1.0047) model_time 0.9329 (1.0016) loss 1.4269 (1.2110) grad_norm 14.2174 (9.7454/3.4389) mem 68106MB [2022-12-19 07:14:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][150/1519] eta 0:22:55 lr 0.000033 time 0.9288 (1.0048) model_time 0.9286 (1.0019) loss 1.2732 (1.2105) grad_norm 6.2880 (9.5873/3.3961) mem 68106MB [2022-12-19 07:15:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][160/1519] eta 0:22:45 lr 0.000033 time 0.9278 (1.0044) model_time 0.9277 (1.0017) loss 1.1729 (1.2095) grad_norm 5.8965 (9.4665/3.3539) mem 68106MB [2022-12-19 07:15:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][170/1519] eta 0:22:34 lr 0.000033 time 0.9357 (1.0042) model_time 0.9354 (1.0016) loss 0.8822 (1.2119) grad_norm 11.9311 (9.4977/3.3326) mem 68106MB [2022-12-19 07:15:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][180/1519] eta 0:22:24 lr 0.000033 time 0.9298 (1.0043) model_time 0.9297 (1.0019) loss 1.5402 (1.2158) grad_norm 10.4843 (9.3870/3.2948) mem 68106MB [2022-12-19 07:15:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][190/1519] eta 0:22:14 lr 0.000033 time 0.9305 (1.0041) model_time 0.9304 (1.0017) loss 0.8363 (1.2186) grad_norm 12.6309 (9.7784/3.9891) mem 68106MB [2022-12-19 07:15:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][200/1519] eta 0:22:03 lr 0.000033 time 0.9321 (1.0038) model_time 0.9320 (1.0015) loss 1.2851 (1.2134) grad_norm 8.8374 (9.8393/3.9563) mem 68106MB [2022-12-19 07:15:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][210/1519] eta 0:21:54 lr 0.000033 time 0.9257 (1.0044) model_time 0.9256 (1.0022) loss 1.0797 (1.2076) grad_norm 11.8254 (9.8640/3.9201) mem 68106MB [2022-12-19 07:16:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][220/1519] eta 0:21:44 lr 0.000033 time 0.9322 (1.0043) model_time 0.9319 (1.0022) loss 1.1850 (1.2008) grad_norm 9.1348 (9.7878/3.8870) mem 68106MB [2022-12-19 07:16:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][230/1519] eta 0:21:34 lr 0.000033 time 0.9291 (1.0041) model_time 0.9289 (1.0021) loss 1.0826 (1.1977) grad_norm 6.3434 (9.7644/3.8238) mem 68106MB [2022-12-19 07:16:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][240/1519] eta 0:21:24 lr 0.000033 time 0.9436 (1.0041) model_time 0.9434 (1.0021) loss 1.5316 (1.1974) grad_norm 8.7079 (9.7761/3.8390) mem 68106MB [2022-12-19 07:16:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][250/1519] eta 0:21:14 lr 0.000033 time 0.9312 (1.0040) model_time 0.9311 (1.0021) loss 1.1862 (1.1952) grad_norm 11.6278 (9.9453/3.8876) mem 68106MB [2022-12-19 07:16:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][260/1519] eta 0:21:03 lr 0.000033 time 0.9204 (1.0038) model_time 0.9202 (1.0020) loss 0.9159 (1.1932) grad_norm 7.9127 (9.8677/3.8423) mem 68106MB [2022-12-19 07:16:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][270/1519] eta 0:20:53 lr 0.000033 time 0.9275 (1.0038) model_time 0.9274 (1.0021) loss 1.1409 (1.1937) grad_norm 14.2754 (9.8785/3.8657) mem 68106MB [2022-12-19 07:17:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][280/1519] eta 0:20:43 lr 0.000033 time 0.9295 (1.0036) model_time 0.9293 (1.0019) loss 1.1066 (1.1911) grad_norm 13.8444 (9.8673/3.8205) mem 68106MB [2022-12-19 07:17:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][290/1519] eta 0:20:33 lr 0.000033 time 0.9701 (1.0036) model_time 0.9698 (1.0019) loss 1.4837 (1.1942) grad_norm 8.5556 (9.9130/3.8754) mem 68106MB [2022-12-19 07:17:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][300/1519] eta 0:20:23 lr 0.000033 time 0.9262 (1.0034) model_time 0.9260 (1.0018) loss 1.4356 (1.1940) grad_norm 9.3098 (10.1008/4.1630) mem 68106MB [2022-12-19 07:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][310/1519] eta 0:20:12 lr 0.000033 time 0.9319 (1.0033) model_time 0.9318 (1.0017) loss 0.8947 (1.1893) grad_norm 8.0562 (10.0822/4.1190) mem 68106MB [2022-12-19 07:17:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][320/1519] eta 0:20:02 lr 0.000033 time 0.9321 (1.0032) model_time 0.9320 (1.0016) loss 0.7690 (1.1879) grad_norm 8.0310 (10.0715/4.0687) mem 68106MB [2022-12-19 07:17:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][330/1519] eta 0:19:52 lr 0.000033 time 0.9395 (1.0031) model_time 0.9393 (1.0016) loss 1.7371 (1.1905) grad_norm 9.9145 (10.0882/4.0273) mem 68106MB [2022-12-19 07:18:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][340/1519] eta 0:19:43 lr 0.000033 time 0.9171 (1.0035) model_time 0.9169 (1.0020) loss 0.7755 (1.1875) grad_norm 13.8530 (10.2056/4.1368) mem 68106MB [2022-12-19 07:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][350/1519] eta 0:19:32 lr 0.000033 time 0.9213 (1.0033) model_time 0.9211 (1.0019) loss 1.2995 (1.1879) grad_norm 15.6891 (10.2810/4.1151) mem 68106MB [2022-12-19 07:18:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][360/1519] eta 0:19:23 lr 0.000033 time 0.9203 (1.0035) model_time 0.9202 (1.0020) loss 1.3357 (1.1902) grad_norm 13.9533 (10.4047/4.1514) mem 68106MB [2022-12-19 07:18:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][370/1519] eta 0:19:12 lr 0.000033 time 0.9259 (1.0034) model_time 0.9258 (1.0020) loss 0.8736 (1.1916) grad_norm 9.4202 (10.3259/4.1270) mem 68106MB [2022-12-19 07:18:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][380/1519] eta 0:19:02 lr 0.000033 time 0.9514 (1.0033) model_time 0.9513 (1.0019) loss 1.5842 (1.1939) grad_norm 14.9661 (10.3672/4.1994) mem 68106MB [2022-12-19 07:18:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][390/1519] eta 0:18:52 lr 0.000033 time 0.9269 (1.0035) model_time 0.9268 (1.0022) loss 1.6276 (1.1952) grad_norm 7.9943 (10.3467/4.1665) mem 68106MB [2022-12-19 07:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][400/1519] eta 0:18:42 lr 0.000033 time 0.9298 (1.0034) model_time 0.9296 (1.0021) loss 0.9165 (1.1936) grad_norm 12.9303 (10.3051/4.1401) mem 68106MB [2022-12-19 07:19:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][410/1519] eta 0:18:32 lr 0.000033 time 0.9299 (1.0033) model_time 0.9298 (1.0020) loss 0.7527 (1.1907) grad_norm 6.7327 (10.2803/4.1281) mem 68106MB [2022-12-19 07:19:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][420/1519] eta 0:18:22 lr 0.000033 time 0.9299 (1.0032) model_time 0.9294 (1.0019) loss 1.5696 (1.1896) grad_norm 6.5805 (10.2536/4.0943) mem 68106MB [2022-12-19 07:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][430/1519] eta 0:18:12 lr 0.000033 time 0.9376 (1.0032) model_time 0.9374 (1.0019) loss 1.3892 (1.1894) grad_norm 22.2000 (10.3024/4.1633) mem 68106MB [2022-12-19 07:19:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][440/1519] eta 0:18:02 lr 0.000033 time 0.9347 (1.0030) model_time 0.9345 (1.0018) loss 1.0784 (1.1901) grad_norm 6.9288 (10.3652/4.2956) mem 68106MB [2022-12-19 07:19:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][450/1519] eta 0:17:52 lr 0.000033 time 0.9280 (1.0029) model_time 0.9279 (1.0017) loss 1.0997 (1.1889) grad_norm 8.5101 (10.3971/4.2687) mem 68106MB [2022-12-19 07:20:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][460/1519] eta 0:17:41 lr 0.000033 time 0.9311 (1.0028) model_time 0.9309 (1.0016) loss 1.1676 (1.1885) grad_norm 9.3589 (10.4226/4.2955) mem 68106MB [2022-12-19 07:20:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][470/1519] eta 0:17:32 lr 0.000033 time 1.0157 (1.0029) model_time 1.0156 (1.0017) loss 1.0758 (1.1888) grad_norm 6.9907 (10.3828/4.2686) mem 68106MB [2022-12-19 07:20:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][480/1519] eta 0:17:22 lr 0.000033 time 0.9215 (1.0029) model_time 0.9213 (1.0018) loss 1.5165 (1.1920) grad_norm 10.1650 (10.3884/4.2355) mem 68106MB [2022-12-19 07:20:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][490/1519] eta 0:17:11 lr 0.000033 time 0.9347 (1.0029) model_time 0.9346 (1.0017) loss 1.0787 (1.1898) grad_norm 15.1563 (10.3724/4.2219) mem 68106MB [2022-12-19 07:20:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][500/1519] eta 0:17:01 lr 0.000033 time 0.9305 (1.0028) model_time 0.9304 (1.0016) loss 1.0957 (1.1891) grad_norm 6.2119 (10.3340/4.2158) mem 68106MB [2022-12-19 07:20:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][510/1519] eta 0:16:51 lr 0.000033 time 0.9258 (1.0027) model_time 0.9255 (1.0016) loss 1.0606 (1.1883) grad_norm 20.6911 (10.3436/4.2345) mem 68106MB [2022-12-19 07:21:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][520/1519] eta 0:16:42 lr 0.000033 time 0.9252 (1.0032) model_time 0.9250 (1.0021) loss 1.7022 (1.1913) grad_norm 6.7044 (10.3004/4.2068) mem 68106MB [2022-12-19 07:21:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][530/1519] eta 0:16:32 lr 0.000033 time 0.9337 (1.0031) model_time 0.9336 (1.0020) loss 0.8902 (1.1910) grad_norm 6.6587 (10.2619/4.1821) mem 68106MB [2022-12-19 07:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][540/1519] eta 0:16:21 lr 0.000033 time 0.9356 (1.0030) model_time 0.9355 (1.0019) loss 1.0513 (1.1910) grad_norm 7.4901 (10.2485/4.1731) mem 68106MB [2022-12-19 07:21:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][550/1519] eta 0:16:11 lr 0.000033 time 0.9252 (1.0030) model_time 0.9250 (1.0019) loss 1.1051 (1.1877) grad_norm 8.7956 (10.2112/4.1493) mem 68106MB [2022-12-19 07:21:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][560/1519] eta 0:16:01 lr 0.000033 time 0.9294 (1.0029) model_time 0.9292 (1.0018) loss 0.9005 (1.1868) grad_norm 10.6177 (10.1855/4.1237) mem 68106MB [2022-12-19 07:21:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][570/1519] eta 0:15:51 lr 0.000033 time 0.9360 (1.0029) model_time 0.9359 (1.0018) loss 1.0971 (1.1887) grad_norm 10.3084 (10.1850/4.1038) mem 68106MB [2022-12-19 07:22:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][580/1519] eta 0:15:41 lr 0.000033 time 0.9331 (1.0028) model_time 0.9330 (1.0018) loss 1.0336 (1.1884) grad_norm 11.7967 (10.2045/4.1066) mem 68106MB [2022-12-19 07:22:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][590/1519] eta 0:15:31 lr 0.000033 time 0.9352 (1.0028) model_time 0.9351 (1.0017) loss 0.9939 (1.1882) grad_norm 5.4621 (10.1979/4.0867) mem 68106MB [2022-12-19 07:22:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][600/1519] eta 0:15:21 lr 0.000033 time 0.9415 (1.0027) model_time 0.9413 (1.0017) loss 1.1831 (1.1873) grad_norm 10.6516 (10.1997/4.0547) mem 68106MB [2022-12-19 07:22:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][610/1519] eta 0:15:11 lr 0.000033 time 0.9267 (1.0027) model_time 0.9265 (1.0017) loss 1.0639 (1.1865) grad_norm 8.6844 (10.1845/4.0538) mem 68106MB [2022-12-19 07:22:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][620/1519] eta 0:15:01 lr 0.000033 time 0.9281 (1.0026) model_time 0.9280 (1.0016) loss 1.1339 (1.1863) grad_norm 11.3925 (10.2650/4.0770) mem 68106MB [2022-12-19 07:22:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][630/1519] eta 0:14:51 lr 0.000033 time 0.9302 (1.0025) model_time 0.9301 (1.0016) loss 0.9485 (1.1856) grad_norm 8.2430 (10.3357/4.0762) mem 68106MB [2022-12-19 07:23:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][640/1519] eta 0:14:41 lr 0.000033 time 0.9285 (1.0025) model_time 0.9283 (1.0015) loss 1.0950 (1.1851) grad_norm 10.4831 (10.3577/4.0687) mem 68106MB [2022-12-19 07:23:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][650/1519] eta 0:14:31 lr 0.000033 time 0.9313 (1.0025) model_time 0.9312 (1.0015) loss 0.9663 (1.1843) grad_norm 9.0861 (10.3546/4.0661) mem 68106MB [2022-12-19 07:23:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][660/1519] eta 0:14:21 lr 0.000033 time 0.9314 (1.0026) model_time 0.9313 (1.0016) loss 0.9419 (1.1819) grad_norm 6.7447 (10.3261/4.0718) mem 68106MB [2022-12-19 07:23:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][670/1519] eta 0:14:11 lr 0.000033 time 0.9147 (1.0028) model_time 0.9145 (1.0019) loss 1.5248 (1.1808) grad_norm 7.4140 (10.3062/4.0861) mem 68106MB [2022-12-19 07:23:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][680/1519] eta 0:14:01 lr 0.000033 time 0.9326 (1.0027) model_time 0.9324 (1.0018) loss 1.0915 (1.1802) grad_norm 9.9508 (10.2846/4.0873) mem 68106MB [2022-12-19 07:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][690/1519] eta 0:13:51 lr 0.000033 time 0.9344 (1.0027) model_time 0.9342 (1.0018) loss 1.2321 (1.1811) grad_norm 7.7381 (10.2454/4.0778) mem 68106MB [2022-12-19 07:24:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][700/1519] eta 0:13:41 lr 0.000033 time 0.9245 (1.0031) model_time 0.9244 (1.0022) loss 1.1534 (1.1840) grad_norm 12.3883 (10.2236/4.0818) mem 68106MB [2022-12-19 07:24:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][710/1519] eta 0:13:31 lr 0.000033 time 0.9307 (1.0030) model_time 0.9305 (1.0021) loss 1.1034 (1.1825) grad_norm 6.5440 (10.1559/3.9734) mem 68106MB [2022-12-19 07:24:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][720/1519] eta 0:13:21 lr 0.000033 time 0.9272 (1.0030) model_time 0.9270 (1.0021) loss 1.0340 (1.1810) grad_norm 17.4442 (10.2026/3.9958) mem 68106MB [2022-12-19 07:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][730/1519] eta 0:13:11 lr 0.000033 time 0.9397 (1.0029) model_time 0.9395 (1.0020) loss 1.2423 (1.1810) grad_norm 7.0326 (10.1693/3.9760) mem 68106MB [2022-12-19 07:24:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][740/1519] eta 0:13:01 lr 0.000033 time 0.9698 (1.0029) model_time 0.9697 (1.0020) loss 1.3160 (1.1815) grad_norm 12.5112 (10.2032/3.9986) mem 68106MB [2022-12-19 07:24:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][750/1519] eta 0:12:51 lr 0.000033 time 0.9267 (1.0028) model_time 0.9265 (1.0020) loss 1.6030 (1.1805) grad_norm 12.2712 (10.2244/3.9956) mem 68106MB [2022-12-19 07:25:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][760/1519] eta 0:12:41 lr 0.000033 time 0.9308 (1.0028) model_time 0.9303 (1.0019) loss 0.9866 (1.1811) grad_norm 6.1640 (10.2292/3.9904) mem 68106MB [2022-12-19 07:25:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][770/1519] eta 0:12:31 lr 0.000033 time 0.9302 (1.0027) model_time 0.9299 (1.0019) loss 1.5049 (1.1812) grad_norm 13.3947 (10.2194/3.9829) mem 68106MB [2022-12-19 07:25:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][780/1519] eta 0:12:20 lr 0.000033 time 0.9315 (1.0027) model_time 0.9313 (1.0018) loss 0.9323 (1.1815) grad_norm 7.0989 (10.2355/3.9695) mem 68106MB [2022-12-19 07:25:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][790/1519] eta 0:12:10 lr 0.000033 time 0.9320 (1.0027) model_time 0.9317 (1.0018) loss 0.9496 (1.1805) grad_norm 8.5857 (10.1336/3.7818) mem 68106MB [2022-12-19 07:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][800/1519] eta 0:12:00 lr 0.000033 time 0.9370 (1.0027) model_time 0.9368 (1.0018) loss 1.0112 (1.1801) grad_norm 7.0426 (10.0710/3.7797) mem 68106MB [2022-12-19 07:25:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][810/1519] eta 0:11:50 lr 0.000033 time 0.9312 (1.0026) model_time 0.9310 (1.0018) loss 1.3673 (1.1789) grad_norm 6.8766 (10.0600/3.8102) mem 68106MB [2022-12-19 07:26:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][820/1519] eta 0:11:40 lr 0.000033 time 0.9291 (1.0025) model_time 0.9289 (1.0017) loss 1.1209 (1.1785) grad_norm 8.8856 (10.0938/3.7968) mem 68106MB [2022-12-19 07:26:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][830/1519] eta 0:11:30 lr 0.000033 time 0.9284 (1.0025) model_time 0.9281 (1.0017) loss 0.9707 (1.1795) grad_norm 13.1304 (10.0974/3.8019) mem 68106MB [2022-12-19 07:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][840/1519] eta 0:11:20 lr 0.000033 time 0.9339 (1.0024) model_time 0.9337 (1.0016) loss 1.0316 (1.1798) grad_norm 10.5723 (10.0867/3.7686) mem 68106MB [2022-12-19 07:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][850/1519] eta 0:11:10 lr 0.000033 time 0.9309 (1.0024) model_time 0.9307 (1.0016) loss 1.2382 (1.1798) grad_norm 6.3419 (9.9857/3.7317) mem 68106MB [2022-12-19 07:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][860/1519] eta 0:11:00 lr 0.000033 time 0.9328 (1.0024) model_time 0.9326 (1.0015) loss 1.4153 (1.1803) grad_norm 19.1708 (10.0507/3.7662) mem 68106MB [2022-12-19 07:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][870/1519] eta 0:10:50 lr 0.000033 time 0.9335 (1.0023) model_time 0.9334 (1.0015) loss 1.1334 (1.1792) grad_norm 10.8045 (10.0342/3.7322) mem 68106MB [2022-12-19 07:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][880/1519] eta 0:10:40 lr 0.000033 time 0.9320 (1.0023) model_time 0.9318 (1.0015) loss 1.0976 (1.1792) grad_norm 10.1551 (10.0582/3.7663) mem 68106MB [2022-12-19 07:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][890/1519] eta 0:10:30 lr 0.000033 time 0.9316 (1.0022) model_time 0.9314 (1.0014) loss 1.0874 (1.1794) grad_norm 5.0917 (10.0093/3.7201) mem 68106MB [2022-12-19 07:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][900/1519] eta 0:10:20 lr 0.000033 time 0.9348 (1.0022) model_time 0.9346 (1.0014) loss 1.1041 (1.1795) grad_norm 12.4243 (9.9478/3.5481) mem 68106MB [2022-12-19 07:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][910/1519] eta 0:10:10 lr 0.000033 time 0.9299 (1.0022) model_time 0.9298 (1.0014) loss 1.1954 (1.1781) grad_norm 12.5913 (9.9432/3.5472) mem 68106MB [2022-12-19 07:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][920/1519] eta 0:10:00 lr 0.000033 time 0.9306 (1.0021) model_time 0.9304 (1.0013) loss 1.6727 (1.1793) grad_norm 9.1095 (9.9280/3.5453) mem 68106MB [2022-12-19 07:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][930/1519] eta 0:09:50 lr 0.000033 time 0.9365 (1.0021) model_time 0.9362 (1.0013) loss 1.4161 (1.1796) grad_norm 12.1425 (9.9343/3.5447) mem 68106MB [2022-12-19 07:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][940/1519] eta 0:09:40 lr 0.000033 time 0.9301 (1.0021) model_time 0.9299 (1.0013) loss 1.0702 (1.1793) grad_norm 12.2401 (9.8568/3.4382) mem 68106MB [2022-12-19 07:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][950/1519] eta 0:09:30 lr 0.000033 time 0.9287 (1.0020) model_time 0.9285 (1.0012) loss 0.9732 (1.1806) grad_norm 6.4638 (9.7984/3.4297) mem 68106MB [2022-12-19 07:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][960/1519] eta 0:09:20 lr 0.000033 time 0.9332 (1.0023) model_time 0.9331 (1.0015) loss 1.1758 (1.1803) grad_norm 5.7484 (9.6875/3.3631) mem 68106MB [2022-12-19 07:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][970/1519] eta 0:09:10 lr 0.000033 time 0.9142 (1.0023) model_time 0.9140 (1.0015) loss 0.8863 (1.1790) grad_norm 12.5975 (9.7113/3.3692) mem 68106MB [2022-12-19 07:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][980/1519] eta 0:09:00 lr 0.000033 time 0.9320 (1.0023) model_time 0.9319 (1.0015) loss 1.0253 (1.1785) grad_norm 7.2504 (9.6747/3.2751) mem 68106MB [2022-12-19 07:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][990/1519] eta 0:08:50 lr 0.000033 time 0.9319 (1.0023) model_time 0.9318 (1.0016) loss 1.1277 (1.1786) grad_norm 6.5027 (9.6735/3.2879) mem 68106MB [2022-12-19 07:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1000/1519] eta 0:08:40 lr 0.000033 time 0.9364 (1.0023) model_time 0.9363 (1.0015) loss 1.1533 (1.1788) grad_norm 8.4499 (9.6842/3.2782) mem 68106MB [2022-12-19 07:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1010/1519] eta 0:08:30 lr 0.000033 time 0.9329 (1.0023) model_time 0.9328 (1.0015) loss 1.1651 (1.1799) grad_norm 7.3945 (9.6769/3.2677) mem 68106MB [2022-12-19 07:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1020/1519] eta 0:08:20 lr 0.000033 time 0.9380 (1.0026) model_time 0.9379 (1.0018) loss 1.1844 (1.1791) grad_norm 12.1222 (9.6718/3.2710) mem 68106MB [2022-12-19 07:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1030/1519] eta 0:08:10 lr 0.000033 time 0.9290 (1.0025) model_time 0.9289 (1.0018) loss 1.1070 (1.1787) grad_norm 10.0940 (9.6379/3.1688) mem 68106MB [2022-12-19 07:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1040/1519] eta 0:08:00 lr 0.000033 time 0.9261 (1.0025) model_time 0.9260 (1.0017) loss 1.1289 (1.1785) grad_norm 12.7712 (9.5765/2.9863) mem 68106MB [2022-12-19 07:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1050/1519] eta 0:07:50 lr 0.000033 time 0.9295 (1.0024) model_time 0.9293 (1.0017) loss 1.4835 (1.1793) grad_norm 12.0532 (9.5562/2.9700) mem 68106MB [2022-12-19 07:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1060/1519] eta 0:07:40 lr 0.000033 time 0.9293 (1.0025) model_time 0.9292 (1.0017) loss 0.8032 (1.1792) grad_norm 7.1792 (9.5422/2.9295) mem 68106MB [2022-12-19 07:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1070/1519] eta 0:07:30 lr 0.000033 time 0.9352 (1.0024) model_time 0.9351 (1.0017) loss 1.0372 (1.1796) grad_norm 10.6128 (9.6042/2.9822) mem 68106MB [2022-12-19 07:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1080/1519] eta 0:07:20 lr 0.000033 time 0.9320 (1.0024) model_time 0.9319 (1.0017) loss 1.1762 (1.1791) grad_norm 9.7846 (9.5896/3.0137) mem 68106MB [2022-12-19 07:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1090/1519] eta 0:07:10 lr 0.000033 time 0.9326 (1.0023) model_time 0.9324 (1.0016) loss 1.2632 (1.1784) grad_norm 9.4483 (9.5748/2.9899) mem 68106MB [2022-12-19 07:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1100/1519] eta 0:06:59 lr 0.000033 time 0.9278 (1.0023) model_time 0.9277 (1.0016) loss 1.4216 (1.1779) grad_norm 8.1856 (9.5684/2.9655) mem 68106MB [2022-12-19 07:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1110/1519] eta 0:06:49 lr 0.000033 time 0.9254 (1.0023) model_time 0.9252 (1.0016) loss 0.8930 (1.1777) grad_norm 13.9917 (9.6171/2.9596) mem 68106MB [2022-12-19 07:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1120/1519] eta 0:06:39 lr 0.000033 time 0.9347 (1.0022) model_time 0.9346 (1.0015) loss 0.7577 (1.1775) grad_norm 7.2288 (9.6048/2.9687) mem 68106MB [2022-12-19 07:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1130/1519] eta 0:06:29 lr 0.000033 time 0.9265 (1.0022) model_time 0.9263 (1.0015) loss 1.0310 (1.1779) grad_norm 11.3760 (9.6975/3.0468) mem 68106MB [2022-12-19 07:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1140/1519] eta 0:06:19 lr 0.000033 time 0.9311 (1.0022) model_time 0.9310 (1.0015) loss 0.9240 (1.1768) grad_norm 17.1550 (9.7277/3.0508) mem 68106MB [2022-12-19 07:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1150/1519] eta 0:06:09 lr 0.000033 time 0.9345 (1.0022) model_time 0.9343 (1.0015) loss 1.1316 (1.1767) grad_norm 9.4609 (9.7896/3.0695) mem 68106MB [2022-12-19 07:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1160/1519] eta 0:05:59 lr 0.000033 time 0.9303 (1.0021) model_time 0.9302 (1.0014) loss 1.3625 (1.1758) grad_norm 9.5470 (9.7788/3.0749) mem 68106MB [2022-12-19 07:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1170/1519] eta 0:05:49 lr 0.000033 time 0.9361 (1.0021) model_time 0.9359 (1.0014) loss 1.2507 (1.1767) grad_norm 6.4560 (9.7406/3.0688) mem 68106MB [2022-12-19 07:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1180/1519] eta 0:05:39 lr 0.000033 time 0.9325 (1.0021) model_time 0.9323 (1.0014) loss 1.2922 (1.1772) grad_norm 12.5218 (9.7128/3.0371) mem 68106MB [2022-12-19 07:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1190/1519] eta 0:05:29 lr 0.000033 time 0.9302 (1.0021) model_time 0.9300 (1.0014) loss 1.3478 (1.1775) grad_norm 6.3712 (9.6821/3.0320) mem 68106MB [2022-12-19 07:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1200/1519] eta 0:05:19 lr 0.000033 time 0.9315 (1.0020) model_time 0.9314 (1.0013) loss 1.4711 (1.1768) grad_norm 9.0246 (9.6675/3.0569) mem 68106MB [2022-12-19 07:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1210/1519] eta 0:05:09 lr 0.000033 time 0.9311 (1.0020) model_time 0.9309 (1.0013) loss 1.2168 (1.1777) grad_norm 8.1943 (9.7348/3.1544) mem 68106MB [2022-12-19 07:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1220/1519] eta 0:04:59 lr 0.000033 time 0.9273 (1.0020) model_time 0.9271 (1.0013) loss 0.9619 (1.1770) grad_norm 6.9840 (9.6962/3.1390) mem 68106MB [2022-12-19 07:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1230/1519] eta 0:04:49 lr 0.000033 time 0.9257 (1.0020) model_time 0.9256 (1.0013) loss 1.0257 (1.1775) grad_norm 11.1093 (9.6664/3.1041) mem 68106MB [2022-12-19 07:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1240/1519] eta 0:04:39 lr 0.000033 time 0.9438 (1.0022) model_time 0.9436 (1.0015) loss 0.9202 (1.1765) grad_norm 6.7240 (9.6489/3.1810) mem 68106MB [2022-12-19 07:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1250/1519] eta 0:04:29 lr 0.000033 time 0.9294 (1.0021) model_time 0.9293 (1.0015) loss 0.9205 (1.1764) grad_norm 8.3583 (9.6493/3.1828) mem 68106MB [2022-12-19 07:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1260/1519] eta 0:04:19 lr 0.000033 time 0.9291 (1.0021) model_time 0.9290 (1.0015) loss 1.4190 (1.1757) grad_norm 6.0976 (9.7315/3.2369) mem 68106MB [2022-12-19 07:33:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9396 (1.0021) model_time 0.9395 (1.0014) loss 0.8867 (1.1753) grad_norm 13.3274 (9.7615/3.2277) mem 68106MB [2022-12-19 07:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9297 (1.0021) model_time 0.9296 (1.0014) loss 1.3344 (1.1757) grad_norm 9.9309 (9.7666/3.2347) mem 68106MB [2022-12-19 07:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9277 (1.0021) model_time 0.9276 (1.0014) loss 1.6064 (1.1761) grad_norm 8.4297 (9.7771/3.2322) mem 68106MB [2022-12-19 07:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9425 (1.0022) model_time 0.9424 (1.0016) loss 1.1569 (1.1757) grad_norm 6.8869 (9.7408/3.2376) mem 68106MB [2022-12-19 07:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9315 (1.0022) model_time 0.9314 (1.0015) loss 1.1815 (1.1759) grad_norm 11.5892 (9.7513/3.2371) mem 68106MB [2022-12-19 07:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9321 (1.0022) model_time 0.9320 (1.0015) loss 0.9280 (1.1755) grad_norm 6.8485 (9.7138/3.2087) mem 68106MB [2022-12-19 07:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9554 (1.0023) model_time 0.9553 (1.0017) loss 1.2324 (1.1762) grad_norm 6.0533 (9.7050/3.2074) mem 68106MB [2022-12-19 07:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9311 (1.0023) model_time 0.9309 (1.0017) loss 1.0441 (1.1762) grad_norm 8.5887 (9.6940/3.2048) mem 68106MB [2022-12-19 07:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9365 (1.0023) model_time 0.9364 (1.0016) loss 1.2418 (1.1762) grad_norm 12.7973 (9.7274/3.1991) mem 68106MB [2022-12-19 07:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9654 (1.0023) model_time 0.9653 (1.0016) loss 0.9110 (1.1750) grad_norm 9.6893 (9.7517/3.1919) mem 68106MB [2022-12-19 07:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9278 (1.0023) model_time 0.9276 (1.0016) loss 0.8887 (1.1752) grad_norm 6.9586 (9.7368/3.1901) mem 68106MB [2022-12-19 07:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9314 (1.0022) model_time 0.9313 (1.0016) loss 1.3057 (1.1751) grad_norm 6.8791 (9.7381/3.1935) mem 68106MB [2022-12-19 07:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9461 (1.0022) model_time 0.9460 (1.0016) loss 1.4011 (1.1758) grad_norm 7.0424 (9.7046/3.1747) mem 68106MB [2022-12-19 07:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9279 (1.0022) model_time 0.9278 (1.0016) loss 0.9362 (1.1757) grad_norm 9.8068 (9.7591/3.1976) mem 68106MB [2022-12-19 07:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9306 (1.0022) model_time 0.9304 (1.0016) loss 1.1059 (1.1759) grad_norm 10.1616 (9.7381/3.1570) mem 68106MB [2022-12-19 07:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9341 (1.0022) model_time 0.9339 (1.0015) loss 1.4238 (1.1765) grad_norm 6.4278 (9.7064/3.1617) mem 68106MB [2022-12-19 07:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9265 (1.0021) model_time 0.9263 (1.0015) loss 1.0594 (1.1762) grad_norm 9.3562 (9.7059/3.1610) mem 68106MB [2022-12-19 07:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9369 (1.0022) model_time 0.9367 (1.0015) loss 1.2862 (1.1769) grad_norm 7.1064 (9.7047/3.1709) mem 68106MB [2022-12-19 07:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.8905 (1.0022) model_time 0.8903 (1.0016) loss 1.1865 (1.1769) grad_norm 16.8272 (9.7521/3.2042) mem 68106MB [2022-12-19 07:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9345 (1.0022) model_time 0.9343 (1.0015) loss 1.1184 (1.1767) grad_norm 11.7846 (9.7343/3.1629) mem 68106MB [2022-12-19 07:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9290 (1.0021) model_time 0.9289 (1.0015) loss 0.9355 (1.1767) grad_norm 8.3734 (9.7513/3.2219) mem 68106MB [2022-12-19 07:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9347 (1.0022) model_time 0.9345 (1.0015) loss 1.4122 (1.1774) grad_norm 7.0357 (9.6844/3.1867) mem 68106MB [2022-12-19 07:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9340 (1.0021) model_time 0.9339 (1.0015) loss 1.2929 (1.1766) grad_norm 7.1289 (9.6790/3.1929) mem 68106MB [2022-12-19 07:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9283 (1.0022) model_time 0.9282 (1.0015) loss 1.1006 (1.1763) grad_norm 5.8452 (9.6525/3.1829) mem 68106MB [2022-12-19 07:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [7/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9250 (1.0022) model_time 0.9249 (1.0016) loss 1.4491 (1.1764) grad_norm 7.1954 (9.6402/3.1851) mem 68106MB [2022-12-19 07:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 7 training takes 0:25:22 [2022-12-19 07:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_7.pth saving...... [2022-12-19 07:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_7.pth saved !!! [2022-12-19 07:38:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.664 (0.664) Loss 1.1349 (1.1349) Acc@1 75.347 (75.347) Acc@5 94.097 (94.097) Mem 68106MB [2022-12-19 07:38:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.305 (0.330) Loss 1.1563 (1.1455) Acc@1 81.250 (77.683) Acc@5 95.139 (94.571) Mem 68106MB [2022-12-19 07:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.315) Loss 1.0772 (1.1406) Acc@1 78.125 (77.265) Acc@5 96.181 (94.726) Mem 68106MB [2022-12-19 07:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.309) Loss 1.2106 (1.1436) Acc@1 76.042 (77.162) Acc@5 93.403 (94.612) Mem 68106MB [2022-12-19 07:38:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.306) Loss 1.0847 (1.1324) Acc@1 79.861 (77.083) Acc@5 94.444 (94.749) Mem 68106MB [2022-12-19 07:38:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.305) Loss 1.1675 (1.1264) Acc@1 76.389 (77.267) Acc@5 94.444 (94.853) Mem 68106MB [2022-12-19 07:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.304) Loss 1.1815 (1.1274) Acc@1 74.306 (77.083) Acc@5 95.486 (94.888) Mem 68106MB [2022-12-19 07:38:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 1.2880 (1.1332) Acc@1 75.347 (76.946) Acc@5 91.319 (94.875) Mem 68106MB [2022-12-19 07:38:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.302) Loss 1.0967 (1.1343) Acc@1 77.778 (76.895) Acc@5 95.833 (94.877) Mem 68106MB [2022-12-19 07:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:7] * Acc@1 76.895 Acc@5 94.863 [2022-12-19 07:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 76.9% [2022-12-19 07:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 07:39:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 07:39:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 76.90% [2022-12-19 07:39:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][0/1519] eta 0:33:51 lr 0.000033 time 1.3375 (1.3375) model_time 0.9604 (0.9604) loss 1.0961 (1.0961) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 07:39:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][10/1519] eta 0:26:04 lr 0.000033 time 0.9492 (1.0370) model_time 0.9491 (1.0023) loss 0.9876 (0.9885) grad_norm 8.8117 (9.6919/3.4100) mem 68106MB [2022-12-19 07:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][20/1519] eta 0:25:30 lr 0.000033 time 0.9370 (1.0211) model_time 0.9369 (1.0028) loss 1.0255 (0.9991) grad_norm 10.0778 (10.6383/3.7852) mem 68106MB [2022-12-19 07:39:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][30/1519] eta 0:25:13 lr 0.000033 time 0.9403 (1.0167) model_time 0.9401 (1.0042) loss 1.2119 (1.0390) grad_norm 12.6399 (10.5270/3.4637) mem 68106MB [2022-12-19 07:39:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][40/1519] eta 0:24:58 lr 0.000033 time 0.9298 (1.0129) model_time 0.9296 (1.0033) loss 0.9499 (1.0436) grad_norm 9.0692 (10.2963/3.1385) mem 68106MB [2022-12-19 07:39:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][50/1519] eta 0:24:45 lr 0.000033 time 0.9315 (1.0110) model_time 0.9314 (1.0032) loss 1.1415 (1.0710) grad_norm 5.9860 (9.8590/3.1168) mem 68106MB [2022-12-19 07:40:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][60/1519] eta 0:24:33 lr 0.000033 time 0.9355 (1.0096) model_time 0.9353 (1.0030) loss 1.2534 (1.0729) grad_norm 6.8720 (9.7133/3.0397) mem 68106MB [2022-12-19 07:40:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][70/1519] eta 0:24:23 lr 0.000033 time 0.9303 (1.0101) model_time 0.9298 (1.0043) loss 1.0002 (1.0724) grad_norm 5.9567 (9.3979/3.0283) mem 68106MB [2022-12-19 07:40:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][80/1519] eta 0:24:12 lr 0.000033 time 0.9306 (1.0091) model_time 0.9304 (1.0040) loss 1.1336 (1.0869) grad_norm 15.1890 (9.3437/3.0504) mem 68106MB [2022-12-19 07:40:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][90/1519] eta 0:24:00 lr 0.000033 time 0.9300 (1.0078) model_time 0.9299 (1.0033) loss 1.0058 (1.0873) grad_norm 13.2908 (9.5826/2.9829) mem 68106MB [2022-12-19 07:40:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][100/1519] eta 0:23:48 lr 0.000033 time 0.9309 (1.0070) model_time 0.9307 (1.0028) loss 1.2161 (1.1003) grad_norm 7.0888 (9.3397/2.9626) mem 68106MB [2022-12-19 07:40:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][110/1519] eta 0:23:38 lr 0.000033 time 0.9296 (1.0067) model_time 0.9294 (1.0029) loss 1.1414 (1.1154) grad_norm 7.0865 (9.3885/2.8794) mem 68106MB [2022-12-19 07:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][120/1519] eta 0:23:28 lr 0.000033 time 0.9289 (1.0065) model_time 0.9287 (1.0029) loss 1.2278 (1.1195) grad_norm 12.1311 (9.3740/2.8693) mem 68106MB [2022-12-19 07:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][130/1519] eta 0:23:17 lr 0.000033 time 0.9400 (1.0060) model_time 0.9399 (1.0027) loss 1.0584 (1.1255) grad_norm 6.6419 (10.0127/6.3373) mem 68106MB [2022-12-19 07:41:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][140/1519] eta 0:23:06 lr 0.000033 time 0.9329 (1.0055) model_time 0.9326 (1.0024) loss 1.1712 (1.1324) grad_norm 10.8014 (10.0740/6.1534) mem 68106MB [2022-12-19 07:41:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][150/1519] eta 0:22:55 lr 0.000033 time 0.9323 (1.0051) model_time 0.9321 (1.0022) loss 1.2966 (1.1378) grad_norm 8.4740 (10.0221/5.9782) mem 68106MB [2022-12-19 07:41:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][160/1519] eta 0:22:45 lr 0.000033 time 0.9274 (1.0049) model_time 0.9272 (1.0021) loss 1.6328 (1.1438) grad_norm 11.3380 (9.9927/5.8299) mem 68106MB [2022-12-19 07:41:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][170/1519] eta 0:22:35 lr 0.000033 time 0.9292 (1.0046) model_time 0.9291 (1.0020) loss 1.0274 (1.1398) grad_norm 18.4595 (9.9985/5.7851) mem 68106MB [2022-12-19 07:42:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][180/1519] eta 0:22:24 lr 0.000033 time 0.9269 (1.0041) model_time 0.9267 (1.0017) loss 1.3951 (1.1426) grad_norm 17.8833 (10.0126/5.7230) mem 68106MB [2022-12-19 07:42:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][190/1519] eta 0:22:14 lr 0.000033 time 0.9283 (1.0039) model_time 0.9280 (1.0015) loss 1.2022 (1.1450) grad_norm 11.6243 (10.1316/5.6807) mem 68106MB [2022-12-19 07:42:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][200/1519] eta 0:22:03 lr 0.000033 time 0.9287 (1.0036) model_time 0.9285 (1.0013) loss 1.0773 (1.1413) grad_norm 8.8509 (10.0768/5.5645) mem 68106MB [2022-12-19 07:42:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][210/1519] eta 0:21:53 lr 0.000033 time 0.9304 (1.0035) model_time 0.9303 (1.0013) loss 1.1522 (1.1444) grad_norm 12.5115 (10.0272/5.4487) mem 68106MB [2022-12-19 07:42:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][220/1519] eta 0:21:43 lr 0.000033 time 0.9304 (1.0032) model_time 0.9303 (1.0011) loss 1.1064 (1.1452) grad_norm 6.6251 (10.0104/5.3433) mem 68106MB [2022-12-19 07:42:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][230/1519] eta 0:21:32 lr 0.000033 time 0.9305 (1.0030) model_time 0.9304 (1.0009) loss 1.2502 (1.1451) grad_norm 13.3376 (10.1622/5.4176) mem 68106MB [2022-12-19 07:43:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][240/1519] eta 0:21:22 lr 0.000033 time 0.9328 (1.0029) model_time 0.9326 (1.0009) loss 0.7914 (1.1467) grad_norm 13.6511 (10.2652/5.4369) mem 68106MB [2022-12-19 07:43:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][250/1519] eta 0:21:12 lr 0.000033 time 0.9350 (1.0028) model_time 0.9348 (1.0009) loss 0.8600 (1.1421) grad_norm 6.6335 (10.2324/5.3553) mem 68106MB [2022-12-19 07:43:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][260/1519] eta 0:21:02 lr 0.000033 time 0.9317 (1.0028) model_time 0.9316 (1.0010) loss 1.1995 (1.1425) grad_norm 10.1958 (10.2826/5.2889) mem 68106MB [2022-12-19 07:43:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][270/1519] eta 0:20:52 lr 0.000033 time 0.9273 (1.0028) model_time 0.9271 (1.0010) loss 1.4270 (1.1432) grad_norm 11.9082 (10.2147/5.2138) mem 68106MB [2022-12-19 07:43:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][280/1519] eta 0:20:42 lr 0.000033 time 0.9313 (1.0026) model_time 0.9311 (1.0009) loss 0.9649 (1.1439) grad_norm 8.2919 (10.1381/5.1434) mem 68106MB [2022-12-19 07:43:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][290/1519] eta 0:20:31 lr 0.000033 time 0.9259 (1.0024) model_time 0.9257 (1.0008) loss 1.1394 (1.1439) grad_norm 9.8266 (10.1052/5.0647) mem 68106MB [2022-12-19 07:44:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][300/1519] eta 0:20:22 lr 0.000033 time 0.9324 (1.0026) model_time 0.9322 (1.0010) loss 0.8006 (1.1436) grad_norm 6.2250 (10.0531/5.0145) mem 68106MB [2022-12-19 07:44:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][310/1519] eta 0:20:12 lr 0.000033 time 0.9286 (1.0025) model_time 0.9284 (1.0009) loss 1.0886 (1.1427) grad_norm 6.1754 (9.9913/4.9584) mem 68106MB [2022-12-19 07:44:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][320/1519] eta 0:20:01 lr 0.000033 time 0.9335 (1.0024) model_time 0.9333 (1.0009) loss 1.1081 (1.1431) grad_norm 7.7264 (9.9099/4.9050) mem 68106MB [2022-12-19 07:44:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][330/1519] eta 0:19:51 lr 0.000033 time 0.9279 (1.0022) model_time 0.9278 (1.0007) loss 1.3070 (1.1430) grad_norm 9.4937 (9.9787/4.9380) mem 68106MB [2022-12-19 07:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][340/1519] eta 0:19:42 lr 0.000033 time 0.9286 (1.0028) model_time 0.9284 (1.0013) loss 1.1929 (1.1413) grad_norm 7.9134 (9.9231/4.8804) mem 68106MB [2022-12-19 07:44:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][350/1519] eta 0:19:32 lr 0.000033 time 0.9297 (1.0027) model_time 0.9296 (1.0013) loss 0.9128 (1.1423) grad_norm 6.0685 (9.8948/4.8531) mem 68106MB [2022-12-19 07:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][360/1519] eta 0:19:22 lr 0.000033 time 0.9306 (1.0027) model_time 0.9305 (1.0012) loss 0.9780 (1.1433) grad_norm 11.3003 (9.9882/4.8349) mem 68106MB [2022-12-19 07:45:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][370/1519] eta 0:19:12 lr 0.000033 time 0.9352 (1.0027) model_time 0.9351 (1.0013) loss 0.9725 (1.1431) grad_norm 14.7504 (10.0689/4.8568) mem 68106MB [2022-12-19 07:45:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][380/1519] eta 0:19:01 lr 0.000033 time 0.9292 (1.0026) model_time 0.9291 (1.0012) loss 0.9016 (1.1441) grad_norm 10.0381 (10.0617/4.8130) mem 68106MB [2022-12-19 07:45:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][390/1519] eta 0:18:52 lr 0.000033 time 0.9356 (1.0030) model_time 0.9355 (1.0016) loss 1.1555 (1.1443) grad_norm 20.6637 (10.2892/5.2311) mem 68106MB [2022-12-19 07:45:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][400/1519] eta 0:18:42 lr 0.000033 time 0.9271 (1.0028) model_time 0.9270 (1.0015) loss 1.2346 (1.1485) grad_norm 9.6475 (10.3610/5.2342) mem 68106MB [2022-12-19 07:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][410/1519] eta 0:18:32 lr 0.000033 time 0.9322 (1.0028) model_time 0.9321 (1.0015) loss 1.4267 (1.1487) grad_norm 9.0694 (10.3601/5.1847) mem 68106MB [2022-12-19 07:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][420/1519] eta 0:18:21 lr 0.000033 time 0.9294 (1.0027) model_time 0.9292 (1.0014) loss 0.8939 (1.1486) grad_norm 8.4866 (10.3776/5.1420) mem 68106MB [2022-12-19 07:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][430/1519] eta 0:18:12 lr 0.000033 time 0.9922 (1.0028) model_time 0.9920 (1.0015) loss 1.1249 (1.1488) grad_norm 9.4697 (10.3446/5.0996) mem 68106MB [2022-12-19 07:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][440/1519] eta 0:18:02 lr 0.000033 time 0.9291 (1.0028) model_time 0.9290 (1.0016) loss 0.7770 (1.1473) grad_norm 5.9293 (10.2870/5.0632) mem 68106MB [2022-12-19 07:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][450/1519] eta 0:17:51 lr 0.000033 time 0.9356 (1.0027) model_time 0.9354 (1.0015) loss 1.0606 (1.1465) grad_norm 7.0618 (10.2985/5.0469) mem 68106MB [2022-12-19 07:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][460/1519] eta 0:17:41 lr 0.000033 time 0.9349 (1.0027) model_time 0.9348 (1.0015) loss 1.3877 (1.1479) grad_norm 9.9484 (10.3325/5.0093) mem 68106MB [2022-12-19 07:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][470/1519] eta 0:17:31 lr 0.000033 time 0.9342 (1.0026) model_time 0.9341 (1.0014) loss 0.8661 (1.1486) grad_norm 8.2659 (10.3247/4.9864) mem 68106MB [2022-12-19 07:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][480/1519] eta 0:17:21 lr 0.000033 time 0.9362 (1.0025) model_time 0.9361 (1.0014) loss 1.3090 (1.1473) grad_norm 6.6032 (10.3271/4.9748) mem 68106MB [2022-12-19 07:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][490/1519] eta 0:17:11 lr 0.000033 time 0.9338 (1.0024) model_time 0.9337 (1.0013) loss 1.1772 (1.1453) grad_norm 11.8175 (10.3094/4.9308) mem 68106MB [2022-12-19 07:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][500/1519] eta 0:17:01 lr 0.000033 time 0.9297 (1.0023) model_time 0.9295 (1.0012) loss 1.4024 (1.1438) grad_norm 7.9305 (10.2902/4.9303) mem 68106MB [2022-12-19 07:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][510/1519] eta 0:16:51 lr 0.000033 time 0.9353 (1.0022) model_time 0.9352 (1.0011) loss 0.9112 (1.1430) grad_norm 8.6562 (10.2626/4.8916) mem 68106MB [2022-12-19 07:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][520/1519] eta 0:16:41 lr 0.000033 time 0.9326 (1.0023) model_time 0.9325 (1.0012) loss 1.0948 (1.1424) grad_norm 6.5245 (10.2097/4.8644) mem 68106MB [2022-12-19 07:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][530/1519] eta 0:16:31 lr 0.000033 time 0.9323 (1.0022) model_time 0.9322 (1.0012) loss 1.1517 (1.1448) grad_norm 10.4375 (10.2302/4.8349) mem 68106MB [2022-12-19 07:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][540/1519] eta 0:16:21 lr 0.000033 time 0.9278 (1.0021) model_time 0.9277 (1.0011) loss 1.1284 (1.1444) grad_norm 7.3626 (10.2059/4.8016) mem 68106MB [2022-12-19 07:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][550/1519] eta 0:16:11 lr 0.000033 time 0.9252 (1.0021) model_time 0.9250 (1.0011) loss 0.9366 (1.1441) grad_norm 14.5679 (10.2032/4.7773) mem 68106MB [2022-12-19 07:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][560/1519] eta 0:16:00 lr 0.000033 time 0.9287 (1.0020) model_time 0.9285 (1.0010) loss 1.3139 (1.1445) grad_norm 7.8645 (10.1864/4.7520) mem 68106MB [2022-12-19 07:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][570/1519] eta 0:15:50 lr 0.000033 time 0.9334 (1.0020) model_time 0.9333 (1.0010) loss 1.1049 (1.1435) grad_norm 8.6403 (10.1629/4.7292) mem 68106MB [2022-12-19 07:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][580/1519] eta 0:15:40 lr 0.000033 time 0.9911 (1.0021) model_time 0.9909 (1.0011) loss 0.9332 (1.1410) grad_norm 16.0670 (10.1386/4.7180) mem 68106MB [2022-12-19 07:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][590/1519] eta 0:15:30 lr 0.000033 time 0.9308 (1.0020) model_time 0.9306 (1.0010) loss 1.3191 (1.1405) grad_norm 26.3343 (10.2026/4.7942) mem 68106MB [2022-12-19 07:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][600/1519] eta 0:15:20 lr 0.000033 time 0.9441 (1.0020) model_time 0.9440 (1.0010) loss 1.0212 (1.1406) grad_norm 18.6329 (10.2256/4.7840) mem 68106MB [2022-12-19 07:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][610/1519] eta 0:15:11 lr 0.000033 time 1.2009 (1.0024) model_time 1.2007 (1.0014) loss 1.2157 (1.1389) grad_norm 9.0001 (10.2212/4.7923) mem 68106MB [2022-12-19 07:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][620/1519] eta 0:15:01 lr 0.000033 time 0.9298 (1.0023) model_time 0.9296 (1.0013) loss 0.9879 (1.1394) grad_norm 7.1878 (10.2737/4.9020) mem 68106MB [2022-12-19 07:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][630/1519] eta 0:14:50 lr 0.000033 time 0.9197 (1.0022) model_time 0.9195 (1.0012) loss 1.0799 (1.1399) grad_norm 6.7051 (10.2380/4.9014) mem 68106MB [2022-12-19 07:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][640/1519] eta 0:14:40 lr 0.000033 time 0.9305 (1.0021) model_time 0.9303 (1.0012) loss 1.2955 (1.1393) grad_norm 10.4765 (10.2105/4.9083) mem 68106MB [2022-12-19 07:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][650/1519] eta 0:14:30 lr 0.000033 time 0.9313 (1.0020) model_time 0.9311 (1.0011) loss 1.0574 (1.1391) grad_norm 10.2823 (10.2684/4.9077) mem 68106MB [2022-12-19 07:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][660/1519] eta 0:14:20 lr 0.000033 time 0.9306 (1.0020) model_time 0.9305 (1.0010) loss 0.8774 (1.1384) grad_norm 14.0573 (10.2729/4.9188) mem 68106MB [2022-12-19 07:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][670/1519] eta 0:14:10 lr 0.000033 time 0.9686 (1.0020) model_time 0.9685 (1.0011) loss 1.1736 (1.1401) grad_norm 6.4991 (10.2734/4.9138) mem 68106MB [2022-12-19 07:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][680/1519] eta 0:14:00 lr 0.000033 time 0.9306 (1.0020) model_time 0.9305 (1.0011) loss 1.2447 (1.1414) grad_norm 12.6264 (10.2612/4.9105) mem 68106MB [2022-12-19 07:50:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][690/1519] eta 0:13:50 lr 0.000033 time 0.9302 (1.0019) model_time 0.9301 (1.0010) loss 1.3256 (1.1414) grad_norm 22.1415 (10.2555/4.9594) mem 68106MB [2022-12-19 07:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][700/1519] eta 0:13:40 lr 0.000033 time 0.9351 (1.0022) model_time 0.9350 (1.0013) loss 1.3456 (1.1425) grad_norm 5.8751 (10.2492/4.9607) mem 68106MB [2022-12-19 07:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][710/1519] eta 0:13:30 lr 0.000033 time 0.9275 (1.0021) model_time 0.9274 (1.0012) loss 1.3379 (1.1436) grad_norm 24.1994 (10.2882/5.0324) mem 68106MB [2022-12-19 07:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][720/1519] eta 0:13:20 lr 0.000033 time 0.9336 (1.0021) model_time 0.9335 (1.0012) loss 1.1499 (1.1451) grad_norm 7.8880 (10.2844/5.0338) mem 68106MB [2022-12-19 07:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][730/1519] eta 0:13:10 lr 0.000033 time 0.9281 (1.0021) model_time 0.9279 (1.0012) loss 1.0280 (1.1463) grad_norm 6.9521 (10.1985/4.3740) mem 68106MB [2022-12-19 07:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][740/1519] eta 0:13:00 lr 0.000033 time 0.9302 (1.0020) model_time 0.9300 (1.0011) loss 1.0061 (1.1453) grad_norm 6.7791 (10.1568/4.3907) mem 68106MB [2022-12-19 07:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][750/1519] eta 0:12:50 lr 0.000033 time 0.8992 (1.0021) model_time 0.8990 (1.0012) loss 1.1784 (1.1440) grad_norm 6.9845 (10.1669/4.3873) mem 68106MB [2022-12-19 07:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][760/1519] eta 0:12:40 lr 0.000033 time 1.0182 (1.0022) model_time 1.0181 (1.0013) loss 0.9396 (1.1441) grad_norm 8.5027 (10.1716/4.3773) mem 68106MB [2022-12-19 07:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][770/1519] eta 0:12:30 lr 0.000033 time 0.9329 (1.0022) model_time 0.9328 (1.0013) loss 1.3363 (1.1435) grad_norm 10.6159 (10.1726/4.3307) mem 68106MB [2022-12-19 07:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][780/1519] eta 0:12:20 lr 0.000033 time 0.9314 (1.0021) model_time 0.9313 (1.0013) loss 0.9608 (1.1432) grad_norm 12.8248 (10.1847/4.3090) mem 68106MB [2022-12-19 07:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][790/1519] eta 0:12:10 lr 0.000033 time 1.0339 (1.0023) model_time 1.0337 (1.0014) loss 1.2414 (1.1428) grad_norm 13.4408 (10.1598/4.2795) mem 68106MB [2022-12-19 07:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][800/1519] eta 0:12:00 lr 0.000033 time 0.9322 (1.0022) model_time 0.9320 (1.0014) loss 0.9756 (1.1413) grad_norm 9.5856 (10.2347/4.3240) mem 68106MB [2022-12-19 07:52:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][810/1519] eta 0:11:50 lr 0.000033 time 0.9342 (1.0021) model_time 0.9341 (1.0013) loss 0.7976 (1.1399) grad_norm 17.1331 (10.3205/4.3563) mem 68106MB [2022-12-19 07:52:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][820/1519] eta 0:11:40 lr 0.000033 time 0.9365 (1.0021) model_time 0.9364 (1.0013) loss 1.0267 (1.1394) grad_norm 13.4697 (10.3432/4.3607) mem 68106MB [2022-12-19 07:52:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][830/1519] eta 0:11:30 lr 0.000033 time 0.9304 (1.0020) model_time 0.9302 (1.0012) loss 1.1453 (1.1400) grad_norm 12.5864 (10.2849/4.2889) mem 68106MB [2022-12-19 07:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][840/1519] eta 0:11:20 lr 0.000033 time 0.9328 (1.0021) model_time 0.9326 (1.0013) loss 1.1857 (1.1397) grad_norm 7.4449 (10.2159/4.2365) mem 68106MB [2022-12-19 07:53:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][850/1519] eta 0:11:10 lr 0.000033 time 0.9234 (1.0021) model_time 0.9233 (1.0013) loss 1.0071 (1.1399) grad_norm 12.3345 (10.2215/4.2490) mem 68106MB [2022-12-19 07:53:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][860/1519] eta 0:11:00 lr 0.000033 time 0.9338 (1.0021) model_time 0.9337 (1.0013) loss 0.8873 (1.1392) grad_norm 5.2593 (10.1848/4.2391) mem 68106MB [2022-12-19 07:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][870/1519] eta 0:10:50 lr 0.000033 time 0.9299 (1.0021) model_time 0.9296 (1.0013) loss 0.9524 (1.1397) grad_norm 8.0317 (10.1903/4.2369) mem 68106MB [2022-12-19 07:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][880/1519] eta 0:10:40 lr 0.000033 time 0.9295 (1.0020) model_time 0.9293 (1.0012) loss 1.1398 (1.1405) grad_norm 7.0128 (10.1919/4.2371) mem 68106MB [2022-12-19 07:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][890/1519] eta 0:10:30 lr 0.000033 time 0.9287 (1.0020) model_time 0.9285 (1.0012) loss 0.7733 (1.1390) grad_norm 7.8381 (10.1928/4.2406) mem 68106MB [2022-12-19 07:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][900/1519] eta 0:10:20 lr 0.000033 time 0.9347 (1.0019) model_time 0.9346 (1.0012) loss 1.0737 (1.1395) grad_norm 11.4761 (10.2131/4.2334) mem 68106MB [2022-12-19 07:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][910/1519] eta 0:10:10 lr 0.000033 time 0.9408 (1.0019) model_time 0.9407 (1.0012) loss 1.0874 (1.1394) grad_norm 6.5673 (10.2211/4.2337) mem 68106MB [2022-12-19 07:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][920/1519] eta 0:10:00 lr 0.000033 time 0.9289 (1.0019) model_time 0.9287 (1.0011) loss 1.2583 (1.1390) grad_norm 6.1225 (10.2197/4.2359) mem 68106MB [2022-12-19 07:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][930/1519] eta 0:09:50 lr 0.000033 time 0.9382 (1.0020) model_time 0.9381 (1.0012) loss 1.0733 (1.1387) grad_norm 9.7108 (10.1637/4.1761) mem 68106MB [2022-12-19 07:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][940/1519] eta 0:09:40 lr 0.000033 time 0.9276 (1.0019) model_time 0.9274 (1.0012) loss 1.2217 (1.1377) grad_norm 9.4521 (10.1793/4.1806) mem 68106MB [2022-12-19 07:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][950/1519] eta 0:09:30 lr 0.000033 time 0.9322 (1.0019) model_time 0.9320 (1.0012) loss 1.1067 (1.1379) grad_norm 11.9240 (10.1995/4.1541) mem 68106MB [2022-12-19 07:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][960/1519] eta 0:09:20 lr 0.000033 time 0.9342 (1.0019) model_time 0.9341 (1.0011) loss 0.9158 (1.1381) grad_norm 6.3131 (10.1079/4.1396) mem 68106MB [2022-12-19 07:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][970/1519] eta 0:09:10 lr 0.000033 time 0.9354 (1.0019) model_time 0.9353 (1.0011) loss 0.9554 (1.1374) grad_norm 7.9654 (10.0265/4.0856) mem 68106MB [2022-12-19 07:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][980/1519] eta 0:08:59 lr 0.000033 time 0.9370 (1.0019) model_time 0.9368 (1.0011) loss 1.1171 (1.1360) grad_norm 5.6775 (10.0055/4.1152) mem 68106MB [2022-12-19 07:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][990/1519] eta 0:08:49 lr 0.000033 time 0.9358 (1.0019) model_time 0.9354 (1.0011) loss 1.2349 (1.1364) grad_norm 5.9095 (9.8141/3.7301) mem 68106MB [2022-12-19 07:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1000/1519] eta 0:08:39 lr 0.000033 time 0.9321 (1.0019) model_time 0.9320 (1.0011) loss 1.0211 (1.1363) grad_norm 6.0292 (9.7284/3.6703) mem 68106MB [2022-12-19 07:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1010/1519] eta 0:08:30 lr 0.000033 time 0.9304 (1.0020) model_time 0.9303 (1.0013) loss 1.2074 (1.1356) grad_norm 6.4474 (9.7073/3.6687) mem 68106MB [2022-12-19 07:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1020/1519] eta 0:08:19 lr 0.000033 time 0.9328 (1.0020) model_time 0.9327 (1.0013) loss 1.6875 (1.1354) grad_norm 11.3061 (9.6686/3.6552) mem 68106MB [2022-12-19 07:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1030/1519] eta 0:08:09 lr 0.000033 time 0.9394 (1.0020) model_time 0.9393 (1.0012) loss 1.0016 (1.1354) grad_norm 6.8548 (9.6589/3.6544) mem 68106MB [2022-12-19 07:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1040/1519] eta 0:07:59 lr 0.000033 time 0.9364 (1.0019) model_time 0.9362 (1.0012) loss 0.9403 (1.1348) grad_norm 6.3049 (9.6616/3.6517) mem 68106MB [2022-12-19 07:56:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1050/1519] eta 0:07:49 lr 0.000033 time 0.9308 (1.0019) model_time 0.9306 (1.0012) loss 1.1445 (1.1341) grad_norm 31.7260 (9.7094/3.8421) mem 68106MB [2022-12-19 07:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1060/1519] eta 0:07:39 lr 0.000033 time 0.9288 (1.0019) model_time 0.9287 (1.0012) loss 1.3080 (1.1358) grad_norm 8.5858 (9.6498/3.8266) mem 68106MB [2022-12-19 07:56:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1070/1519] eta 0:07:29 lr 0.000033 time 0.9357 (1.0019) model_time 0.9356 (1.0012) loss 1.3059 (1.1365) grad_norm 6.5307 (9.6525/3.8218) mem 68106MB [2022-12-19 07:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1080/1519] eta 0:07:19 lr 0.000033 time 0.9289 (1.0022) model_time 0.9288 (1.0015) loss 1.0691 (1.1380) grad_norm 9.1920 (9.6088/3.7909) mem 68106MB [2022-12-19 07:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1090/1519] eta 0:07:09 lr 0.000033 time 0.9312 (1.0021) model_time 0.9311 (1.0014) loss 1.2447 (1.1386) grad_norm 7.6886 (9.6019/3.7952) mem 68106MB [2022-12-19 07:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1100/1519] eta 0:06:59 lr 0.000033 time 0.9302 (1.0022) model_time 0.9300 (1.0015) loss 0.9944 (1.1382) grad_norm 6.6403 (9.5615/3.7607) mem 68106MB [2022-12-19 07:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1110/1519] eta 0:06:49 lr 0.000033 time 0.9299 (1.0022) model_time 0.9297 (1.0015) loss 0.8206 (1.1379) grad_norm 11.4834 (9.5764/3.7625) mem 68106MB [2022-12-19 07:57:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1120/1519] eta 0:06:39 lr 0.000033 time 0.9286 (1.0022) model_time 0.9285 (1.0015) loss 0.8837 (1.1380) grad_norm 10.4362 (9.5843/3.7560) mem 68106MB [2022-12-19 07:57:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1130/1519] eta 0:06:29 lr 0.000033 time 0.9341 (1.0022) model_time 0.9339 (1.0015) loss 1.2098 (1.1373) grad_norm 8.7417 (9.5352/3.7414) mem 68106MB [2022-12-19 07:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1140/1519] eta 0:06:19 lr 0.000033 time 0.9312 (1.0022) model_time 0.9311 (1.0015) loss 1.1975 (1.1381) grad_norm 7.7851 (9.5763/3.7930) mem 68106MB [2022-12-19 07:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1150/1519] eta 0:06:09 lr 0.000033 time 0.9293 (1.0024) model_time 0.9292 (1.0017) loss 1.0371 (1.1379) grad_norm 12.1504 (9.5979/3.7871) mem 68106MB [2022-12-19 07:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1160/1519] eta 0:05:59 lr 0.000033 time 0.9386 (1.0023) model_time 0.9384 (1.0017) loss 1.0785 (1.1369) grad_norm 7.4445 (9.6341/3.8347) mem 68106MB [2022-12-19 07:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1170/1519] eta 0:05:49 lr 0.000033 time 0.9314 (1.0024) model_time 0.9313 (1.0017) loss 1.1379 (1.1364) grad_norm 19.1569 (9.6651/3.8593) mem 68106MB [2022-12-19 07:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1180/1519] eta 0:05:39 lr 0.000033 time 0.9273 (1.0023) model_time 0.9271 (1.0016) loss 1.1182 (1.1359) grad_norm 10.1583 (9.6607/3.8333) mem 68106MB [2022-12-19 07:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1190/1519] eta 0:05:29 lr 0.000033 time 0.9277 (1.0024) model_time 0.9276 (1.0017) loss 0.9545 (1.1355) grad_norm 23.1649 (9.6534/3.7884) mem 68106MB [2022-12-19 07:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1200/1519] eta 0:05:19 lr 0.000033 time 0.9311 (1.0024) model_time 0.9310 (1.0017) loss 0.9539 (1.1352) grad_norm 5.3424 (9.5838/3.7634) mem 68106MB [2022-12-19 07:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1210/1519] eta 0:05:09 lr 0.000033 time 0.9287 (1.0024) model_time 0.9286 (1.0017) loss 1.1306 (1.1351) grad_norm 7.0888 (9.5744/3.7498) mem 68106MB [2022-12-19 07:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1220/1519] eta 0:04:59 lr 0.000033 time 0.9399 (1.0024) model_time 0.9398 (1.0017) loss 1.3283 (1.1349) grad_norm 10.8450 (9.4854/3.5539) mem 68106MB [2022-12-19 07:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1230/1519] eta 0:04:49 lr 0.000033 time 0.9247 (1.0023) model_time 0.9246 (1.0017) loss 0.9635 (1.1342) grad_norm 8.7339 (9.4955/3.5544) mem 68106MB [2022-12-19 07:59:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1240/1519] eta 0:04:39 lr 0.000033 time 0.9328 (1.0023) model_time 0.9327 (1.0017) loss 0.8315 (1.1332) grad_norm 7.8202 (9.5048/3.5507) mem 68106MB [2022-12-19 07:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1250/1519] eta 0:04:29 lr 0.000033 time 0.9294 (1.0023) model_time 0.9293 (1.0017) loss 1.0934 (1.1338) grad_norm 11.9824 (9.4673/3.5336) mem 68106MB [2022-12-19 08:00:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1260/1519] eta 0:04:19 lr 0.000033 time 0.9342 (1.0023) model_time 0.9341 (1.0017) loss 0.7699 (1.1338) grad_norm 7.5835 (9.4661/3.5117) mem 68106MB [2022-12-19 08:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9305 (1.0023) model_time 0.9303 (1.0017) loss 1.0928 (1.1334) grad_norm 7.5336 (9.4774/3.5102) mem 68106MB [2022-12-19 08:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9337 (1.0023) model_time 0.9336 (1.0016) loss 1.5401 (1.1335) grad_norm 8.0628 (9.5137/3.5036) mem 68106MB [2022-12-19 08:00:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9347 (1.0023) model_time 0.9345 (1.0016) loss 0.7696 (1.1332) grad_norm 8.2593 (9.4500/3.4385) mem 68106MB [2022-12-19 08:00:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9340 (1.0023) model_time 0.9338 (1.0016) loss 1.2538 (1.1328) grad_norm 9.0864 (9.4887/3.4241) mem 68106MB [2022-12-19 08:00:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9288 (1.0022) model_time 0.9286 (1.0016) loss 1.4388 (1.1328) grad_norm 11.6077 (9.4646/3.3343) mem 68106MB [2022-12-19 08:01:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9293 (1.0022) model_time 0.9291 (1.0016) loss 1.2132 (1.1322) grad_norm 6.9637 (9.4601/3.3311) mem 68106MB [2022-12-19 08:01:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9286 (1.0022) model_time 0.9285 (1.0015) loss 1.0208 (1.1314) grad_norm 10.4739 (9.4313/3.2283) mem 68106MB [2022-12-19 08:01:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9334 (1.0022) model_time 0.9332 (1.0015) loss 1.0854 (1.1315) grad_norm 11.2336 (9.4445/3.2022) mem 68106MB [2022-12-19 08:01:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9348 (1.0021) model_time 0.9346 (1.0015) loss 1.2699 (1.1314) grad_norm 9.0175 (9.4632/3.2216) mem 68106MB [2022-12-19 08:01:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9282 (1.0021) model_time 0.9281 (1.0015) loss 0.9794 (1.1316) grad_norm 7.4989 (9.4763/3.2601) mem 68106MB [2022-12-19 08:01:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9351 (1.0021) model_time 0.9350 (1.0014) loss 0.9952 (1.1316) grad_norm 7.0756 (9.5141/3.3724) mem 68106MB [2022-12-19 08:02:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9302 (1.0021) model_time 0.9300 (1.0015) loss 1.4571 (1.1314) grad_norm 9.2481 (9.4710/3.3530) mem 68106MB [2022-12-19 08:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9334 (1.0021) model_time 0.9333 (1.0015) loss 0.9695 (1.1316) grad_norm 11.5222 (9.4570/3.3704) mem 68106MB [2022-12-19 08:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9414 (1.0021) model_time 0.9413 (1.0015) loss 1.2497 (1.1318) grad_norm 6.4888 (9.3663/3.2957) mem 68106MB [2022-12-19 08:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9388 (1.0022) model_time 0.9386 (1.0016) loss 1.0714 (1.1332) grad_norm 10.7851 (9.2662/3.2303) mem 68106MB [2022-12-19 08:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9284 (1.0024) model_time 0.9283 (1.0018) loss 0.8457 (1.1332) grad_norm 9.5786 (9.2609/3.2244) mem 68106MB [2022-12-19 08:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9337 (1.0024) model_time 0.9336 (1.0017) loss 0.9600 (1.1331) grad_norm 5.0405 (9.2302/3.2265) mem 68106MB [2022-12-19 08:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9416 (1.0023) model_time 0.9415 (1.0017) loss 0.9059 (1.1325) grad_norm 7.8057 (9.2455/3.2338) mem 68106MB [2022-12-19 08:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9301 (1.0023) model_time 0.9299 (1.0017) loss 0.9746 (1.1317) grad_norm 7.9708 (9.2186/3.2019) mem 68106MB [2022-12-19 08:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9384 (1.0023) model_time 0.9383 (1.0017) loss 1.1765 (1.1316) grad_norm 9.2759 (9.2122/3.1995) mem 68106MB [2022-12-19 08:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9333 (1.0023) model_time 0.9332 (1.0017) loss 1.4354 (1.1314) grad_norm 7.2846 (9.2175/3.2031) mem 68106MB [2022-12-19 08:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9272 (1.0023) model_time 0.9271 (1.0017) loss 1.0872 (1.1322) grad_norm 9.9573 (9.2344/3.1966) mem 68106MB [2022-12-19 08:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9236 (1.0023) model_time 0.9234 (1.0017) loss 1.0693 (1.1328) grad_norm 7.8066 (9.2139/3.1942) mem 68106MB [2022-12-19 08:04:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9206 (1.0025) model_time 0.9204 (1.0019) loss 1.0991 (1.1329) grad_norm 10.0046 (9.2199/3.1980) mem 68106MB [2022-12-19 08:04:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [8/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9297 (1.0024) model_time 0.9295 (1.0018) loss 1.1913 (1.1327) grad_norm 7.4180 (9.2024/3.1945) mem 68106MB [2022-12-19 08:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 8 training takes 0:25:22 [2022-12-19 08:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_8.pth saving...... [2022-12-19 08:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_8.pth saved !!! [2022-12-19 08:04:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.698 (0.698) Loss 0.9832 (0.9832) Acc@1 80.556 (80.556) Acc@5 95.833 (95.833) Mem 68106MB [2022-12-19 08:04:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.334) Loss 1.0128 (1.0020) Acc@1 81.597 (80.492) Acc@5 95.833 (95.360) Mem 68106MB [2022-12-19 08:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.317) Loss 0.9553 (0.9968) Acc@1 81.597 (79.894) Acc@5 95.833 (95.503) Mem 68106MB [2022-12-19 08:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 1.0308 (0.9971) Acc@1 79.861 (79.738) Acc@5 93.750 (95.441) Mem 68106MB [2022-12-19 08:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.308) Loss 0.9405 (0.9829) Acc@1 80.903 (80.014) Acc@5 94.792 (95.664) Mem 68106MB [2022-12-19 08:05:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.306) Loss 1.0108 (0.9776) Acc@1 79.861 (80.276) Acc@5 95.833 (95.765) Mem 68106MB [2022-12-19 08:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 1.0060 (0.9777) Acc@1 80.208 (80.168) Acc@5 95.139 (95.811) Mem 68106MB [2022-12-19 08:05:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 1.0865 (0.9811) Acc@1 80.208 (80.174) Acc@5 95.486 (95.824) Mem 68106MB [2022-12-19 08:05:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.9297 (0.9808) Acc@1 80.556 (80.144) Acc@5 95.833 (95.838) Mem 68106MB [2022-12-19 08:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:8] * Acc@1 80.149 Acc@5 95.846 [2022-12-19 08:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 80.1% [2022-12-19 08:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 08:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 08:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 80.15% [2022-12-19 08:05:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][0/1519] eta 0:34:55 lr 0.000033 time 1.3794 (1.3794) model_time 0.9376 (0.9376) loss 1.2447 (1.2447) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 08:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][10/1519] eta 0:26:00 lr 0.000033 time 0.9201 (1.0344) model_time 0.9200 (0.9939) loss 1.0229 (1.1070) grad_norm 6.7709 (10.4145/2.9486) mem 68106MB [2022-12-19 08:05:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][20/1519] eta 0:25:25 lr 0.000033 time 0.9271 (1.0175) model_time 0.9270 (0.9961) loss 1.0406 (1.0976) grad_norm 14.9221 (10.0979/2.9150) mem 68106MB [2022-12-19 08:06:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][30/1519] eta 0:25:07 lr 0.000033 time 0.9375 (1.0125) model_time 0.9374 (0.9979) loss 1.1011 (1.1282) grad_norm 9.3284 (9.6273/2.6267) mem 68106MB [2022-12-19 08:06:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][40/1519] eta 0:24:52 lr 0.000033 time 0.9240 (1.0088) model_time 0.9232 (0.9977) loss 1.0581 (1.1281) grad_norm 6.3008 (9.2935/2.5966) mem 68106MB [2022-12-19 08:06:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][50/1519] eta 0:24:39 lr 0.000033 time 0.9369 (1.0073) model_time 0.9367 (0.9983) loss 1.4756 (1.1222) grad_norm 5.9983 (9.4803/2.6453) mem 68106MB [2022-12-19 08:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][60/1519] eta 0:24:27 lr 0.000033 time 0.9185 (1.0060) model_time 0.9184 (0.9984) loss 0.8376 (1.1317) grad_norm 17.8074 (9.4666/3.0374) mem 68106MB [2022-12-19 08:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][70/1519] eta 0:24:16 lr 0.000033 time 0.9173 (1.0050) model_time 0.9172 (0.9983) loss 1.0919 (1.1317) grad_norm 13.7700 (9.4414/2.9590) mem 68106MB [2022-12-19 08:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][80/1519] eta 0:24:05 lr 0.000033 time 0.9259 (1.0045) model_time 0.9257 (0.9986) loss 1.5284 (1.1322) grad_norm 11.9369 (9.4260/2.8123) mem 68106MB [2022-12-19 08:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][90/1519] eta 0:23:55 lr 0.000033 time 0.9342 (1.0043) model_time 0.9341 (0.9990) loss 0.9950 (1.1275) grad_norm 12.4686 (9.5032/2.8284) mem 68106MB [2022-12-19 08:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][100/1519] eta 0:23:44 lr 0.000033 time 0.9941 (1.0039) model_time 0.9940 (0.9992) loss 1.1416 (1.1282) grad_norm 8.8877 (9.4967/2.6970) mem 68106MB [2022-12-19 08:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][110/1519] eta 0:23:35 lr 0.000033 time 0.9313 (1.0043) model_time 0.9312 (1.0000) loss 1.2438 (1.1214) grad_norm 10.8408 (9.4408/2.6258) mem 68106MB [2022-12-19 08:07:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][120/1519] eta 0:23:25 lr 0.000033 time 0.9397 (1.0048) model_time 0.9395 (1.0007) loss 0.9382 (1.1186) grad_norm 5.9956 (9.3822/2.6902) mem 68106MB [2022-12-19 08:07:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][130/1519] eta 0:23:14 lr 0.000033 time 0.9351 (1.0043) model_time 0.9349 (1.0005) loss 1.2869 (1.1294) grad_norm 9.4483 (9.3148/2.6379) mem 68106MB [2022-12-19 08:07:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][140/1519] eta 0:23:04 lr 0.000033 time 0.9337 (1.0038) model_time 0.9334 (1.0003) loss 1.0165 (1.1249) grad_norm 14.0083 (9.6153/2.7856) mem 68106MB [2022-12-19 08:08:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][150/1519] eta 0:22:53 lr 0.000033 time 0.9311 (1.0033) model_time 0.9309 (1.0000) loss 0.9841 (1.1278) grad_norm 11.6091 (9.5501/2.7442) mem 68106MB [2022-12-19 08:08:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][160/1519] eta 0:22:43 lr 0.000033 time 0.9291 (1.0031) model_time 0.9290 (1.0000) loss 1.4659 (1.1311) grad_norm 10.4708 (9.6408/2.7700) mem 68106MB [2022-12-19 08:08:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][170/1519] eta 0:22:33 lr 0.000033 time 0.9223 (1.0034) model_time 0.9221 (1.0005) loss 0.9334 (1.1246) grad_norm 9.8039 (9.5647/2.7271) mem 68106MB [2022-12-19 08:08:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][180/1519] eta 0:22:23 lr 0.000033 time 0.9344 (1.0032) model_time 0.9342 (1.0003) loss 1.3919 (1.1254) grad_norm 16.4374 (9.6447/2.8592) mem 68106MB [2022-12-19 08:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][190/1519] eta 0:22:13 lr 0.000033 time 0.9319 (1.0030) model_time 0.9318 (1.0003) loss 1.3363 (1.1242) grad_norm 21.3558 (9.6864/3.0632) mem 68106MB [2022-12-19 08:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][200/1519] eta 0:22:02 lr 0.000033 time 0.9288 (1.0029) model_time 0.9287 (1.0003) loss 1.0770 (1.1246) grad_norm 13.4200 (9.8213/3.1656) mem 68106MB [2022-12-19 08:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][210/1519] eta 0:21:54 lr 0.000033 time 0.9004 (1.0039) model_time 0.9002 (1.0014) loss 1.2926 (1.1254) grad_norm 17.1776 (9.8244/3.2184) mem 68106MB [2022-12-19 08:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][220/1519] eta 0:21:44 lr 0.000033 time 0.9417 (1.0038) model_time 0.9415 (1.0015) loss 1.0119 (1.1223) grad_norm 14.0857 (9.8700/3.2175) mem 68106MB [2022-12-19 08:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][230/1519] eta 0:21:33 lr 0.000033 time 0.9267 (1.0037) model_time 0.9265 (1.0014) loss 0.8839 (1.1211) grad_norm 7.3497 (9.9240/3.3628) mem 68106MB [2022-12-19 08:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][240/1519] eta 0:21:23 lr 0.000033 time 0.9290 (1.0035) model_time 0.9288 (1.0013) loss 1.1688 (1.1203) grad_norm 7.9436 (9.9457/3.3134) mem 68106MB [2022-12-19 08:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][250/1519] eta 0:21:13 lr 0.000033 time 0.9286 (1.0036) model_time 0.9285 (1.0014) loss 1.3796 (1.1238) grad_norm 9.0393 (9.9392/3.2732) mem 68106MB [2022-12-19 08:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][260/1519] eta 0:21:03 lr 0.000033 time 0.9301 (1.0035) model_time 0.9299 (1.0014) loss 1.4299 (1.1228) grad_norm 10.3052 (9.9310/3.2601) mem 68106MB [2022-12-19 08:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][270/1519] eta 0:20:53 lr 0.000033 time 0.9162 (1.0035) model_time 0.9160 (1.0015) loss 0.9815 (1.1193) grad_norm 8.3986 (9.8734/3.2233) mem 68106MB [2022-12-19 08:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][280/1519] eta 0:20:43 lr 0.000033 time 0.9312 (1.0033) model_time 0.9311 (1.0013) loss 1.1151 (1.1207) grad_norm 6.5255 (9.8313/3.1888) mem 68106MB [2022-12-19 08:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][290/1519] eta 0:20:32 lr 0.000033 time 0.9344 (1.0032) model_time 0.9343 (1.0013) loss 0.7526 (1.1193) grad_norm 6.1320 (9.7304/3.1806) mem 68106MB [2022-12-19 08:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][300/1519] eta 0:20:23 lr 0.000033 time 0.9313 (1.0036) model_time 0.9312 (1.0018) loss 0.9228 (1.1149) grad_norm 7.0587 (9.6741/3.1535) mem 68106MB [2022-12-19 08:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][310/1519] eta 0:20:13 lr 0.000033 time 0.9318 (1.0034) model_time 0.9315 (1.0016) loss 0.7556 (1.1142) grad_norm 6.5155 (9.6708/3.1176) mem 68106MB [2022-12-19 08:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][320/1519] eta 0:20:02 lr 0.000033 time 0.9317 (1.0033) model_time 0.9315 (1.0015) loss 0.9171 (1.1120) grad_norm 6.6869 (9.6498/3.0898) mem 68106MB [2022-12-19 08:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][330/1519] eta 0:19:52 lr 0.000033 time 0.9232 (1.0031) model_time 0.9231 (1.0014) loss 1.5959 (1.1130) grad_norm 11.6397 (9.6244/3.0684) mem 68106MB [2022-12-19 08:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][340/1519] eta 0:19:42 lr 0.000033 time 0.9304 (1.0030) model_time 0.9302 (1.0013) loss 0.7572 (1.1113) grad_norm 9.3137 (9.5852/3.0352) mem 68106MB [2022-12-19 08:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][350/1519] eta 0:19:32 lr 0.000033 time 0.9228 (1.0028) model_time 0.9225 (1.0012) loss 1.1798 (1.1137) grad_norm 7.7151 (9.5252/3.0267) mem 68106MB [2022-12-19 08:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][360/1519] eta 0:19:22 lr 0.000033 time 0.9225 (1.0029) model_time 0.9224 (1.0013) loss 1.1988 (1.1138) grad_norm 7.8705 (9.5062/3.0193) mem 68106MB [2022-12-19 08:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][370/1519] eta 0:19:11 lr 0.000033 time 0.9169 (1.0024) model_time 0.9168 (1.0008) loss 1.3312 (1.1124) grad_norm 8.8331 (9.5028/2.9966) mem 68106MB [2022-12-19 08:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][380/1519] eta 0:19:01 lr 0.000033 time 0.9176 (1.0023) model_time 0.9175 (1.0007) loss 1.0428 (1.1100) grad_norm 8.2780 (9.4938/2.9808) mem 68106MB [2022-12-19 08:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][390/1519] eta 0:18:51 lr 0.000033 time 0.9285 (1.0022) model_time 0.9284 (1.0007) loss 1.1556 (1.1100) grad_norm 9.5482 (9.5178/2.9727) mem 68106MB [2022-12-19 08:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][400/1519] eta 0:18:41 lr 0.000033 time 0.9217 (1.0021) model_time 0.9216 (1.0006) loss 1.3238 (1.1116) grad_norm 12.7171 (9.5798/3.0992) mem 68106MB [2022-12-19 08:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][410/1519] eta 0:18:31 lr 0.000033 time 0.9256 (1.0020) model_time 0.9255 (1.0006) loss 1.1831 (1.1128) grad_norm 7.2691 (9.6005/3.1072) mem 68106MB [2022-12-19 08:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][420/1519] eta 0:18:21 lr 0.000033 time 0.9233 (1.0022) model_time 0.9232 (1.0008) loss 0.9471 (1.1143) grad_norm 21.5529 (9.7766/3.3803) mem 68106MB [2022-12-19 08:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][430/1519] eta 0:18:11 lr 0.000033 time 0.9301 (1.0022) model_time 0.9299 (1.0008) loss 1.2138 (1.1154) grad_norm 13.2884 (9.7669/3.3728) mem 68106MB [2022-12-19 08:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][440/1519] eta 0:18:01 lr 0.000033 time 0.9376 (1.0025) model_time 0.9374 (1.0011) loss 1.0845 (1.1165) grad_norm 16.9349 (9.7898/3.3779) mem 68106MB [2022-12-19 08:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][450/1519] eta 0:17:51 lr 0.000033 time 0.9295 (1.0025) model_time 0.9293 (1.0011) loss 1.1117 (1.1181) grad_norm 5.7273 (9.8145/3.3745) mem 68106MB [2022-12-19 08:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][460/1519] eta 0:17:41 lr 0.000033 time 0.9308 (1.0024) model_time 0.9305 (1.0011) loss 1.0819 (1.1188) grad_norm 11.3451 (9.8365/3.3445) mem 68106MB [2022-12-19 08:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][470/1519] eta 0:17:31 lr 0.000033 time 0.9266 (1.0023) model_time 0.9265 (1.0010) loss 1.3161 (1.1185) grad_norm 10.7871 (9.7953/3.3299) mem 68106MB [2022-12-19 08:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][480/1519] eta 0:17:22 lr 0.000033 time 0.9096 (1.0030) model_time 0.9094 (1.0017) loss 1.0653 (1.1157) grad_norm 5.9489 (9.7813/3.3091) mem 68106MB [2022-12-19 08:13:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][490/1519] eta 0:17:12 lr 0.000033 time 0.9322 (1.0029) model_time 0.9320 (1.0016) loss 1.0251 (1.1148) grad_norm 9.6034 (9.7383/3.2975) mem 68106MB [2022-12-19 08:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][500/1519] eta 0:17:01 lr 0.000033 time 0.9294 (1.0029) model_time 0.9292 (1.0016) loss 0.9630 (1.1121) grad_norm 11.6331 (9.7467/3.3012) mem 68106MB [2022-12-19 08:14:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][510/1519] eta 0:16:51 lr 0.000033 time 0.9281 (1.0028) model_time 0.9280 (1.0015) loss 0.9010 (1.1131) grad_norm 10.5011 (9.7469/3.2753) mem 68106MB [2022-12-19 08:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][520/1519] eta 0:16:41 lr 0.000033 time 0.9350 (1.0028) model_time 0.9348 (1.0016) loss 0.9901 (1.1106) grad_norm 6.2686 (9.7195/3.2882) mem 68106MB [2022-12-19 08:14:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][530/1519] eta 0:16:31 lr 0.000033 time 0.9311 (1.0029) model_time 0.9310 (1.0017) loss 0.8252 (1.1088) grad_norm 6.7754 (9.6696/3.2785) mem 68106MB [2022-12-19 08:14:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][540/1519] eta 0:16:21 lr 0.000033 time 0.9316 (1.0028) model_time 0.9315 (1.0016) loss 1.0173 (1.1090) grad_norm 6.3506 (9.6204/3.2702) mem 68106MB [2022-12-19 08:14:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][550/1519] eta 0:16:11 lr 0.000033 time 0.9370 (1.0027) model_time 0.9368 (1.0015) loss 1.1201 (1.1089) grad_norm 8.3828 (9.6412/3.2759) mem 68106MB [2022-12-19 08:15:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][560/1519] eta 0:16:01 lr 0.000033 time 0.9241 (1.0026) model_time 0.9239 (1.0014) loss 1.0159 (1.1075) grad_norm 9.2537 (9.6621/3.2547) mem 68106MB [2022-12-19 08:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][570/1519] eta 0:15:51 lr 0.000033 time 0.9339 (1.0025) model_time 0.9338 (1.0013) loss 0.9396 (1.1083) grad_norm 7.6144 (9.6495/3.2424) mem 68106MB [2022-12-19 08:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][580/1519] eta 0:15:41 lr 0.000033 time 0.9271 (1.0024) model_time 0.9270 (1.0012) loss 0.8762 (1.1072) grad_norm 12.4202 (9.6552/3.2257) mem 68106MB [2022-12-19 08:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][590/1519] eta 0:15:31 lr 0.000033 time 0.9282 (1.0023) model_time 0.9280 (1.0012) loss 1.0248 (1.1057) grad_norm 14.5058 (9.6607/3.2189) mem 68106MB [2022-12-19 08:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][600/1519] eta 0:15:21 lr 0.000033 time 0.9215 (1.0023) model_time 0.9212 (1.0011) loss 1.1575 (1.1052) grad_norm 20.3809 (9.6970/3.2755) mem 68106MB [2022-12-19 08:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][610/1519] eta 0:15:11 lr 0.000033 time 0.9299 (1.0022) model_time 0.9297 (1.0011) loss 1.1877 (1.1049) grad_norm 11.4726 (9.6749/3.2599) mem 68106MB [2022-12-19 08:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][620/1519] eta 0:15:00 lr 0.000033 time 0.9289 (1.0022) model_time 0.9287 (1.0011) loss 0.9075 (1.1059) grad_norm 8.4791 (9.6644/3.2724) mem 68106MB [2022-12-19 08:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][630/1519] eta 0:14:50 lr 0.000033 time 0.9288 (1.0022) model_time 0.9286 (1.0011) loss 0.7867 (1.1062) grad_norm 11.5710 (9.6869/3.2706) mem 68106MB [2022-12-19 08:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][640/1519] eta 0:14:40 lr 0.000033 time 0.9314 (1.0021) model_time 0.9312 (1.0010) loss 1.1549 (1.1080) grad_norm 11.3093 (9.7226/3.2715) mem 68106MB [2022-12-19 08:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][650/1519] eta 0:14:30 lr 0.000033 time 0.9308 (1.0020) model_time 0.9306 (1.0010) loss 0.9152 (1.1093) grad_norm 7.5351 (9.7278/3.3351) mem 68106MB [2022-12-19 08:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][660/1519] eta 0:14:20 lr 0.000033 time 0.9256 (1.0020) model_time 0.9255 (1.0009) loss 1.0187 (1.1093) grad_norm 8.3546 (9.7428/3.3299) mem 68106MB [2022-12-19 08:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][670/1519] eta 0:14:10 lr 0.000033 time 0.9311 (1.0019) model_time 0.9309 (1.0008) loss 0.8660 (1.1100) grad_norm 14.1474 (9.7774/3.3459) mem 68106MB [2022-12-19 08:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][680/1519] eta 0:14:00 lr 0.000033 time 0.9300 (1.0018) model_time 0.9299 (1.0008) loss 1.1423 (1.1088) grad_norm 9.9321 (9.7871/3.3685) mem 68106MB [2022-12-19 08:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][690/1519] eta 0:13:50 lr 0.000033 time 0.9320 (1.0018) model_time 0.9319 (1.0007) loss 1.0301 (1.1092) grad_norm 7.1450 (9.7740/3.3664) mem 68106MB [2022-12-19 08:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][700/1519] eta 0:13:40 lr 0.000033 time 0.9273 (1.0017) model_time 0.9271 (1.0007) loss 1.4569 (1.1093) grad_norm 7.4852 (9.7490/3.3729) mem 68106MB [2022-12-19 08:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][710/1519] eta 0:13:30 lr 0.000033 time 0.9316 (1.0016) model_time 0.9314 (1.0006) loss 1.1997 (1.1100) grad_norm 15.6779 (9.7989/3.4011) mem 68106MB [2022-12-19 08:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][720/1519] eta 0:13:20 lr 0.000033 time 0.9298 (1.0016) model_time 0.9296 (1.0006) loss 1.2160 (1.1110) grad_norm 18.9371 (9.8227/3.4326) mem 68106MB [2022-12-19 08:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][730/1519] eta 0:13:10 lr 0.000033 time 0.9226 (1.0017) model_time 0.9225 (1.0007) loss 1.0804 (1.1109) grad_norm 9.4559 (9.8396/3.4217) mem 68106MB [2022-12-19 08:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][740/1519] eta 0:13:00 lr 0.000033 time 0.9328 (1.0017) model_time 0.9327 (1.0007) loss 1.1556 (1.1107) grad_norm 8.2878 (9.7713/3.3968) mem 68106MB [2022-12-19 08:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][750/1519] eta 0:12:50 lr 0.000033 time 0.9311 (1.0018) model_time 0.9310 (1.0008) loss 1.0633 (1.1112) grad_norm 7.8382 (9.7401/3.3950) mem 68106MB [2022-12-19 08:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][760/1519] eta 0:12:40 lr 0.000033 time 0.9292 (1.0017) model_time 0.9291 (1.0008) loss 0.9127 (1.1104) grad_norm 7.8085 (9.6944/3.3744) mem 68106MB [2022-12-19 08:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][770/1519] eta 0:12:30 lr 0.000033 time 0.9348 (1.0018) model_time 0.9347 (1.0008) loss 0.9989 (1.1098) grad_norm 6.9902 (9.6864/3.3785) mem 68106MB [2022-12-19 08:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][780/1519] eta 0:12:20 lr 0.000033 time 0.9299 (1.0017) model_time 0.9297 (1.0008) loss 1.2508 (1.1108) grad_norm 10.4007 (9.6611/3.3712) mem 68106MB [2022-12-19 08:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][790/1519] eta 0:12:10 lr 0.000033 time 0.9275 (1.0018) model_time 0.9273 (1.0009) loss 0.9899 (1.1102) grad_norm 9.1607 (9.6672/3.3682) mem 68106MB [2022-12-19 08:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][800/1519] eta 0:12:00 lr 0.000033 time 0.9292 (1.0019) model_time 0.9291 (1.0010) loss 0.9667 (1.1102) grad_norm 13.5593 (9.5965/3.2710) mem 68106MB [2022-12-19 08:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][810/1519] eta 0:11:50 lr 0.000033 time 0.9284 (1.0019) model_time 0.9283 (1.0010) loss 0.7746 (1.1086) grad_norm 5.7264 (9.5843/3.2625) mem 68106MB [2022-12-19 08:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][820/1519] eta 0:11:40 lr 0.000033 time 0.9309 (1.0019) model_time 0.9307 (1.0009) loss 0.8434 (1.1078) grad_norm 7.2409 (9.5334/3.2267) mem 68106MB [2022-12-19 08:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][830/1519] eta 0:11:30 lr 0.000033 time 0.9320 (1.0019) model_time 0.9319 (1.0010) loss 0.9511 (1.1083) grad_norm 9.4243 (9.4916/3.1528) mem 68106MB [2022-12-19 08:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][840/1519] eta 0:11:20 lr 0.000033 time 0.9318 (1.0019) model_time 0.9317 (1.0010) loss 0.8689 (1.1075) grad_norm 6.9815 (9.4454/3.1549) mem 68106MB [2022-12-19 08:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][850/1519] eta 0:11:10 lr 0.000033 time 0.9307 (1.0020) model_time 0.9306 (1.0011) loss 1.3486 (1.1067) grad_norm 5.5644 (9.4418/3.1550) mem 68106MB [2022-12-19 08:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][860/1519] eta 0:11:00 lr 0.000033 time 0.9272 (1.0019) model_time 0.9271 (1.0010) loss 1.1949 (1.1069) grad_norm 11.5898 (9.4404/3.1404) mem 68106MB [2022-12-19 08:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][870/1519] eta 0:10:50 lr 0.000033 time 0.9323 (1.0019) model_time 0.9322 (1.0010) loss 0.8893 (1.1083) grad_norm 18.0293 (9.4624/3.2004) mem 68106MB [2022-12-19 08:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][880/1519] eta 0:10:40 lr 0.000033 time 0.9302 (1.0018) model_time 0.9301 (1.0009) loss 1.2050 (1.1090) grad_norm 14.0885 (9.4714/3.2109) mem 68106MB [2022-12-19 08:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][890/1519] eta 0:10:30 lr 0.000033 time 0.9317 (1.0018) model_time 0.9316 (1.0009) loss 0.9585 (1.1070) grad_norm 14.6186 (9.5673/3.2590) mem 68106MB [2022-12-19 08:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][900/1519] eta 0:10:20 lr 0.000033 time 0.9326 (1.0017) model_time 0.9324 (1.0009) loss 1.1128 (1.1060) grad_norm 7.7258 (9.6048/3.2617) mem 68106MB [2022-12-19 08:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][910/1519] eta 0:10:10 lr 0.000033 time 0.9276 (1.0018) model_time 0.9274 (1.0009) loss 1.4663 (1.1053) grad_norm 8.0434 (9.5743/3.2678) mem 68106MB [2022-12-19 08:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][920/1519] eta 0:10:00 lr 0.000033 time 0.9273 (1.0018) model_time 0.9271 (1.0009) loss 0.9090 (1.1048) grad_norm 7.2672 (9.5980/3.2864) mem 68106MB [2022-12-19 08:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][930/1519] eta 0:09:50 lr 0.000033 time 0.9297 (1.0018) model_time 0.9295 (1.0009) loss 0.9490 (1.1046) grad_norm 6.8704 (9.5925/3.2830) mem 68106MB [2022-12-19 08:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][940/1519] eta 0:09:40 lr 0.000033 time 0.9385 (1.0018) model_time 0.9384 (1.0009) loss 0.9821 (1.1051) grad_norm 9.0921 (9.5756/3.2883) mem 68106MB [2022-12-19 08:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][950/1519] eta 0:09:30 lr 0.000033 time 0.9320 (1.0018) model_time 0.9319 (1.0009) loss 1.2541 (1.1056) grad_norm 6.0410 (9.5915/3.2945) mem 68106MB [2022-12-19 08:21:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][960/1519] eta 0:09:20 lr 0.000033 time 0.9292 (1.0018) model_time 0.9291 (1.0010) loss 1.2994 (1.1061) grad_norm 9.7306 (9.5951/3.2769) mem 68106MB [2022-12-19 08:21:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][970/1519] eta 0:09:09 lr 0.000033 time 0.9315 (1.0018) model_time 0.9314 (1.0009) loss 0.9193 (1.1061) grad_norm 23.8868 (9.7501/3.4927) mem 68106MB [2022-12-19 08:22:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][980/1519] eta 0:08:59 lr 0.000033 time 0.9406 (1.0018) model_time 0.9405 (1.0009) loss 0.9347 (1.1070) grad_norm 11.0399 (9.7693/3.4897) mem 68106MB [2022-12-19 08:22:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][990/1519] eta 0:08:49 lr 0.000033 time 0.9308 (1.0017) model_time 0.9306 (1.0009) loss 1.0482 (1.1066) grad_norm 8.3521 (9.7326/3.4851) mem 68106MB [2022-12-19 08:22:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1000/1519] eta 0:08:39 lr 0.000033 time 0.9266 (1.0017) model_time 0.9264 (1.0009) loss 0.9251 (1.1065) grad_norm 6.2578 (9.6835/3.4123) mem 68106MB [2022-12-19 08:22:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1010/1519] eta 0:08:29 lr 0.000033 time 0.9315 (1.0018) model_time 0.9314 (1.0009) loss 0.8737 (1.1072) grad_norm 15.9207 (9.7567/3.5981) mem 68106MB [2022-12-19 08:22:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1020/1519] eta 0:08:19 lr 0.000033 time 0.9393 (1.0018) model_time 0.9392 (1.0010) loss 1.1403 (1.1068) grad_norm 13.8951 (9.6853/3.4428) mem 68106MB [2022-12-19 08:22:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1030/1519] eta 0:08:09 lr 0.000033 time 0.9306 (1.0018) model_time 0.9304 (1.0010) loss 1.1643 (1.1067) grad_norm 9.0950 (9.7227/3.4583) mem 68106MB [2022-12-19 08:23:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1040/1519] eta 0:07:59 lr 0.000033 time 0.9311 (1.0018) model_time 0.9310 (1.0010) loss 0.9971 (1.1066) grad_norm 11.9535 (9.7493/3.5109) mem 68106MB [2022-12-19 08:23:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1050/1519] eta 0:07:49 lr 0.000033 time 0.9550 (1.0019) model_time 0.9548 (1.0011) loss 0.9802 (1.1066) grad_norm 4.8053 (9.6892/3.5091) mem 68106MB [2022-12-19 08:23:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1060/1519] eta 0:07:39 lr 0.000033 time 0.9716 (1.0021) model_time 0.9715 (1.0013) loss 1.4377 (1.1069) grad_norm 8.1450 (9.6338/3.5171) mem 68106MB [2022-12-19 08:23:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1070/1519] eta 0:07:29 lr 0.000033 time 0.9292 (1.0021) model_time 0.9291 (1.0013) loss 1.1302 (1.1074) grad_norm 5.3572 (9.6341/3.5150) mem 68106MB [2022-12-19 08:23:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1080/1519] eta 0:07:19 lr 0.000033 time 0.9467 (1.0021) model_time 0.9466 (1.0014) loss 1.0142 (1.1080) grad_norm 5.8283 (9.5967/3.5274) mem 68106MB [2022-12-19 08:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1090/1519] eta 0:07:09 lr 0.000033 time 0.9337 (1.0021) model_time 0.9336 (1.0013) loss 1.0267 (1.1078) grad_norm 7.2165 (9.6248/3.5283) mem 68106MB [2022-12-19 08:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1100/1519] eta 0:06:59 lr 0.000033 time 0.9415 (1.0023) model_time 0.9414 (1.0015) loss 1.0616 (1.1077) grad_norm 12.2999 (9.6105/3.5201) mem 68106MB [2022-12-19 08:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1110/1519] eta 0:06:50 lr 0.000033 time 0.9248 (1.0026) model_time 0.9247 (1.0018) loss 1.2797 (1.1079) grad_norm 13.5582 (9.5758/3.5466) mem 68106MB [2022-12-19 08:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1120/1519] eta 0:06:40 lr 0.000033 time 0.9558 (1.0026) model_time 0.9556 (1.0018) loss 1.1064 (1.1086) grad_norm 7.0424 (9.5817/3.5213) mem 68106MB [2022-12-19 08:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1130/1519] eta 0:06:29 lr 0.000033 time 0.9300 (1.0025) model_time 0.9299 (1.0018) loss 1.1832 (1.1085) grad_norm 7.3098 (9.6071/3.5300) mem 68106MB [2022-12-19 08:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1140/1519] eta 0:06:19 lr 0.000033 time 0.9331 (1.0025) model_time 0.9329 (1.0018) loss 1.0974 (1.1078) grad_norm 10.4701 (9.6385/3.5256) mem 68106MB [2022-12-19 08:24:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1150/1519] eta 0:06:09 lr 0.000033 time 0.9537 (1.0026) model_time 0.9536 (1.0018) loss 1.1806 (1.1078) grad_norm 7.7473 (9.6181/3.5063) mem 68106MB [2022-12-19 08:25:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1160/1519] eta 0:05:59 lr 0.000033 time 0.9319 (1.0026) model_time 0.9318 (1.0018) loss 0.8991 (1.1074) grad_norm 11.4155 (9.6003/3.5137) mem 68106MB [2022-12-19 08:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1170/1519] eta 0:05:49 lr 0.000033 time 0.9414 (1.0026) model_time 0.9413 (1.0018) loss 1.0207 (1.1064) grad_norm 9.9843 (9.5939/3.5050) mem 68106MB [2022-12-19 08:25:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1180/1519] eta 0:05:39 lr 0.000033 time 0.9375 (1.0026) model_time 0.9374 (1.0018) loss 0.7569 (1.1062) grad_norm 9.8980 (9.6234/3.5141) mem 68106MB [2022-12-19 08:25:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1190/1519] eta 0:05:29 lr 0.000033 time 0.9290 (1.0025) model_time 0.9289 (1.0018) loss 0.9072 (1.1069) grad_norm 7.8809 (9.6144/3.4992) mem 68106MB [2022-12-19 08:25:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1200/1519] eta 0:05:19 lr 0.000033 time 0.9261 (1.0025) model_time 0.9260 (1.0017) loss 1.4981 (1.1063) grad_norm 7.8046 (9.5935/3.4361) mem 68106MB [2022-12-19 08:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1210/1519] eta 0:05:09 lr 0.000033 time 0.9435 (1.0025) model_time 0.9434 (1.0018) loss 1.1330 (1.1063) grad_norm 7.4939 (9.6034/3.4483) mem 68106MB [2022-12-19 08:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1220/1519] eta 0:04:59 lr 0.000033 time 0.9370 (1.0026) model_time 0.9368 (1.0018) loss 1.3750 (1.1061) grad_norm 11.8982 (9.6092/3.4446) mem 68106MB [2022-12-19 08:26:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1230/1519] eta 0:04:49 lr 0.000033 time 0.9310 (1.0026) model_time 0.9309 (1.0018) loss 1.2771 (1.1058) grad_norm 6.7393 (9.6200/3.4526) mem 68106MB [2022-12-19 08:26:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1240/1519] eta 0:04:39 lr 0.000033 time 0.9312 (1.0026) model_time 0.9310 (1.0019) loss 0.9908 (1.1063) grad_norm 8.0966 (9.5806/3.4739) mem 68106MB [2022-12-19 08:26:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1250/1519] eta 0:04:29 lr 0.000033 time 0.9349 (1.0026) model_time 0.9348 (1.0019) loss 0.9965 (1.1064) grad_norm 7.2577 (9.5299/3.4127) mem 68106MB [2022-12-19 08:26:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1260/1519] eta 0:04:19 lr 0.000033 time 0.9332 (1.0026) model_time 0.9331 (1.0019) loss 1.1782 (1.1069) grad_norm 15.1789 (9.5606/3.4184) mem 68106MB [2022-12-19 08:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9281 (1.0026) model_time 0.9279 (1.0019) loss 1.2666 (1.1072) grad_norm 11.0126 (9.5000/3.3738) mem 68106MB [2022-12-19 08:27:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9345 (1.0026) model_time 0.9344 (1.0018) loss 0.7405 (1.1070) grad_norm 12.7242 (9.5216/3.3966) mem 68106MB [2022-12-19 08:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9404 (1.0027) model_time 0.9403 (1.0019) loss 1.5994 (1.1074) grad_norm 8.1745 (9.5379/3.4057) mem 68106MB [2022-12-19 08:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9395 (1.0026) model_time 0.9394 (1.0019) loss 1.3737 (1.1076) grad_norm 10.4040 (9.5863/3.4201) mem 68106MB [2022-12-19 08:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9399 (1.0026) model_time 0.9397 (1.0019) loss 1.0420 (1.1078) grad_norm 12.6200 (9.5957/3.4782) mem 68106MB [2022-12-19 08:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9164 (1.0027) model_time 0.9163 (1.0020) loss 1.2067 (1.1076) grad_norm 6.8042 (9.5506/3.4382) mem 68106MB [2022-12-19 08:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9339 (1.0028) model_time 0.9338 (1.0020) loss 1.5245 (1.1082) grad_norm 9.0547 (9.5701/3.4348) mem 68106MB [2022-12-19 08:28:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9290 (1.0027) model_time 0.9289 (1.0020) loss 0.8957 (1.1078) grad_norm 11.4575 (9.5853/3.4370) mem 68106MB [2022-12-19 08:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9326 (1.0027) model_time 0.9325 (1.0020) loss 1.0302 (1.1074) grad_norm 5.1144 (9.5894/3.4394) mem 68106MB [2022-12-19 08:28:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9306 (1.0027) model_time 0.9305 (1.0020) loss 1.4650 (1.1084) grad_norm 6.0775 (9.5770/3.4589) mem 68106MB [2022-12-19 08:28:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9402 (1.0027) model_time 0.9400 (1.0020) loss 1.1963 (1.1087) grad_norm 11.0714 (9.5948/3.4575) mem 68106MB [2022-12-19 08:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9347 (1.0027) model_time 0.9345 (1.0019) loss 1.3808 (1.1089) grad_norm 6.9439 (9.5946/3.4457) mem 68106MB [2022-12-19 08:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9350 (1.0027) model_time 0.9348 (1.0019) loss 1.1498 (1.1083) grad_norm 6.5028 (9.5906/3.4468) mem 68106MB [2022-12-19 08:29:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9333 (1.0026) model_time 0.9332 (1.0019) loss 1.7089 (1.1087) grad_norm 6.8476 (9.5592/3.4359) mem 68106MB [2022-12-19 08:29:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1410/1519] eta 0:01:49 lr 0.000033 time 1.1265 (1.0027) model_time 1.1264 (1.0020) loss 0.9296 (1.1088) grad_norm 9.5738 (9.5748/3.4264) mem 68106MB [2022-12-19 08:29:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9251 (1.0027) model_time 0.9249 (1.0020) loss 1.0157 (1.1083) grad_norm 6.9605 (9.5767/3.4429) mem 68106MB [2022-12-19 08:29:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9326 (1.0027) model_time 0.9325 (1.0020) loss 0.8253 (1.1075) grad_norm 13.0795 (9.5782/3.4340) mem 68106MB [2022-12-19 08:29:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9305 (1.0027) model_time 0.9304 (1.0020) loss 1.0818 (1.1077) grad_norm 13.8955 (9.6643/3.4856) mem 68106MB [2022-12-19 08:29:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9369 (1.0027) model_time 0.9368 (1.0020) loss 1.1014 (1.1075) grad_norm 13.3578 (9.7034/3.4998) mem 68106MB [2022-12-19 08:30:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9303 (1.0027) model_time 0.9302 (1.0020) loss 1.1493 (1.1070) grad_norm 7.7398 (9.6910/3.5042) mem 68106MB [2022-12-19 08:30:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9388 (1.0026) model_time 0.9387 (1.0020) loss 1.2657 (1.1069) grad_norm 11.5905 (9.6863/3.4662) mem 68106MB [2022-12-19 08:30:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9280 (1.0026) model_time 0.9278 (1.0019) loss 0.7092 (1.1066) grad_norm 9.7791 (9.6950/3.4537) mem 68106MB [2022-12-19 08:30:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9313 (1.0026) model_time 0.9312 (1.0019) loss 0.9999 (1.1063) grad_norm 11.8203 (9.6396/3.4119) mem 68106MB [2022-12-19 08:30:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9363 (1.0026) model_time 0.9362 (1.0019) loss 0.8913 (1.1067) grad_norm 8.7571 (9.6364/3.4017) mem 68106MB [2022-12-19 08:30:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [9/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9332 (1.0025) model_time 0.9331 (1.0019) loss 0.9355 (1.1066) grad_norm 9.3769 (9.7040/3.5038) mem 68106MB [2022-12-19 08:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 9 training takes 0:25:22 [2022-12-19 08:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_9.pth saving...... [2022-12-19 08:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_9.pth saved !!! [2022-12-19 08:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.676 (0.676) Loss 0.8747 (0.8747) Acc@1 81.250 (81.250) Acc@5 96.528 (96.528) Mem 68106MB [2022-12-19 08:31:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.300 (0.331) Loss 0.8898 (0.8757) Acc@1 83.681 (82.670) Acc@5 96.181 (96.559) Mem 68106MB [2022-12-19 08:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.315) Loss 0.8218 (0.8708) Acc@1 84.722 (82.474) Acc@5 97.222 (96.544) Mem 68106MB [2022-12-19 08:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.9315 (0.8731) Acc@1 81.944 (82.460) Acc@5 95.486 (96.371) Mem 68106MB [2022-12-19 08:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.307) Loss 0.7991 (0.8612) Acc@1 84.028 (82.520) Acc@5 95.833 (96.485) Mem 68106MB [2022-12-19 08:31:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.8755 (0.8564) Acc@1 79.167 (82.530) Acc@5 95.833 (96.555) Mem 68106MB [2022-12-19 08:31:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.305) Loss 0.8958 (0.8570) Acc@1 82.639 (82.599) Acc@5 95.139 (96.562) Mem 68106MB [2022-12-19 08:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.304 (0.304) Loss 0.9707 (0.8605) Acc@1 81.597 (82.600) Acc@5 94.792 (96.528) Mem 68106MB [2022-12-19 08:31:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.8165 (0.8605) Acc@1 84.028 (82.587) Acc@5 97.222 (96.562) Mem 68106MB [2022-12-19 08:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:9] * Acc@1 82.633 Acc@5 96.566 [2022-12-19 08:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 82.6% [2022-12-19 08:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 08:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 08:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 82.63% [2022-12-19 08:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][0/1519] eta 0:34:56 lr 0.000033 time 1.3803 (1.3803) model_time 0.9646 (0.9646) loss 0.9223 (0.9223) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 08:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][10/1519] eta 0:26:04 lr 0.000033 time 0.9245 (1.0366) model_time 0.9244 (0.9985) loss 1.2911 (1.0825) grad_norm 4.3216 (8.2957/2.2759) mem 68106MB [2022-12-19 08:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][20/1519] eta 0:25:36 lr 0.000033 time 0.9441 (1.0248) model_time 0.9439 (1.0047) loss 0.8711 (1.0565) grad_norm 8.2777 (8.9955/2.0299) mem 68106MB [2022-12-19 08:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][30/1519] eta 0:25:22 lr 0.000033 time 1.0419 (1.0223) model_time 1.0418 (1.0086) loss 1.2111 (1.0974) grad_norm 7.7725 (9.2265/1.8224) mem 68106MB [2022-12-19 08:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][40/1519] eta 0:25:04 lr 0.000033 time 0.9245 (1.0175) model_time 0.9243 (1.0070) loss 0.8226 (1.0962) grad_norm 9.2252 (9.0805/1.7352) mem 68106MB [2022-12-19 08:33:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][50/1519] eta 0:24:58 lr 0.000033 time 0.9411 (1.0198) model_time 0.9410 (1.0112) loss 0.9826 (1.0691) grad_norm 8.3354 (8.9670/1.7744) mem 68106MB [2022-12-19 08:33:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][60/1519] eta 0:24:42 lr 0.000033 time 0.9245 (1.0161) model_time 0.9244 (1.0090) loss 1.1846 (1.0868) grad_norm 8.3058 (10.2358/5.8446) mem 68106MB [2022-12-19 08:33:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][70/1519] eta 0:24:31 lr 0.000033 time 0.9327 (1.0152) model_time 0.9325 (1.0085) loss 0.7747 (1.0893) grad_norm 8.0302 (10.1246/5.4349) mem 68106MB [2022-12-19 08:33:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][80/1519] eta 0:24:21 lr 0.000033 time 0.9260 (1.0159) model_time 0.9258 (1.0099) loss 0.9554 (1.0884) grad_norm 12.9222 (9.9719/5.1815) mem 68106MB [2022-12-19 08:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][90/1519] eta 0:24:09 lr 0.000033 time 0.9309 (1.0146) model_time 0.9307 (1.0092) loss 1.2654 (1.0958) grad_norm 14.2264 (10.2653/5.1632) mem 68106MB [2022-12-19 08:33:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][100/1519] eta 0:23:58 lr 0.000033 time 0.9382 (1.0134) model_time 0.9381 (1.0085) loss 1.0599 (1.1075) grad_norm 5.0486 (10.0109/4.9945) mem 68106MB [2022-12-19 08:34:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][110/1519] eta 0:23:46 lr 0.000033 time 0.9258 (1.0124) model_time 0.9256 (1.0079) loss 1.0285 (1.0953) grad_norm 6.6085 (9.8938/4.8707) mem 68106MB [2022-12-19 08:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][120/1519] eta 0:23:39 lr 0.000033 time 1.1655 (1.0148) model_time 1.1653 (1.0107) loss 1.1249 (1.0985) grad_norm 7.3900 (9.7201/4.7101) mem 68106MB [2022-12-19 08:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][130/1519] eta 0:23:28 lr 0.000033 time 0.9221 (1.0139) model_time 0.9220 (1.0100) loss 1.3964 (1.1014) grad_norm 9.2398 (9.6644/4.6014) mem 68106MB [2022-12-19 08:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][140/1519] eta 0:23:16 lr 0.000033 time 0.9347 (1.0130) model_time 0.9343 (1.0094) loss 0.7766 (1.1075) grad_norm 6.2577 (9.7632/4.5782) mem 68106MB [2022-12-19 08:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][150/1519] eta 0:23:05 lr 0.000033 time 0.9227 (1.0120) model_time 0.9226 (1.0086) loss 1.4528 (1.1050) grad_norm 11.1792 (9.6521/4.4677) mem 68106MB [2022-12-19 08:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][160/1519] eta 0:22:54 lr 0.000033 time 0.9321 (1.0113) model_time 0.9320 (1.0080) loss 1.1591 (1.1050) grad_norm 11.3175 (9.5206/4.3824) mem 68106MB [2022-12-19 08:35:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][170/1519] eta 0:22:43 lr 0.000033 time 0.9204 (1.0105) model_time 0.9202 (1.0074) loss 0.8164 (1.1023) grad_norm 12.5623 (9.5652/4.2647) mem 68106MB [2022-12-19 08:35:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][180/1519] eta 0:22:32 lr 0.000033 time 0.9202 (1.0098) model_time 0.9199 (1.0069) loss 0.8416 (1.1010) grad_norm 8.6049 (9.5083/4.1722) mem 68106MB [2022-12-19 08:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][190/1519] eta 0:22:21 lr 0.000033 time 0.9299 (1.0094) model_time 0.9298 (1.0066) loss 1.3016 (1.0938) grad_norm 7.3066 (9.4175/4.0802) mem 68106MB [2022-12-19 08:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][200/1519] eta 0:22:12 lr 0.000033 time 0.9379 (1.0103) model_time 0.9377 (1.0077) loss 1.1185 (1.0887) grad_norm 8.0023 (9.4077/4.0473) mem 68106MB [2022-12-19 08:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][210/1519] eta 0:22:02 lr 0.000033 time 1.0329 (1.0104) model_time 1.0328 (1.0078) loss 1.0285 (1.0888) grad_norm 6.0608 (9.3054/3.9845) mem 68106MB [2022-12-19 08:35:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][220/1519] eta 0:21:51 lr 0.000033 time 0.9279 (1.0098) model_time 0.9277 (1.0074) loss 0.9990 (1.0890) grad_norm 7.6237 (9.2373/3.9162) mem 68106MB [2022-12-19 08:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][230/1519] eta 0:21:41 lr 0.000033 time 0.9226 (1.0097) model_time 0.9225 (1.0074) loss 0.9911 (1.0937) grad_norm 17.6481 (9.3197/3.9261) mem 68106MB [2022-12-19 08:36:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][240/1519] eta 0:21:31 lr 0.000033 time 0.9227 (1.0094) model_time 0.9225 (1.0071) loss 1.2011 (1.0923) grad_norm 6.1176 (9.3735/3.9121) mem 68106MB [2022-12-19 08:36:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][250/1519] eta 0:21:20 lr 0.000033 time 0.9311 (1.0090) model_time 0.9309 (1.0068) loss 0.9322 (1.0927) grad_norm 10.3912 (9.4534/4.0291) mem 68106MB [2022-12-19 08:36:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][260/1519] eta 0:21:11 lr 0.000033 time 0.9312 (1.0098) model_time 0.9311 (1.0077) loss 1.4086 (1.0920) grad_norm 9.0226 (9.4483/3.9636) mem 68106MB [2022-12-19 08:36:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][270/1519] eta 0:21:00 lr 0.000033 time 0.9199 (1.0094) model_time 0.9197 (1.0073) loss 1.0921 (1.0911) grad_norm 8.3744 (9.3640/3.9210) mem 68106MB [2022-12-19 08:36:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][280/1519] eta 0:20:50 lr 0.000033 time 0.9306 (1.0090) model_time 0.9304 (1.0071) loss 1.1080 (1.0894) grad_norm 7.6785 (9.3213/3.8590) mem 68106MB [2022-12-19 08:37:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][290/1519] eta 0:20:39 lr 0.000033 time 0.9272 (1.0087) model_time 0.9271 (1.0068) loss 0.8477 (1.0900) grad_norm 5.9718 (9.2453/3.8200) mem 68106MB [2022-12-19 08:37:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][300/1519] eta 0:20:29 lr 0.000033 time 1.0047 (1.0088) model_time 1.0046 (1.0069) loss 0.8229 (1.0866) grad_norm 14.1562 (9.2521/3.7875) mem 68106MB [2022-12-19 08:37:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][310/1519] eta 0:20:19 lr 0.000033 time 0.9285 (1.0085) model_time 0.9284 (1.0067) loss 1.0070 (1.0835) grad_norm 7.6767 (9.2946/3.7570) mem 68106MB [2022-12-19 08:37:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][320/1519] eta 0:20:08 lr 0.000033 time 0.9267 (1.0083) model_time 0.9266 (1.0065) loss 0.9527 (1.0822) grad_norm 7.0006 (9.3049/3.7254) mem 68106MB [2022-12-19 08:37:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][330/1519] eta 0:19:58 lr 0.000033 time 0.9195 (1.0080) model_time 0.9192 (1.0063) loss 0.7312 (1.0801) grad_norm 5.6548 (9.2665/3.6886) mem 68106MB [2022-12-19 08:37:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][340/1519] eta 0:19:48 lr 0.000033 time 0.9274 (1.0078) model_time 0.9272 (1.0061) loss 1.1215 (1.0802) grad_norm 9.2040 (9.3083/3.7360) mem 68106MB [2022-12-19 08:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][350/1519] eta 0:19:37 lr 0.000033 time 0.9339 (1.0076) model_time 0.9337 (1.0060) loss 1.3295 (1.0816) grad_norm 8.9969 (9.2890/3.6926) mem 68106MB [2022-12-19 08:38:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][360/1519] eta 0:19:27 lr 0.000033 time 0.9370 (1.0074) model_time 0.9368 (1.0058) loss 0.8934 (1.0798) grad_norm 9.3583 (9.2965/3.6435) mem 68106MB [2022-12-19 08:38:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][370/1519] eta 0:19:17 lr 0.000033 time 0.9302 (1.0072) model_time 0.9300 (1.0057) loss 0.8326 (1.0800) grad_norm 8.3630 (9.2238/3.6237) mem 68106MB [2022-12-19 08:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][380/1519] eta 0:19:07 lr 0.000033 time 0.9276 (1.0073) model_time 0.9275 (1.0058) loss 1.2411 (1.0789) grad_norm 10.7810 (9.2252/3.5902) mem 68106MB [2022-12-19 08:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][390/1519] eta 0:18:57 lr 0.000033 time 0.9764 (1.0073) model_time 0.9762 (1.0058) loss 1.1222 (1.0789) grad_norm 6.7018 (9.2951/3.6096) mem 68106MB [2022-12-19 08:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][400/1519] eta 0:18:46 lr 0.000033 time 0.9296 (1.0071) model_time 0.9294 (1.0056) loss 0.9422 (1.0795) grad_norm 8.6341 (9.3033/3.5935) mem 68106MB [2022-12-19 08:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][410/1519] eta 0:18:36 lr 0.000033 time 0.9239 (1.0070) model_time 0.9238 (1.0055) loss 0.8603 (1.0788) grad_norm 5.5912 (9.3751/3.6368) mem 68106MB [2022-12-19 08:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][420/1519] eta 0:18:26 lr 0.000033 time 0.9397 (1.0069) model_time 0.9396 (1.0054) loss 0.9274 (1.0758) grad_norm 6.3132 (9.3525/3.6015) mem 68106MB [2022-12-19 08:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][430/1519] eta 0:18:16 lr 0.000033 time 0.9298 (1.0067) model_time 0.9296 (1.0053) loss 0.9211 (1.0734) grad_norm 6.8796 (9.2844/3.5900) mem 68106MB [2022-12-19 08:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][440/1519] eta 0:18:06 lr 0.000033 time 0.9309 (1.0069) model_time 0.9307 (1.0055) loss 0.9269 (1.0729) grad_norm 8.2487 (9.3316/3.6988) mem 68106MB [2022-12-19 08:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][450/1519] eta 0:17:56 lr 0.000033 time 0.9231 (1.0067) model_time 0.9230 (1.0053) loss 1.1615 (1.0719) grad_norm 11.6121 (9.3437/3.6906) mem 68106MB [2022-12-19 08:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][460/1519] eta 0:17:46 lr 0.000033 time 0.9398 (1.0066) model_time 0.9396 (1.0053) loss 0.9194 (1.0720) grad_norm 8.0488 (9.3189/3.6548) mem 68106MB [2022-12-19 08:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][470/1519] eta 0:17:35 lr 0.000033 time 0.9374 (1.0065) model_time 0.9372 (1.0052) loss 1.6745 (1.0740) grad_norm 5.3372 (9.2939/3.6313) mem 68106MB [2022-12-19 08:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][480/1519] eta 0:17:25 lr 0.000033 time 0.9403 (1.0065) model_time 0.9401 (1.0052) loss 1.0166 (1.0729) grad_norm 9.6667 (9.2696/3.6054) mem 68106MB [2022-12-19 08:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][490/1519] eta 0:17:15 lr 0.000033 time 0.9314 (1.0063) model_time 0.9311 (1.0050) loss 1.0145 (1.0730) grad_norm 6.8459 (9.2904/3.6368) mem 68106MB [2022-12-19 08:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][500/1519] eta 0:17:05 lr 0.000033 time 0.9306 (1.0062) model_time 0.9304 (1.0049) loss 1.0213 (1.0739) grad_norm 7.3316 (9.2949/3.6099) mem 68106MB [2022-12-19 08:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][510/1519] eta 0:16:55 lr 0.000033 time 0.9292 (1.0063) model_time 0.9290 (1.0050) loss 1.1039 (1.0750) grad_norm 9.3469 (9.3062/3.5970) mem 68106MB [2022-12-19 08:40:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][520/1519] eta 0:16:45 lr 0.000033 time 0.9224 (1.0062) model_time 0.9222 (1.0049) loss 0.8136 (1.0736) grad_norm 10.1444 (9.2789/3.5736) mem 68106MB [2022-12-19 08:41:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][530/1519] eta 0:16:35 lr 0.000033 time 0.9310 (1.0063) model_time 0.9308 (1.0051) loss 1.3535 (1.0743) grad_norm 8.5733 (9.2723/3.5420) mem 68106MB [2022-12-19 08:41:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][540/1519] eta 0:16:25 lr 0.000033 time 0.8984 (1.0067) model_time 0.8981 (1.0055) loss 1.2771 (1.0752) grad_norm 9.0670 (9.2921/3.5360) mem 68106MB [2022-12-19 08:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][550/1519] eta 0:16:15 lr 0.000033 time 0.9301 (1.0066) model_time 0.9300 (1.0054) loss 1.5173 (1.0759) grad_norm 12.0396 (9.3526/3.5698) mem 68106MB [2022-12-19 08:41:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][560/1519] eta 0:16:05 lr 0.000033 time 0.9310 (1.0065) model_time 0.9309 (1.0053) loss 1.0142 (1.0769) grad_norm 7.9757 (9.3166/3.5527) mem 68106MB [2022-12-19 08:41:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][570/1519] eta 0:15:55 lr 0.000033 time 0.9918 (1.0064) model_time 0.9917 (1.0053) loss 1.0319 (1.0774) grad_norm 7.8719 (9.2713/3.5395) mem 68106MB [2022-12-19 08:41:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][580/1519] eta 0:15:45 lr 0.000033 time 0.9323 (1.0065) model_time 0.9322 (1.0053) loss 0.7700 (1.0759) grad_norm 15.4961 (9.3224/3.5465) mem 68106MB [2022-12-19 08:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][590/1519] eta 0:15:34 lr 0.000033 time 0.9295 (1.0064) model_time 0.9293 (1.0053) loss 1.2560 (1.0764) grad_norm 6.5802 (9.3299/3.5355) mem 68106MB [2022-12-19 08:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][600/1519] eta 0:15:24 lr 0.000033 time 0.9258 (1.0063) model_time 0.9256 (1.0052) loss 0.9486 (1.0767) grad_norm 11.8305 (9.3485/3.5129) mem 68106MB [2022-12-19 08:42:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][610/1519] eta 0:15:14 lr 0.000033 time 0.9295 (1.0062) model_time 0.9294 (1.0051) loss 0.9844 (1.0760) grad_norm 7.5836 (9.3815/3.5287) mem 68106MB [2022-12-19 08:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][620/1519] eta 0:15:05 lr 0.000033 time 0.9344 (1.0069) model_time 0.9342 (1.0058) loss 1.0582 (1.0771) grad_norm 8.2984 (9.3674/3.5356) mem 68106MB [2022-12-19 08:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][630/1519] eta 0:14:55 lr 0.000033 time 0.9390 (1.0068) model_time 0.9388 (1.0057) loss 0.9021 (1.0768) grad_norm 13.4632 (9.4148/3.5801) mem 68106MB [2022-12-19 08:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][640/1519] eta 0:14:44 lr 0.000033 time 0.9285 (1.0066) model_time 0.9283 (1.0056) loss 1.2745 (1.0756) grad_norm 8.8275 (9.4142/3.5765) mem 68106MB [2022-12-19 08:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][650/1519] eta 0:14:34 lr 0.000033 time 0.9277 (1.0065) model_time 0.9276 (1.0054) loss 1.0749 (1.0755) grad_norm 9.6240 (9.4247/3.5822) mem 68106MB [2022-12-19 08:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][660/1519] eta 0:14:24 lr 0.000033 time 0.9378 (1.0065) model_time 0.9377 (1.0054) loss 1.1485 (1.0750) grad_norm 16.3445 (9.3549/3.1635) mem 68106MB [2022-12-19 08:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][670/1519] eta 0:14:14 lr 0.000033 time 0.9304 (1.0064) model_time 0.9303 (1.0053) loss 0.9659 (1.0741) grad_norm 13.5709 (9.4073/3.2141) mem 68106MB [2022-12-19 08:43:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][680/1519] eta 0:14:04 lr 0.000033 time 0.9297 (1.0062) model_time 0.9296 (1.0052) loss 1.1676 (1.0732) grad_norm 13.4501 (9.4464/3.2207) mem 68106MB [2022-12-19 08:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][690/1519] eta 0:13:54 lr 0.000033 time 1.0140 (1.0063) model_time 1.0139 (1.0052) loss 1.0863 (1.0731) grad_norm 11.9402 (9.4289/3.1962) mem 68106MB [2022-12-19 08:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][700/1519] eta 0:13:44 lr 0.000033 time 0.9307 (1.0062) model_time 0.9306 (1.0052) loss 1.2537 (1.0748) grad_norm 6.2227 (9.4204/3.1948) mem 68106MB [2022-12-19 08:44:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][710/1519] eta 0:13:33 lr 0.000033 time 0.9321 (1.0061) model_time 0.9319 (1.0051) loss 1.4190 (1.0742) grad_norm 9.5206 (9.4348/3.1815) mem 68106MB [2022-12-19 08:44:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][720/1519] eta 0:13:23 lr 0.000033 time 0.9304 (1.0060) model_time 0.9302 (1.0050) loss 1.1186 (1.0744) grad_norm 9.7217 (9.5074/3.2472) mem 68106MB [2022-12-19 08:44:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][730/1519] eta 0:13:13 lr 0.000033 time 0.9285 (1.0058) model_time 0.9283 (1.0048) loss 1.0773 (1.0746) grad_norm 10.2011 (9.5511/3.2643) mem 68106MB [2022-12-19 08:44:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][740/1519] eta 0:13:03 lr 0.000033 time 0.9310 (1.0058) model_time 0.9307 (1.0048) loss 1.0798 (1.0755) grad_norm 7.9692 (9.4854/3.2320) mem 68106MB [2022-12-19 08:44:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][750/1519] eta 0:12:53 lr 0.000033 time 0.9313 (1.0060) model_time 0.9311 (1.0050) loss 1.0768 (1.0748) grad_norm 14.0077 (9.5311/3.2398) mem 68106MB [2022-12-19 08:45:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][760/1519] eta 0:12:43 lr 0.000033 time 0.9309 (1.0059) model_time 0.9307 (1.0049) loss 1.1500 (1.0766) grad_norm 7.3577 (9.5675/3.2535) mem 68106MB [2022-12-19 08:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][770/1519] eta 0:12:33 lr 0.000033 time 0.9265 (1.0058) model_time 0.9263 (1.0048) loss 1.1547 (1.0762) grad_norm 11.0222 (9.5474/3.2597) mem 68106MB [2022-12-19 08:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][780/1519] eta 0:12:23 lr 0.000033 time 0.9322 (1.0057) model_time 0.9321 (1.0048) loss 1.1131 (1.0773) grad_norm 8.6162 (9.5328/3.2611) mem 68106MB [2022-12-19 08:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][790/1519] eta 0:12:13 lr 0.000033 time 0.9377 (1.0056) model_time 0.9376 (1.0047) loss 1.3206 (1.0785) grad_norm 6.2654 (9.5398/3.2807) mem 68106MB [2022-12-19 08:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][800/1519] eta 0:12:02 lr 0.000033 time 0.9291 (1.0055) model_time 0.9289 (1.0046) loss 1.0242 (1.0802) grad_norm 12.5069 (9.6274/3.3944) mem 68106MB [2022-12-19 08:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][810/1519] eta 0:11:52 lr 0.000033 time 0.9367 (1.0055) model_time 0.9365 (1.0045) loss 1.3917 (1.0801) grad_norm 7.3019 (9.6457/3.3899) mem 68106MB [2022-12-19 08:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][820/1519] eta 0:11:43 lr 0.000033 time 0.9285 (1.0057) model_time 0.9284 (1.0048) loss 1.0542 (1.0811) grad_norm 7.6984 (9.6407/3.3900) mem 68106MB [2022-12-19 08:46:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][830/1519] eta 0:11:32 lr 0.000033 time 0.9294 (1.0057) model_time 0.9293 (1.0047) loss 1.8076 (1.0814) grad_norm 9.7451 (9.6234/3.3677) mem 68106MB [2022-12-19 08:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][840/1519] eta 0:11:22 lr 0.000033 time 0.9351 (1.0057) model_time 0.9349 (1.0048) loss 1.1017 (1.0806) grad_norm 16.0522 (9.6105/3.3641) mem 68106MB [2022-12-19 08:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][850/1519] eta 0:11:12 lr 0.000033 time 0.9344 (1.0056) model_time 0.9343 (1.0047) loss 1.1455 (1.0811) grad_norm 6.1403 (9.5520/3.2886) mem 68106MB [2022-12-19 08:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][860/1519] eta 0:11:02 lr 0.000033 time 0.9376 (1.0056) model_time 0.9374 (1.0047) loss 1.1286 (1.0814) grad_norm 6.3420 (9.5211/3.2948) mem 68106MB [2022-12-19 08:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][870/1519] eta 0:10:52 lr 0.000033 time 0.9312 (1.0055) model_time 0.9311 (1.0047) loss 1.0807 (1.0812) grad_norm 9.0232 (9.5267/3.2890) mem 68106MB [2022-12-19 08:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][880/1519] eta 0:10:42 lr 0.000033 time 0.9306 (1.0055) model_time 0.9305 (1.0046) loss 1.3031 (1.0809) grad_norm 9.8945 (9.5204/3.2986) mem 68106MB [2022-12-19 08:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][890/1519] eta 0:10:32 lr 0.000033 time 0.9322 (1.0056) model_time 0.9320 (1.0047) loss 1.0842 (1.0804) grad_norm 6.0545 (9.5296/3.2973) mem 68106MB [2022-12-19 08:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][900/1519] eta 0:10:22 lr 0.000033 time 0.9563 (1.0055) model_time 0.9562 (1.0047) loss 1.2014 (1.0807) grad_norm 9.5461 (9.4937/3.2988) mem 68106MB [2022-12-19 08:47:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][910/1519] eta 0:10:12 lr 0.000033 time 0.9292 (1.0055) model_time 0.9290 (1.0046) loss 1.4262 (1.0809) grad_norm 10.6653 (9.4578/3.2912) mem 68106MB [2022-12-19 08:47:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][920/1519] eta 0:10:02 lr 0.000033 time 0.9329 (1.0054) model_time 0.9328 (1.0045) loss 1.1561 (1.0813) grad_norm 6.6397 (9.4923/3.4668) mem 68106MB [2022-12-19 08:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][930/1519] eta 0:09:52 lr 0.000033 time 0.9300 (1.0056) model_time 0.9299 (1.0047) loss 1.1694 (1.0811) grad_norm 7.3593 (9.5036/3.4586) mem 68106MB [2022-12-19 08:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][940/1519] eta 0:09:42 lr 0.000033 time 0.9322 (1.0055) model_time 0.9320 (1.0047) loss 0.7745 (1.0809) grad_norm 12.1506 (9.4779/3.4082) mem 68106MB [2022-12-19 08:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][950/1519] eta 0:09:32 lr 0.000033 time 0.9285 (1.0055) model_time 0.9283 (1.0046) loss 1.4741 (1.0816) grad_norm 8.4345 (9.4762/3.4053) mem 68106MB [2022-12-19 08:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][960/1519] eta 0:09:22 lr 0.000033 time 0.9314 (1.0054) model_time 0.9313 (1.0046) loss 0.9865 (1.0818) grad_norm 8.6386 (9.4508/3.4142) mem 68106MB [2022-12-19 08:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][970/1519] eta 0:09:11 lr 0.000033 time 0.9308 (1.0053) model_time 0.9306 (1.0045) loss 0.9006 (1.0807) grad_norm 6.7808 (9.4683/3.4043) mem 68106MB [2022-12-19 08:48:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][980/1519] eta 0:09:01 lr 0.000033 time 0.9308 (1.0053) model_time 0.9306 (1.0045) loss 1.1991 (1.0811) grad_norm 8.3157 (9.4373/3.4072) mem 68106MB [2022-12-19 08:48:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][990/1519] eta 0:08:51 lr 0.000033 time 0.9365 (1.0052) model_time 0.9364 (1.0044) loss 1.4018 (1.0818) grad_norm 6.6042 (9.4029/3.3714) mem 68106MB [2022-12-19 08:49:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1000/1519] eta 0:08:41 lr 0.000033 time 0.9340 (1.0052) model_time 0.9338 (1.0044) loss 1.0135 (1.0812) grad_norm 6.3080 (9.3714/3.3601) mem 68106MB [2022-12-19 08:49:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1010/1519] eta 0:08:31 lr 0.000033 time 0.9842 (1.0053) model_time 0.9841 (1.0045) loss 0.8260 (1.0798) grad_norm 7.8794 (9.3012/3.3031) mem 68106MB [2022-12-19 08:49:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1020/1519] eta 0:08:21 lr 0.000033 time 0.9529 (1.0053) model_time 0.9528 (1.0044) loss 1.1600 (1.0798) grad_norm 7.1560 (9.2817/3.3076) mem 68106MB [2022-12-19 08:49:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1030/1519] eta 0:08:11 lr 0.000033 time 0.9320 (1.0052) model_time 0.9319 (1.0044) loss 1.3963 (1.0806) grad_norm 8.3989 (9.3415/3.3050) mem 68106MB [2022-12-19 08:49:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1040/1519] eta 0:08:01 lr 0.000033 time 0.9330 (1.0052) model_time 0.9328 (1.0043) loss 0.8268 (1.0796) grad_norm 7.3955 (9.2970/3.1901) mem 68106MB [2022-12-19 08:49:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1050/1519] eta 0:07:51 lr 0.000033 time 0.9289 (1.0051) model_time 0.9288 (1.0043) loss 1.3232 (1.0790) grad_norm 6.5055 (9.2718/3.1706) mem 68106MB [2022-12-19 08:50:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1060/1519] eta 0:07:41 lr 0.000033 time 0.9344 (1.0051) model_time 0.9343 (1.0043) loss 1.0614 (1.0789) grad_norm 9.2774 (9.2675/3.1813) mem 68106MB [2022-12-19 08:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1070/1519] eta 0:07:31 lr 0.000033 time 0.9294 (1.0051) model_time 0.9293 (1.0042) loss 0.9599 (1.0787) grad_norm 7.7880 (9.2590/3.1759) mem 68106MB [2022-12-19 08:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1080/1519] eta 0:07:21 lr 0.000033 time 0.9327 (1.0050) model_time 0.9325 (1.0042) loss 1.3946 (1.0786) grad_norm 8.8046 (9.2489/3.1758) mem 68106MB [2022-12-19 08:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1090/1519] eta 0:07:11 lr 0.000033 time 0.9313 (1.0050) model_time 0.9311 (1.0042) loss 0.8814 (1.0784) grad_norm 6.6651 (9.2435/3.1363) mem 68106MB [2022-12-19 08:50:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1100/1519] eta 0:07:01 lr 0.000033 time 0.9299 (1.0050) model_time 0.9297 (1.0041) loss 0.9437 (1.0786) grad_norm 7.5104 (9.2259/3.1444) mem 68106MB [2022-12-19 08:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1110/1519] eta 0:06:51 lr 0.000033 time 0.9022 (1.0049) model_time 0.9021 (1.0041) loss 1.0412 (1.0788) grad_norm 6.4409 (9.1771/3.1366) mem 68106MB [2022-12-19 08:51:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1120/1519] eta 0:06:40 lr 0.000033 time 0.9342 (1.0049) model_time 0.9341 (1.0041) loss 0.7699 (1.0783) grad_norm 5.7103 (9.1905/3.1564) mem 68106MB [2022-12-19 08:51:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1130/1519] eta 0:06:30 lr 0.000033 time 1.0130 (1.0049) model_time 1.0129 (1.0041) loss 1.0379 (1.0772) grad_norm 9.5454 (9.2027/3.1606) mem 68106MB [2022-12-19 08:51:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1140/1519] eta 0:06:20 lr 0.000033 time 0.9345 (1.0049) model_time 0.9343 (1.0041) loss 1.2602 (1.0767) grad_norm 22.4381 (9.2268/3.2393) mem 68106MB [2022-12-19 08:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1150/1519] eta 0:06:10 lr 0.000033 time 0.9259 (1.0049) model_time 0.9258 (1.0041) loss 0.9723 (1.0756) grad_norm 6.5823 (9.1694/3.1773) mem 68106MB [2022-12-19 08:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1160/1519] eta 0:06:00 lr 0.000033 time 0.9323 (1.0048) model_time 0.9321 (1.0041) loss 0.9350 (1.0744) grad_norm 8.7913 (9.1753/3.1716) mem 68106MB [2022-12-19 08:51:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1170/1519] eta 0:05:50 lr 0.000033 time 0.9290 (1.0048) model_time 0.9288 (1.0041) loss 0.9598 (1.0744) grad_norm 12.1554 (9.2524/3.1837) mem 68106MB [2022-12-19 08:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1180/1519] eta 0:05:40 lr 0.000033 time 0.9294 (1.0048) model_time 0.9293 (1.0040) loss 1.2855 (1.0751) grad_norm 10.0727 (9.1828/3.1504) mem 68106MB [2022-12-19 08:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1190/1519] eta 0:05:30 lr 0.000033 time 0.9874 (1.0049) model_time 0.9872 (1.0041) loss 1.0271 (1.0745) grad_norm 7.3715 (9.1482/3.1425) mem 68106MB [2022-12-19 08:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1200/1519] eta 0:05:20 lr 0.000033 time 0.9315 (1.0049) model_time 0.9314 (1.0041) loss 1.3873 (1.0743) grad_norm 7.3144 (9.1281/3.1497) mem 68106MB [2022-12-19 08:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1210/1519] eta 0:05:10 lr 0.000033 time 0.9294 (1.0048) model_time 0.9293 (1.0041) loss 1.0099 (1.0746) grad_norm 6.1016 (9.0923/3.1211) mem 68106MB [2022-12-19 08:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1220/1519] eta 0:05:00 lr 0.000033 time 0.9360 (1.0048) model_time 0.9358 (1.0040) loss 1.3516 (1.0742) grad_norm 7.8023 (9.0729/3.1211) mem 68106MB [2022-12-19 08:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1230/1519] eta 0:04:50 lr 0.000033 time 0.9302 (1.0047) model_time 0.9301 (1.0040) loss 1.3247 (1.0743) grad_norm 7.5422 (8.9897/3.0668) mem 68106MB [2022-12-19 08:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1240/1519] eta 0:04:40 lr 0.000033 time 0.9301 (1.0050) model_time 0.9299 (1.0042) loss 1.3527 (1.0742) grad_norm 11.3076 (8.9871/3.0753) mem 68106MB [2022-12-19 08:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1250/1519] eta 0:04:30 lr 0.000033 time 0.9384 (1.0050) model_time 0.9382 (1.0042) loss 1.2491 (1.0744) grad_norm 6.6225 (8.9847/3.0763) mem 68106MB [2022-12-19 08:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1260/1519] eta 0:04:20 lr 0.000033 time 0.9293 (1.0049) model_time 0.9292 (1.0042) loss 1.3078 (1.0751) grad_norm 8.0039 (8.9486/3.0318) mem 68106MB [2022-12-19 08:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1270/1519] eta 0:04:10 lr 0.000033 time 0.9313 (1.0049) model_time 0.9312 (1.0041) loss 0.9056 (1.0746) grad_norm 10.4679 (8.9201/2.9986) mem 68106MB [2022-12-19 08:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1280/1519] eta 0:04:00 lr 0.000033 time 0.9309 (1.0048) model_time 0.9308 (1.0041) loss 0.9908 (1.0743) grad_norm 7.6943 (8.8631/2.9728) mem 68106MB [2022-12-19 08:53:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1290/1519] eta 0:03:50 lr 0.000033 time 0.9284 (1.0048) model_time 0.9283 (1.0041) loss 0.7625 (1.0738) grad_norm 5.5748 (8.8271/2.9658) mem 68106MB [2022-12-19 08:54:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1300/1519] eta 0:03:40 lr 0.000033 time 0.9346 (1.0048) model_time 0.9345 (1.0040) loss 1.5092 (1.0743) grad_norm 8.0613 (8.8616/2.9694) mem 68106MB [2022-12-19 08:54:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1310/1519] eta 0:03:30 lr 0.000033 time 1.0040 (1.0048) model_time 1.0038 (1.0041) loss 1.0270 (1.0739) grad_norm 6.9123 (8.8360/2.9595) mem 68106MB [2022-12-19 08:54:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9340 (1.0048) model_time 0.9338 (1.0040) loss 0.9249 (1.0740) grad_norm 5.9107 (8.7648/2.8827) mem 68106MB [2022-12-19 08:54:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9237 (1.0048) model_time 0.9235 (1.0041) loss 1.2218 (1.0739) grad_norm 7.6214 (8.6996/2.8351) mem 68106MB [2022-12-19 08:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9551 (1.0048) model_time 0.9550 (1.0041) loss 0.7394 (1.0737) grad_norm 12.3635 (8.7587/2.8473) mem 68106MB [2022-12-19 08:54:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9348 (1.0049) model_time 0.9347 (1.0041) loss 0.8135 (1.0737) grad_norm 8.1538 (8.7189/2.8193) mem 68106MB [2022-12-19 08:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9655 (1.0049) model_time 0.9654 (1.0042) loss 0.8717 (1.0737) grad_norm 7.8811 (8.6998/2.7807) mem 68106MB [2022-12-19 08:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9944 (1.0049) model_time 0.9943 (1.0042) loss 0.9916 (1.0738) grad_norm 5.6961 (8.6674/2.7808) mem 68106MB [2022-12-19 08:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9492 (1.0050) model_time 0.9491 (1.0043) loss 1.4028 (1.0744) grad_norm 8.1190 (8.6754/2.7875) mem 68106MB [2022-12-19 08:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9338 (1.0050) model_time 0.9336 (1.0043) loss 1.4927 (1.0747) grad_norm 9.8949 (8.7223/2.8707) mem 68106MB [2022-12-19 08:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9345 (1.0049) model_time 0.9343 (1.0042) loss 0.9339 (1.0743) grad_norm 7.3107 (8.6188/2.6986) mem 68106MB [2022-12-19 08:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9304 (1.0049) model_time 0.9302 (1.0042) loss 0.9206 (1.0741) grad_norm 6.7927 (8.6176/2.6914) mem 68106MB [2022-12-19 08:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1420/1519] eta 0:01:39 lr 0.000033 time 0.9295 (1.0049) model_time 0.9293 (1.0042) loss 0.9464 (1.0737) grad_norm 9.5702 (8.6798/2.8082) mem 68106MB [2022-12-19 08:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9389 (1.0049) model_time 0.9388 (1.0042) loss 0.8431 (1.0737) grad_norm 8.2548 (8.6589/2.8048) mem 68106MB [2022-12-19 08:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9312 (1.0048) model_time 0.9311 (1.0041) loss 0.7899 (1.0750) grad_norm 9.9957 (8.6451/2.7986) mem 68106MB [2022-12-19 08:56:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9443 (1.0048) model_time 0.9442 (1.0041) loss 1.1996 (1.0746) grad_norm 6.0546 (8.6827/2.8230) mem 68106MB [2022-12-19 08:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9382 (1.0048) model_time 0.9381 (1.0041) loss 1.2087 (1.0746) grad_norm 10.7954 (8.6936/2.8284) mem 68106MB [2022-12-19 08:56:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9307 (1.0048) model_time 0.9305 (1.0041) loss 1.2304 (1.0753) grad_norm 9.6046 (8.7425/2.8440) mem 68106MB [2022-12-19 08:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9356 (1.0048) model_time 0.9354 (1.0041) loss 1.4759 (1.0755) grad_norm 6.3072 (8.7337/2.8395) mem 68106MB [2022-12-19 08:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9352 (1.0047) model_time 0.9351 (1.0040) loss 1.1762 (1.0752) grad_norm 7.9389 (8.7294/2.8329) mem 68106MB [2022-12-19 08:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9358 (1.0048) model_time 0.9356 (1.0041) loss 0.8874 (1.0757) grad_norm 6.0523 (8.7414/2.8291) mem 68106MB [2022-12-19 08:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [10/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9319 (1.0048) model_time 0.9318 (1.0041) loss 0.8112 (1.0753) grad_norm 10.7727 (8.7368/2.8316) mem 68106MB [2022-12-19 08:57:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 10 training takes 0:25:26 [2022-12-19 08:57:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_10.pth saving...... [2022-12-19 08:58:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_10.pth saved !!! [2022-12-19 08:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.8166 (0.8166) Acc@1 82.639 (82.639) Acc@5 95.833 (95.833) Mem 68106MB [2022-12-19 08:58:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.331) Loss 0.8102 (0.7934) Acc@1 86.806 (84.501) Acc@5 96.181 (96.749) Mem 68106MB [2022-12-19 08:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.315) Loss 0.7445 (0.7943) Acc@1 83.681 (84.259) Acc@5 98.264 (96.710) Mem 68106MB [2022-12-19 08:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.8392 (0.7942) Acc@1 84.028 (84.353) Acc@5 95.833 (96.685) Mem 68106MB [2022-12-19 08:58:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.307) Loss 0.7589 (0.7847) Acc@1 83.333 (84.375) Acc@5 96.875 (96.867) Mem 68106MB [2022-12-19 08:58:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.8125 (0.7797) Acc@1 81.944 (84.470) Acc@5 96.875 (96.902) Mem 68106MB [2022-12-19 08:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.301 (0.304) Loss 0.8119 (0.7804) Acc@1 83.333 (84.449) Acc@5 95.833 (96.932) Mem 68106MB [2022-12-19 08:58:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.8832 (0.7831) Acc@1 84.375 (84.370) Acc@5 96.528 (96.948) Mem 68106MB [2022-12-19 08:58:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.7381 (0.7830) Acc@1 82.292 (84.336) Acc@5 97.222 (96.956) Mem 68106MB [2022-12-19 08:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:10] * Acc@1 84.357 Acc@5 96.979 [2022-12-19 08:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 84.4% [2022-12-19 08:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 08:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 08:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 84.36% [2022-12-19 08:58:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][0/1519] eta 0:34:18 lr 0.000033 time 1.3550 (1.3550) model_time 0.9637 (0.9637) loss 1.0617 (1.0617) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 08:59:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][10/1519] eta 0:26:05 lr 0.000033 time 0.9393 (1.0372) model_time 0.9391 (1.0012) loss 0.7999 (0.9856) grad_norm 10.7416 (10.0698/4.3559) mem 68106MB [2022-12-19 08:59:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][20/1519] eta 0:25:41 lr 0.000033 time 0.9391 (1.0284) model_time 0.9387 (1.0094) loss 0.8953 (1.0008) grad_norm 9.5850 (10.6705/3.5719) mem 68106MB [2022-12-19 08:59:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][30/1519] eta 0:25:18 lr 0.000033 time 0.9294 (1.0196) model_time 0.9293 (1.0066) loss 0.9552 (1.0013) grad_norm 7.1183 (10.6399/3.9910) mem 68106MB [2022-12-19 08:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][40/1519] eta 0:25:05 lr 0.000033 time 0.9207 (1.0180) model_time 0.9204 (1.0081) loss 1.0365 (0.9884) grad_norm 12.1215 (10.9325/3.8567) mem 68106MB [2022-12-19 08:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][50/1519] eta 0:24:51 lr 0.000033 time 0.9360 (1.0155) model_time 0.9358 (1.0074) loss 0.7679 (0.9913) grad_norm 8.7103 (10.3534/3.7120) mem 68106MB [2022-12-19 08:59:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][60/1519] eta 0:24:37 lr 0.000033 time 0.9256 (1.0126) model_time 0.9253 (1.0057) loss 0.7009 (0.9726) grad_norm 9.3800 (10.2954/3.5037) mem 68106MB [2022-12-19 09:00:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][70/1519] eta 0:24:24 lr 0.000033 time 0.9262 (1.0109) model_time 0.9256 (1.0050) loss 1.2596 (0.9789) grad_norm 6.7741 (10.0723/3.3295) mem 68106MB [2022-12-19 09:00:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][80/1519] eta 0:24:13 lr 0.000033 time 0.9419 (1.0098) model_time 0.9417 (1.0045) loss 1.0152 (0.9850) grad_norm 11.4597 (9.8907/3.2730) mem 68106MB [2022-12-19 09:00:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][90/1519] eta 0:24:01 lr 0.000033 time 0.9400 (1.0087) model_time 0.9399 (1.0040) loss 0.7872 (0.9829) grad_norm 7.8447 (9.8962/3.4065) mem 68106MB [2022-12-19 09:00:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][100/1519] eta 0:23:50 lr 0.000033 time 0.9348 (1.0079) model_time 0.9347 (1.0035) loss 1.3235 (0.9993) grad_norm 12.9943 (9.7854/3.3203) mem 68106MB [2022-12-19 09:00:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][110/1519] eta 0:23:40 lr 0.000033 time 0.9449 (1.0081) model_time 0.9448 (1.0042) loss 1.0300 (1.0107) grad_norm 5.6264 (9.7199/3.2263) mem 68106MB [2022-12-19 09:00:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][120/1519] eta 0:23:29 lr 0.000033 time 0.9340 (1.0076) model_time 0.9339 (1.0039) loss 1.1874 (1.0188) grad_norm 6.6127 (9.6347/3.2498) mem 68106MB [2022-12-19 09:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][130/1519] eta 0:23:20 lr 0.000033 time 0.9394 (1.0082) model_time 0.9393 (1.0047) loss 0.9326 (1.0173) grad_norm 10.5385 (9.7054/3.6402) mem 68106MB [2022-12-19 09:01:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][140/1519] eta 0:23:09 lr 0.000033 time 0.9350 (1.0076) model_time 0.9348 (1.0044) loss 1.1326 (1.0254) grad_norm 7.2440 (9.5696/3.5521) mem 68106MB [2022-12-19 09:01:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][150/1519] eta 0:22:58 lr 0.000033 time 0.9436 (1.0072) model_time 0.9435 (1.0042) loss 0.9551 (1.0267) grad_norm 5.9114 (9.4274/3.4798) mem 68106MB [2022-12-19 09:01:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][160/1519] eta 0:22:48 lr 0.000033 time 0.9291 (1.0068) model_time 0.9290 (1.0040) loss 0.8574 (1.0193) grad_norm 7.3834 (9.3770/3.4021) mem 68106MB [2022-12-19 09:01:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][170/1519] eta 0:22:39 lr 0.000033 time 1.0674 (1.0078) model_time 1.0673 (1.0050) loss 1.0450 (1.0220) grad_norm 6.1421 (9.3760/3.3635) mem 68106MB [2022-12-19 09:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][180/1519] eta 0:22:28 lr 0.000033 time 0.9301 (1.0073) model_time 0.9299 (1.0047) loss 1.3729 (1.0243) grad_norm 7.9255 (9.3190/3.3056) mem 68106MB [2022-12-19 09:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][190/1519] eta 0:22:18 lr 0.000033 time 0.9288 (1.0068) model_time 0.9286 (1.0044) loss 0.8320 (1.0269) grad_norm 9.5158 (9.2332/3.2454) mem 68106MB [2022-12-19 09:02:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][200/1519] eta 0:22:07 lr 0.000033 time 0.9316 (1.0067) model_time 0.9314 (1.0043) loss 0.9538 (1.0296) grad_norm 10.6317 (9.2007/3.1894) mem 68106MB [2022-12-19 09:02:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][210/1519] eta 0:21:57 lr 0.000033 time 0.9243 (1.0066) model_time 0.9242 (1.0043) loss 0.7774 (1.0325) grad_norm 7.2724 (9.0871/3.1551) mem 68106MB [2022-12-19 09:02:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][220/1519] eta 0:21:47 lr 0.000033 time 0.9375 (1.0068) model_time 0.9374 (1.0046) loss 0.9003 (1.0304) grad_norm 11.1238 (9.1246/3.1219) mem 68106MB [2022-12-19 09:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][230/1519] eta 0:21:37 lr 0.000033 time 0.9293 (1.0064) model_time 0.9290 (1.0043) loss 0.9091 (1.0296) grad_norm 16.7761 (9.1605/3.1557) mem 68106MB [2022-12-19 09:02:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][240/1519] eta 0:21:26 lr 0.000033 time 0.9298 (1.0062) model_time 0.9297 (1.0041) loss 1.0178 (1.0275) grad_norm 9.9148 (9.1146/3.1101) mem 68106MB [2022-12-19 09:03:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][250/1519] eta 0:21:16 lr 0.000033 time 0.9427 (1.0061) model_time 0.9425 (1.0041) loss 1.0880 (1.0277) grad_norm 6.1897 (9.0833/3.0760) mem 68106MB [2022-12-19 09:03:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][260/1519] eta 0:21:06 lr 0.000033 time 0.9342 (1.0059) model_time 0.9340 (1.0040) loss 0.9906 (1.0256) grad_norm 7.4789 (9.1147/3.0605) mem 68106MB [2022-12-19 09:03:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][270/1519] eta 0:20:55 lr 0.000033 time 0.9277 (1.0056) model_time 0.9276 (1.0038) loss 0.8891 (1.0229) grad_norm 6.7225 (9.0700/3.0400) mem 68106MB [2022-12-19 09:03:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][280/1519] eta 0:20:45 lr 0.000033 time 0.9333 (1.0054) model_time 0.9331 (1.0036) loss 1.2373 (1.0232) grad_norm 11.5744 (9.0507/3.0098) mem 68106MB [2022-12-19 09:03:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][290/1519] eta 0:20:35 lr 0.000033 time 0.9328 (1.0054) model_time 0.9325 (1.0037) loss 0.9942 (1.0240) grad_norm 5.7564 (8.9979/2.9821) mem 68106MB [2022-12-19 09:03:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][300/1519] eta 0:20:25 lr 0.000033 time 0.9331 (1.0053) model_time 0.9329 (1.0036) loss 1.8571 (1.0275) grad_norm 7.8526 (8.9864/2.9541) mem 68106MB [2022-12-19 09:04:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][310/1519] eta 0:20:16 lr 0.000033 time 0.9328 (1.0061) model_time 0.9326 (1.0044) loss 0.8373 (1.0278) grad_norm 10.3561 (9.0524/3.0111) mem 68106MB [2022-12-19 09:04:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][320/1519] eta 0:20:06 lr 0.000033 time 0.9308 (1.0059) model_time 0.9306 (1.0043) loss 1.3904 (1.0313) grad_norm 8.6644 (9.0844/2.9891) mem 68106MB [2022-12-19 09:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][330/1519] eta 0:19:56 lr 0.000033 time 0.9346 (1.0060) model_time 0.9344 (1.0044) loss 1.4836 (1.0365) grad_norm 13.0884 (9.1078/2.9619) mem 68106MB [2022-12-19 09:04:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][340/1519] eta 0:19:46 lr 0.000033 time 0.9434 (1.0060) model_time 0.9432 (1.0045) loss 0.8247 (1.0366) grad_norm 7.5973 (9.0471/2.9428) mem 68106MB [2022-12-19 09:04:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][350/1519] eta 0:19:36 lr 0.000033 time 1.0292 (1.0067) model_time 1.0290 (1.0052) loss 0.8865 (1.0360) grad_norm 6.6165 (9.0158/2.9162) mem 68106MB [2022-12-19 09:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][360/1519] eta 0:19:26 lr 0.000033 time 0.9319 (1.0065) model_time 0.9318 (1.0050) loss 1.1810 (1.0361) grad_norm 8.6841 (8.9577/2.9011) mem 68106MB [2022-12-19 09:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][370/1519] eta 0:19:16 lr 0.000033 time 0.9323 (1.0064) model_time 0.9322 (1.0049) loss 0.8127 (1.0364) grad_norm 6.8126 (8.9638/2.8709) mem 68106MB [2022-12-19 09:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][380/1519] eta 0:19:06 lr 0.000033 time 0.9331 (1.0063) model_time 0.9329 (1.0049) loss 0.7983 (1.0375) grad_norm 10.0711 (8.9446/2.8426) mem 68106MB [2022-12-19 09:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][390/1519] eta 0:18:55 lr 0.000033 time 0.9300 (1.0061) model_time 0.9299 (1.0047) loss 1.4299 (1.0384) grad_norm 8.5377 (8.9565/2.8268) mem 68106MB [2022-12-19 09:05:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][400/1519] eta 0:18:45 lr 0.000033 time 0.9281 (1.0060) model_time 0.9280 (1.0046) loss 0.9163 (1.0401) grad_norm 11.7184 (8.9509/2.8069) mem 68106MB [2022-12-19 09:05:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][410/1519] eta 0:18:35 lr 0.000033 time 0.9361 (1.0058) model_time 0.9360 (1.0044) loss 0.9713 (1.0407) grad_norm 7.7363 (8.9945/2.8144) mem 68106MB [2022-12-19 09:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][420/1519] eta 0:18:25 lr 0.000033 time 0.9342 (1.0056) model_time 0.9341 (1.0043) loss 1.1304 (1.0396) grad_norm 7.1332 (8.9985/2.8080) mem 68106MB [2022-12-19 09:06:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][430/1519] eta 0:18:14 lr 0.000033 time 0.9291 (1.0055) model_time 0.9290 (1.0042) loss 0.9427 (1.0391) grad_norm 6.8834 (8.9772/2.7935) mem 68106MB [2022-12-19 09:06:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][440/1519] eta 0:18:04 lr 0.000033 time 0.9444 (1.0054) model_time 0.9442 (1.0041) loss 1.3302 (1.0397) grad_norm 7.7482 (8.9666/2.7683) mem 68106MB [2022-12-19 09:06:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][450/1519] eta 0:17:54 lr 0.000033 time 0.9310 (1.0052) model_time 0.9309 (1.0040) loss 0.9229 (1.0398) grad_norm 9.9944 (8.9800/2.7500) mem 68106MB [2022-12-19 09:06:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][460/1519] eta 0:17:44 lr 0.000033 time 0.9348 (1.0052) model_time 0.9347 (1.0039) loss 0.9851 (1.0404) grad_norm 5.7085 (8.9993/2.8160) mem 68106MB [2022-12-19 09:06:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][470/1519] eta 0:17:34 lr 0.000033 time 0.9297 (1.0050) model_time 0.9295 (1.0038) loss 0.8861 (1.0400) grad_norm 9.6097 (9.0050/2.7949) mem 68106MB [2022-12-19 09:07:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][480/1519] eta 0:17:24 lr 0.000033 time 0.9265 (1.0049) model_time 0.9263 (1.0037) loss 0.8671 (1.0385) grad_norm 10.2428 (9.0028/2.7685) mem 68106MB [2022-12-19 09:07:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][490/1519] eta 0:17:14 lr 0.000033 time 0.9335 (1.0050) model_time 0.9332 (1.0039) loss 0.7649 (1.0360) grad_norm 10.5776 (9.0684/2.7973) mem 68106MB [2022-12-19 09:07:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][500/1519] eta 0:17:04 lr 0.000033 time 0.9333 (1.0049) model_time 0.9332 (1.0038) loss 0.9337 (1.0365) grad_norm 7.1513 (9.0224/2.7889) mem 68106MB [2022-12-19 09:07:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][510/1519] eta 0:16:53 lr 0.000033 time 0.9302 (1.0048) model_time 0.9300 (1.0037) loss 1.3281 (1.0373) grad_norm 8.6491 (9.0184/2.8020) mem 68106MB [2022-12-19 09:07:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][520/1519] eta 0:16:43 lr 0.000033 time 0.9277 (1.0047) model_time 0.9276 (1.0036) loss 0.9498 (1.0375) grad_norm 5.5151 (9.0485/2.8220) mem 68106MB [2022-12-19 09:07:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][530/1519] eta 0:16:33 lr 0.000033 time 0.9356 (1.0046) model_time 0.9354 (1.0035) loss 0.9370 (1.0385) grad_norm 8.7426 (9.0495/2.8093) mem 68106MB [2022-12-19 09:08:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][540/1519] eta 0:16:23 lr 0.000033 time 0.9333 (1.0045) model_time 0.9332 (1.0034) loss 1.0752 (1.0404) grad_norm 5.8221 (9.0451/2.8396) mem 68106MB [2022-12-19 09:08:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][550/1519] eta 0:16:13 lr 0.000033 time 0.9285 (1.0044) model_time 0.9283 (1.0033) loss 1.0832 (1.0412) grad_norm 9.1360 (9.0725/2.8511) mem 68106MB [2022-12-19 09:08:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][560/1519] eta 0:16:03 lr 0.000033 time 0.9289 (1.0043) model_time 0.9288 (1.0032) loss 0.7568 (1.0402) grad_norm 9.5153 (9.0526/2.8339) mem 68106MB [2022-12-19 09:08:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][570/1519] eta 0:15:53 lr 0.000033 time 0.9280 (1.0042) model_time 0.9278 (1.0032) loss 1.1559 (1.0400) grad_norm 7.6547 (9.0579/2.8308) mem 68106MB [2022-12-19 09:08:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][580/1519] eta 0:15:42 lr 0.000033 time 0.9335 (1.0041) model_time 0.9334 (1.0031) loss 0.7029 (1.0387) grad_norm 9.0909 (9.0454/2.8126) mem 68106MB [2022-12-19 09:08:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][590/1519] eta 0:15:32 lr 0.000033 time 0.9331 (1.0040) model_time 0.9329 (1.0030) loss 0.9417 (1.0399) grad_norm 7.0236 (9.0222/2.8000) mem 68106MB [2022-12-19 09:09:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][600/1519] eta 0:15:22 lr 0.000033 time 0.9329 (1.0041) model_time 0.9327 (1.0031) loss 1.5336 (1.0407) grad_norm 8.9508 (9.0413/2.7989) mem 68106MB [2022-12-19 09:09:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][610/1519] eta 0:15:12 lr 0.000033 time 0.9257 (1.0041) model_time 0.9256 (1.0031) loss 0.8115 (1.0389) grad_norm 6.3289 (9.0062/2.7502) mem 68106MB [2022-12-19 09:09:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][620/1519] eta 0:15:02 lr 0.000033 time 0.9339 (1.0042) model_time 0.9337 (1.0032) loss 1.2723 (1.0389) grad_norm 11.2222 (8.9748/2.7381) mem 68106MB [2022-12-19 09:09:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][630/1519] eta 0:14:52 lr 0.000033 time 0.9279 (1.0041) model_time 0.9278 (1.0031) loss 1.2597 (1.0386) grad_norm 5.6562 (8.9523/2.7030) mem 68106MB [2022-12-19 09:09:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][640/1519] eta 0:14:42 lr 0.000033 time 0.9299 (1.0041) model_time 0.9298 (1.0031) loss 0.9967 (1.0384) grad_norm 6.6491 (8.8872/2.6529) mem 68106MB [2022-12-19 09:09:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][650/1519] eta 0:14:32 lr 0.000033 time 0.9355 (1.0042) model_time 0.9353 (1.0032) loss 0.9533 (1.0382) grad_norm 6.6191 (8.9359/2.6912) mem 68106MB [2022-12-19 09:10:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][660/1519] eta 0:14:22 lr 0.000033 time 0.9311 (1.0044) model_time 0.9310 (1.0034) loss 1.2891 (1.0381) grad_norm 9.3105 (8.9810/2.7348) mem 68106MB [2022-12-19 09:10:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][670/1519] eta 0:14:12 lr 0.000033 time 0.9310 (1.0045) model_time 0.9309 (1.0035) loss 1.3864 (1.0375) grad_norm 8.9485 (8.9914/2.7515) mem 68106MB [2022-12-19 09:10:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][680/1519] eta 0:14:02 lr 0.000033 time 0.9399 (1.0044) model_time 0.9398 (1.0035) loss 1.1809 (1.0382) grad_norm 11.5725 (8.9886/2.7442) mem 68106MB [2022-12-19 09:10:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][690/1519] eta 0:13:52 lr 0.000033 time 0.9287 (1.0044) model_time 0.9286 (1.0035) loss 1.2307 (1.0396) grad_norm 7.2509 (8.9662/2.6870) mem 68106MB [2022-12-19 09:10:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][700/1519] eta 0:13:42 lr 0.000033 time 0.9325 (1.0045) model_time 0.9324 (1.0035) loss 1.1506 (1.0412) grad_norm 14.2705 (8.9980/2.7116) mem 68106MB [2022-12-19 09:10:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][710/1519] eta 0:13:32 lr 0.000033 time 0.9698 (1.0045) model_time 0.9697 (1.0036) loss 0.9213 (1.0411) grad_norm 9.4929 (8.9901/2.7034) mem 68106MB [2022-12-19 09:11:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][720/1519] eta 0:13:22 lr 0.000033 time 0.9388 (1.0045) model_time 0.9387 (1.0036) loss 1.1123 (1.0418) grad_norm 6.3066 (9.0272/2.7049) mem 68106MB [2022-12-19 09:11:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][730/1519] eta 0:13:12 lr 0.000033 time 0.9320 (1.0044) model_time 0.9319 (1.0035) loss 0.9438 (1.0412) grad_norm 11.4706 (9.0235/2.5673) mem 68106MB [2022-12-19 09:11:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][740/1519] eta 0:13:02 lr 0.000033 time 0.9380 (1.0044) model_time 0.9379 (1.0035) loss 0.9242 (1.0403) grad_norm 10.0257 (9.0277/2.5678) mem 68106MB [2022-12-19 09:11:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][750/1519] eta 0:12:52 lr 0.000033 time 0.9263 (1.0043) model_time 0.9261 (1.0034) loss 0.7578 (1.0392) grad_norm 7.7915 (9.0680/2.5998) mem 68106MB [2022-12-19 09:11:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][760/1519] eta 0:12:42 lr 0.000033 time 0.9356 (1.0042) model_time 0.9355 (1.0034) loss 1.3206 (1.0381) grad_norm 8.3244 (9.0492/2.6051) mem 68106MB [2022-12-19 09:11:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][770/1519] eta 0:12:32 lr 0.000033 time 0.9352 (1.0042) model_time 0.9346 (1.0033) loss 0.6924 (1.0375) grad_norm 6.3175 (9.0134/2.5969) mem 68106MB [2022-12-19 09:12:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][780/1519] eta 0:12:22 lr 0.000033 time 0.9303 (1.0041) model_time 0.9302 (1.0032) loss 1.1306 (1.0380) grad_norm 5.5961 (9.0168/2.6130) mem 68106MB [2022-12-19 09:12:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][790/1519] eta 0:12:11 lr 0.000033 time 0.9251 (1.0041) model_time 0.9249 (1.0032) loss 1.4073 (1.0381) grad_norm 8.8804 (9.0612/2.6898) mem 68106MB [2022-12-19 09:12:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][800/1519] eta 0:12:01 lr 0.000033 time 0.9252 (1.0041) model_time 0.9251 (1.0032) loss 1.0694 (1.0390) grad_norm 8.5394 (9.0653/2.6940) mem 68106MB [2022-12-19 09:12:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][810/1519] eta 0:11:51 lr 0.000033 time 0.9339 (1.0040) model_time 0.9338 (1.0032) loss 1.1040 (1.0393) grad_norm 12.1232 (9.1024/2.6912) mem 68106MB [2022-12-19 09:12:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][820/1519] eta 0:11:41 lr 0.000033 time 0.9314 (1.0039) model_time 0.9312 (1.0031) loss 1.0354 (1.0379) grad_norm 8.5717 (9.1289/2.7192) mem 68106MB [2022-12-19 09:12:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][830/1519] eta 0:11:31 lr 0.000033 time 0.9179 (1.0039) model_time 0.9178 (1.0031) loss 1.1258 (1.0389) grad_norm 6.6284 (9.0915/2.6882) mem 68106MB [2022-12-19 09:13:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][840/1519] eta 0:11:21 lr 0.000033 time 0.9269 (1.0042) model_time 0.9267 (1.0034) loss 0.8297 (1.0392) grad_norm 5.5841 (9.0983/2.6966) mem 68106MB [2022-12-19 09:13:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][850/1519] eta 0:11:11 lr 0.000033 time 0.9287 (1.0041) model_time 0.9286 (1.0033) loss 1.0301 (1.0397) grad_norm 11.0960 (9.1409/2.7255) mem 68106MB [2022-12-19 09:13:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][860/1519] eta 0:11:01 lr 0.000033 time 0.9301 (1.0041) model_time 0.9299 (1.0033) loss 0.8746 (1.0397) grad_norm 14.4121 (9.1481/2.7314) mem 68106MB [2022-12-19 09:13:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][870/1519] eta 0:10:51 lr 0.000033 time 0.9220 (1.0040) model_time 0.9218 (1.0032) loss 0.7704 (1.0403) grad_norm 8.0543 (9.1738/2.7248) mem 68106MB [2022-12-19 09:13:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][880/1519] eta 0:10:41 lr 0.000033 time 0.9300 (1.0040) model_time 0.9298 (1.0032) loss 1.2403 (1.0417) grad_norm 7.3039 (9.1651/2.7200) mem 68106MB [2022-12-19 09:13:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][890/1519] eta 0:10:31 lr 0.000033 time 0.9366 (1.0039) model_time 0.9365 (1.0031) loss 0.7994 (1.0413) grad_norm 18.7393 (9.2655/2.8429) mem 68106MB [2022-12-19 09:14:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][900/1519] eta 0:10:21 lr 0.000033 time 0.9285 (1.0039) model_time 0.9283 (1.0031) loss 1.2497 (1.0411) grad_norm 6.7761 (9.2807/2.8440) mem 68106MB [2022-12-19 09:14:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][910/1519] eta 0:10:11 lr 0.000033 time 1.0267 (1.0040) model_time 1.0266 (1.0032) loss 1.1444 (1.0409) grad_norm 9.2122 (9.2666/2.8225) mem 68106MB [2022-12-19 09:14:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][920/1519] eta 0:10:01 lr 0.000033 time 0.9251 (1.0039) model_time 0.9245 (1.0031) loss 1.0362 (1.0407) grad_norm 8.0526 (9.2219/2.8208) mem 68106MB [2022-12-19 09:14:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][930/1519] eta 0:09:51 lr 0.000033 time 0.9280 (1.0041) model_time 0.9279 (1.0033) loss 1.0126 (1.0417) grad_norm 6.2193 (9.2017/2.8289) mem 68106MB [2022-12-19 09:14:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][940/1519] eta 0:09:41 lr 0.000033 time 0.9387 (1.0040) model_time 0.9385 (1.0033) loss 1.3612 (1.0420) grad_norm 8.1678 (9.2473/2.8318) mem 68106MB [2022-12-19 09:14:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][950/1519] eta 0:09:31 lr 0.000033 time 0.9627 (1.0040) model_time 0.9626 (1.0033) loss 1.2350 (1.0419) grad_norm 6.9752 (9.2947/2.9078) mem 68106MB [2022-12-19 09:15:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][960/1519] eta 0:09:21 lr 0.000033 time 0.9308 (1.0040) model_time 0.9307 (1.0032) loss 1.2738 (1.0432) grad_norm 8.8525 (9.3277/2.8976) mem 68106MB [2022-12-19 09:15:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][970/1519] eta 0:09:11 lr 0.000033 time 0.9645 (1.0040) model_time 0.9644 (1.0032) loss 1.5362 (1.0436) grad_norm 10.0611 (9.3455/2.9349) mem 68106MB [2022-12-19 09:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][980/1519] eta 0:09:01 lr 0.000033 time 0.9306 (1.0040) model_time 0.9304 (1.0032) loss 0.8167 (1.0432) grad_norm 8.7945 (9.3499/2.9305) mem 68106MB [2022-12-19 09:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][990/1519] eta 0:08:51 lr 0.000033 time 0.9257 (1.0039) model_time 0.9255 (1.0032) loss 1.1648 (1.0443) grad_norm 6.2144 (9.3492/2.9369) mem 68106MB [2022-12-19 09:15:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1000/1519] eta 0:08:41 lr 0.000033 time 0.8831 (1.0040) model_time 0.8830 (1.0032) loss 0.9748 (1.0442) grad_norm 8.4040 (9.3474/2.9322) mem 68106MB [2022-12-19 09:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1010/1519] eta 0:08:30 lr 0.000033 time 0.9306 (1.0039) model_time 0.9305 (1.0031) loss 1.1798 (1.0438) grad_norm 10.0042 (9.3133/2.9128) mem 68106MB [2022-12-19 09:16:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1020/1519] eta 0:08:20 lr 0.000033 time 0.9295 (1.0039) model_time 0.9294 (1.0032) loss 0.9011 (1.0447) grad_norm 7.0675 (9.3278/2.9200) mem 68106MB [2022-12-19 09:16:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1030/1519] eta 0:08:10 lr 0.000033 time 0.9273 (1.0039) model_time 0.9271 (1.0031) loss 1.0962 (1.0457) grad_norm 9.2375 (9.3596/2.9179) mem 68106MB [2022-12-19 09:16:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1040/1519] eta 0:08:00 lr 0.000033 time 0.9312 (1.0038) model_time 0.9311 (1.0031) loss 1.0690 (1.0461) grad_norm 5.9454 (9.3504/2.9280) mem 68106MB [2022-12-19 09:16:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1050/1519] eta 0:07:50 lr 0.000033 time 0.9322 (1.0038) model_time 0.9319 (1.0030) loss 0.7900 (1.0458) grad_norm 6.1943 (9.3426/2.9531) mem 68106MB [2022-12-19 09:16:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1060/1519] eta 0:07:40 lr 0.000033 time 0.9309 (1.0038) model_time 0.9308 (1.0030) loss 1.6694 (1.0454) grad_norm 10.8242 (9.3049/2.9061) mem 68106MB [2022-12-19 09:16:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1070/1519] eta 0:07:30 lr 0.000033 time 0.9319 (1.0037) model_time 0.9317 (1.0030) loss 1.0162 (1.0460) grad_norm 14.5520 (9.3120/2.9243) mem 68106MB [2022-12-19 09:17:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1080/1519] eta 0:07:20 lr 0.000033 time 0.9295 (1.0037) model_time 0.9294 (1.0029) loss 1.2705 (1.0469) grad_norm 6.1912 (9.3242/2.9602) mem 68106MB [2022-12-19 09:17:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1090/1519] eta 0:07:10 lr 0.000033 time 0.9350 (1.0036) model_time 0.9348 (1.0029) loss 1.3418 (1.0478) grad_norm 10.0756 (9.3417/3.2485) mem 68106MB [2022-12-19 09:17:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1100/1519] eta 0:07:00 lr 0.000033 time 0.9425 (1.0036) model_time 0.9423 (1.0029) loss 0.9660 (1.0475) grad_norm 7.4272 (9.3733/3.2342) mem 68106MB [2022-12-19 09:17:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1110/1519] eta 0:06:50 lr 0.000033 time 0.9306 (1.0037) model_time 0.9305 (1.0030) loss 0.8589 (1.0475) grad_norm 14.1979 (9.3602/3.2360) mem 68106MB [2022-12-19 09:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1120/1519] eta 0:06:40 lr 0.000033 time 0.9270 (1.0037) model_time 0.9269 (1.0030) loss 0.8660 (1.0473) grad_norm 11.9728 (9.3207/3.2214) mem 68106MB [2022-12-19 09:17:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1130/1519] eta 0:06:30 lr 0.000033 time 0.9822 (1.0037) model_time 0.9820 (1.0030) loss 1.1104 (1.0476) grad_norm 11.2553 (9.3201/3.2160) mem 68106MB [2022-12-19 09:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1140/1519] eta 0:06:20 lr 0.000033 time 0.9328 (1.0037) model_time 0.9327 (1.0030) loss 1.5971 (1.0469) grad_norm 7.6059 (9.3095/3.1798) mem 68106MB [2022-12-19 09:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1150/1519] eta 0:06:10 lr 0.000033 time 0.9309 (1.0038) model_time 0.9308 (1.0031) loss 0.8491 (1.0471) grad_norm 16.2035 (9.3103/3.1871) mem 68106MB [2022-12-19 09:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1160/1519] eta 0:06:00 lr 0.000033 time 0.9333 (1.0037) model_time 0.9332 (1.0030) loss 0.9217 (1.0470) grad_norm 6.0210 (9.3125/3.2013) mem 68106MB [2022-12-19 09:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1170/1519] eta 0:05:50 lr 0.000033 time 0.9313 (1.0037) model_time 0.9311 (1.0030) loss 1.1387 (1.0478) grad_norm 6.6103 (9.3171/3.1989) mem 68106MB [2022-12-19 09:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1180/1519] eta 0:05:40 lr 0.000033 time 0.9342 (1.0036) model_time 0.9340 (1.0029) loss 0.8194 (1.0476) grad_norm 10.0962 (9.3042/3.2071) mem 68106MB [2022-12-19 09:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1190/1519] eta 0:05:30 lr 0.000033 time 0.9274 (1.0036) model_time 0.9272 (1.0029) loss 1.2128 (1.0482) grad_norm 11.6024 (9.3950/3.2512) mem 68106MB [2022-12-19 09:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1200/1519] eta 0:05:20 lr 0.000033 time 0.9348 (1.0035) model_time 0.9346 (1.0028) loss 0.9661 (1.0486) grad_norm 11.5048 (9.3832/3.2391) mem 68106MB [2022-12-19 09:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1210/1519] eta 0:05:10 lr 0.000033 time 0.9301 (1.0035) model_time 0.9299 (1.0028) loss 1.5919 (1.0483) grad_norm 8.4074 (9.3870/3.2348) mem 68106MB [2022-12-19 09:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1220/1519] eta 0:05:00 lr 0.000033 time 0.9270 (1.0035) model_time 0.9268 (1.0028) loss 1.1681 (1.0484) grad_norm 8.8022 (9.3730/3.2284) mem 68106MB [2022-12-19 09:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1230/1519] eta 0:04:50 lr 0.000033 time 0.9287 (1.0037) model_time 0.9285 (1.0030) loss 1.0285 (1.0490) grad_norm 9.4116 (9.3623/3.1988) mem 68106MB [2022-12-19 09:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1240/1519] eta 0:04:40 lr 0.000033 time 1.0197 (1.0037) model_time 1.0196 (1.0030) loss 1.2406 (1.0490) grad_norm 13.2799 (9.3878/3.2032) mem 68106MB [2022-12-19 09:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1250/1519] eta 0:04:30 lr 0.000033 time 0.9299 (1.0037) model_time 0.9298 (1.0030) loss 1.2701 (1.0496) grad_norm 6.4104 (9.3456/3.1755) mem 68106MB [2022-12-19 09:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1260/1519] eta 0:04:19 lr 0.000033 time 0.9320 (1.0037) model_time 0.9319 (1.0030) loss 1.1974 (1.0501) grad_norm 9.9889 (9.2774/3.1360) mem 68106MB [2022-12-19 09:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1270/1519] eta 0:04:09 lr 0.000033 time 0.9328 (1.0036) model_time 0.9327 (1.0030) loss 1.0400 (1.0505) grad_norm 6.7526 (9.2739/3.1453) mem 68106MB [2022-12-19 09:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1280/1519] eta 0:03:59 lr 0.000033 time 0.9316 (1.0036) model_time 0.9315 (1.0029) loss 1.0174 (1.0507) grad_norm 7.6308 (9.2964/3.1545) mem 68106MB [2022-12-19 09:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1290/1519] eta 0:03:49 lr 0.000033 time 0.9300 (1.0037) model_time 0.9299 (1.0031) loss 0.9855 (1.0504) grad_norm 11.1602 (9.3101/3.1532) mem 68106MB [2022-12-19 09:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1300/1519] eta 0:03:39 lr 0.000033 time 0.9300 (1.0037) model_time 0.9299 (1.0030) loss 1.2850 (1.0508) grad_norm 6.8807 (9.2778/3.1256) mem 68106MB [2022-12-19 09:20:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1310/1519] eta 0:03:29 lr 0.000033 time 0.9296 (1.0037) model_time 0.9294 (1.0030) loss 0.9029 (1.0508) grad_norm 6.7756 (9.2809/3.1251) mem 68106MB [2022-12-19 09:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1320/1519] eta 0:03:19 lr 0.000033 time 0.9316 (1.0037) model_time 0.9315 (1.0030) loss 0.7977 (1.0511) grad_norm 7.3660 (9.2246/3.1057) mem 68106MB [2022-12-19 09:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1330/1519] eta 0:03:09 lr 0.000033 time 0.9815 (1.0037) model_time 0.9813 (1.0030) loss 1.4837 (1.0512) grad_norm 10.7286 (9.2231/3.1094) mem 68106MB [2022-12-19 09:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1340/1519] eta 0:02:59 lr 0.000033 time 0.9325 (1.0036) model_time 0.9324 (1.0030) loss 0.8850 (1.0521) grad_norm 7.8110 (9.2526/3.1230) mem 68106MB [2022-12-19 09:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1350/1519] eta 0:02:49 lr 0.000033 time 0.9248 (1.0036) model_time 0.9247 (1.0029) loss 1.1962 (1.0520) grad_norm 7.4867 (9.2722/3.1177) mem 68106MB [2022-12-19 09:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1360/1519] eta 0:02:39 lr 0.000033 time 0.9433 (1.0036) model_time 0.9432 (1.0029) loss 1.1321 (1.0514) grad_norm 8.2357 (9.2824/3.1083) mem 68106MB [2022-12-19 09:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1370/1519] eta 0:02:29 lr 0.000033 time 0.9283 (1.0036) model_time 0.9281 (1.0029) loss 0.7604 (1.0506) grad_norm 9.1917 (9.2939/3.1065) mem 68106MB [2022-12-19 09:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1380/1519] eta 0:02:19 lr 0.000033 time 0.9330 (1.0035) model_time 0.9328 (1.0029) loss 1.1111 (1.0504) grad_norm 8.4096 (9.2925/3.0857) mem 68106MB [2022-12-19 09:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1390/1519] eta 0:02:09 lr 0.000033 time 0.9295 (1.0035) model_time 0.9294 (1.0028) loss 0.8388 (1.0505) grad_norm 10.5089 (9.2720/3.0209) mem 68106MB [2022-12-19 09:22:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1400/1519] eta 0:01:59 lr 0.000033 time 0.9267 (1.0035) model_time 0.9260 (1.0028) loss 0.7997 (1.0503) grad_norm 11.1981 (9.2856/3.0139) mem 68106MB [2022-12-19 09:22:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1410/1519] eta 0:01:49 lr 0.000033 time 0.9308 (1.0035) model_time 0.9307 (1.0028) loss 1.1034 (1.0502) grad_norm 6.2656 (9.2469/3.0245) mem 68106MB [2022-12-19 09:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1420/1519] eta 0:01:39 lr 0.000033 time 1.0315 (1.0035) model_time 1.0313 (1.0029) loss 1.0991 (1.0501) grad_norm 5.9602 (9.1958/2.9952) mem 68106MB [2022-12-19 09:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1430/1519] eta 0:01:29 lr 0.000033 time 0.9308 (1.0035) model_time 0.9306 (1.0028) loss 0.9550 (1.0498) grad_norm 6.4211 (9.1839/2.9981) mem 68106MB [2022-12-19 09:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1440/1519] eta 0:01:19 lr 0.000033 time 0.9314 (1.0034) model_time 0.9313 (1.0028) loss 0.7631 (1.0497) grad_norm 9.3396 (9.2086/2.9903) mem 68106MB [2022-12-19 09:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1450/1519] eta 0:01:09 lr 0.000033 time 0.9137 (1.0035) model_time 0.9136 (1.0029) loss 1.0688 (1.0499) grad_norm 5.5142 (9.1532/2.9634) mem 68106MB [2022-12-19 09:23:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1460/1519] eta 0:00:59 lr 0.000033 time 0.9301 (1.0035) model_time 0.9299 (1.0028) loss 0.9489 (1.0496) grad_norm 16.5269 (9.1672/2.9952) mem 68106MB [2022-12-19 09:23:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1470/1519] eta 0:00:49 lr 0.000033 time 0.9338 (1.0036) model_time 0.9337 (1.0029) loss 1.5239 (1.0501) grad_norm 5.1523 (9.1742/3.0182) mem 68106MB [2022-12-19 09:23:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1480/1519] eta 0:00:39 lr 0.000033 time 0.9284 (1.0035) model_time 0.9283 (1.0029) loss 0.7875 (1.0497) grad_norm 10.7121 (9.1963/3.0181) mem 68106MB [2022-12-19 09:23:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1490/1519] eta 0:00:29 lr 0.000033 time 0.9336 (1.0035) model_time 0.9335 (1.0029) loss 0.7189 (1.0495) grad_norm 10.2959 (9.1312/2.8996) mem 68106MB [2022-12-19 09:24:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1500/1519] eta 0:00:19 lr 0.000033 time 0.9358 (1.0035) model_time 0.9357 (1.0029) loss 1.6779 (1.0502) grad_norm 7.4369 (9.1074/2.8928) mem 68106MB [2022-12-19 09:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [11/100][1510/1519] eta 0:00:09 lr 0.000033 time 0.9887 (1.0035) model_time 0.9886 (1.0029) loss 1.3234 (1.0501) grad_norm 13.6751 (9.1031/2.8956) mem 68106MB [2022-12-19 09:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 11 training takes 0:25:24 [2022-12-19 09:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_11.pth saving...... [2022-12-19 09:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_11.pth saved !!! [2022-12-19 09:24:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.647 (0.647) Loss 0.7563 (0.7563) Acc@1 84.028 (84.028) Acc@5 96.528 (96.528) Mem 68106MB [2022-12-19 09:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.330) Loss 0.7485 (0.7312) Acc@1 88.194 (85.322) Acc@5 96.875 (97.096) Mem 68106MB [2022-12-19 09:24:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.315) Loss 0.6874 (0.7300) Acc@1 86.458 (85.169) Acc@5 96.875 (96.991) Mem 68106MB [2022-12-19 09:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 0.7739 (0.7344) Acc@1 86.111 (85.204) Acc@5 96.528 (96.953) Mem 68106MB [2022-12-19 09:24:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.294 (0.307) Loss 0.7160 (0.7256) Acc@1 83.681 (85.425) Acc@5 96.181 (97.078) Mem 68106MB [2022-12-19 09:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.306) Loss 0.7405 (0.7203) Acc@1 84.375 (85.539) Acc@5 96.875 (97.127) Mem 68106MB [2022-12-19 09:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.7329 (0.7203) Acc@1 85.764 (85.548) Acc@5 96.181 (97.120) Mem 68106MB [2022-12-19 09:25:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.8059 (0.7226) Acc@1 85.764 (85.514) Acc@5 96.528 (97.124) Mem 68106MB [2022-12-19 09:25:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.305 (0.303) Loss 0.6663 (0.7221) Acc@1 86.111 (85.511) Acc@5 96.875 (97.119) Mem 68106MB [2022-12-19 09:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:11] * Acc@1 85.531 Acc@5 97.151 [2022-12-19 09:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 85.5% [2022-12-19 09:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 09:25:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 09:25:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 85.53% [2022-12-19 09:25:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][0/1519] eta 0:35:02 lr 0.000033 time 1.3843 (1.3843) model_time 0.9347 (0.9347) loss 0.8599 (0.8599) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 09:25:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][10/1519] eta 0:25:58 lr 0.000033 time 0.9240 (1.0328) model_time 0.9238 (0.9916) loss 1.2003 (0.9482) grad_norm 25.8301 (11.7239/7.1049) mem 68106MB [2022-12-19 09:25:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][20/1519] eta 0:25:24 lr 0.000033 time 0.9307 (1.0171) model_time 0.9305 (0.9953) loss 1.2724 (0.9972) grad_norm 13.7744 (10.7854/5.4531) mem 68106MB [2022-12-19 09:26:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][30/1519] eta 0:25:05 lr 0.000033 time 0.9204 (1.0111) model_time 0.9202 (0.9962) loss 1.0525 (0.9846) grad_norm 8.0146 (10.8892/4.7486) mem 68106MB [2022-12-19 09:26:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][40/1519] eta 0:24:51 lr 0.000033 time 0.9309 (1.0084) model_time 0.9307 (0.9971) loss 0.9831 (0.9986) grad_norm 8.4728 (10.8093/4.3804) mem 68106MB [2022-12-19 09:26:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][50/1519] eta 0:24:40 lr 0.000033 time 0.9290 (1.0079) model_time 0.9288 (0.9987) loss 0.9722 (1.0113) grad_norm 8.5099 (10.8099/4.1737) mem 68106MB [2022-12-19 09:26:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][60/1519] eta 0:24:28 lr 0.000033 time 0.9331 (1.0062) model_time 0.9330 (0.9984) loss 0.8824 (1.0104) grad_norm 12.3631 (10.6588/3.9574) mem 68106MB [2022-12-19 09:26:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][70/1519] eta 0:24:15 lr 0.000033 time 0.9375 (1.0048) model_time 0.9374 (0.9981) loss 1.1292 (1.0024) grad_norm 8.2615 (10.6491/3.8541) mem 68106MB [2022-12-19 09:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][80/1519] eta 0:24:07 lr 0.000033 time 1.0284 (1.0058) model_time 1.0282 (0.9998) loss 1.2960 (0.9988) grad_norm 12.0466 (10.4981/3.6737) mem 68106MB [2022-12-19 09:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][90/1519] eta 0:23:57 lr 0.000033 time 0.9241 (1.0060) model_time 0.9240 (1.0006) loss 0.8578 (1.0073) grad_norm 9.1692 (10.3378/3.5869) mem 68106MB [2022-12-19 09:27:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][100/1519] eta 0:23:46 lr 0.000033 time 0.9202 (1.0052) model_time 0.9199 (1.0004) loss 1.1688 (1.0062) grad_norm 9.8555 (10.1479/3.4875) mem 68106MB [2022-12-19 09:27:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][110/1519] eta 0:23:35 lr 0.000033 time 0.9292 (1.0048) model_time 0.9290 (1.0003) loss 1.5849 (1.0111) grad_norm 8.2446 (10.0512/3.4103) mem 68106MB [2022-12-19 09:27:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][120/1519] eta 0:23:24 lr 0.000033 time 0.9017 (1.0041) model_time 0.9015 (1.0000) loss 0.7952 (1.0118) grad_norm 8.5243 (9.9534/3.3303) mem 68106MB [2022-12-19 09:27:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][130/1519] eta 0:23:13 lr 0.000033 time 0.8974 (1.0036) model_time 0.8970 (0.9998) loss 1.3381 (1.0142) grad_norm 6.9075 (9.8933/3.2655) mem 68106MB [2022-12-19 09:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][140/1519] eta 0:23:03 lr 0.000033 time 0.9292 (1.0036) model_time 0.9289 (1.0000) loss 0.9629 (1.0099) grad_norm 8.6563 (9.8121/3.2126) mem 68106MB [2022-12-19 09:28:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][150/1519] eta 0:22:53 lr 0.000033 time 0.9217 (1.0033) model_time 0.9216 (0.9999) loss 1.0474 (1.0033) grad_norm 8.1181 (9.7219/3.1319) mem 68106MB [2022-12-19 09:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][160/1519] eta 0:22:42 lr 0.000033 time 0.9249 (1.0029) model_time 0.9247 (0.9997) loss 1.3168 (1.0091) grad_norm 9.0944 (9.6700/3.0777) mem 68106MB [2022-12-19 09:28:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][170/1519] eta 0:22:32 lr 0.000033 time 0.9276 (1.0026) model_time 0.9274 (0.9996) loss 1.2885 (1.0112) grad_norm 6.5421 (9.6814/3.0903) mem 68106MB [2022-12-19 09:28:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][180/1519] eta 0:22:22 lr 0.000033 time 0.9313 (1.0025) model_time 0.9312 (0.9996) loss 0.9769 (1.0066) grad_norm 6.3240 (9.4822/3.1155) mem 68106MB [2022-12-19 09:28:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][190/1519] eta 0:22:12 lr 0.000033 time 0.9258 (1.0023) model_time 0.9257 (0.9996) loss 0.7688 (1.0068) grad_norm 7.5839 (9.4656/3.0691) mem 68106MB [2022-12-19 09:28:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][200/1519] eta 0:22:02 lr 0.000033 time 0.9315 (1.0024) model_time 0.9313 (0.9998) loss 1.2668 (1.0128) grad_norm 7.0063 (9.4091/3.0179) mem 68106MB [2022-12-19 09:29:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][210/1519] eta 0:21:52 lr 0.000033 time 0.9329 (1.0025) model_time 0.9328 (1.0000) loss 1.0507 (1.0137) grad_norm 9.4712 (9.3435/2.9712) mem 68106MB [2022-12-19 09:29:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][220/1519] eta 0:21:42 lr 0.000033 time 0.9304 (1.0030) model_time 0.9303 (1.0005) loss 1.2000 (1.0166) grad_norm 18.3027 (9.4260/3.0560) mem 68106MB [2022-12-19 09:29:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][230/1519] eta 0:21:32 lr 0.000033 time 0.9313 (1.0029) model_time 0.9310 (1.0006) loss 0.7542 (1.0153) grad_norm 7.2866 (9.4315/3.0173) mem 68106MB [2022-12-19 09:29:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][240/1519] eta 0:21:22 lr 0.000033 time 0.9326 (1.0029) model_time 0.9325 (1.0006) loss 0.8274 (1.0159) grad_norm 7.4827 (9.3514/2.9934) mem 68106MB [2022-12-19 09:29:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][250/1519] eta 0:21:12 lr 0.000033 time 0.9282 (1.0027) model_time 0.9281 (1.0005) loss 1.1687 (1.0219) grad_norm 7.5688 (9.3371/2.9587) mem 68106MB [2022-12-19 09:29:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][260/1519] eta 0:21:04 lr 0.000033 time 1.1737 (1.0042) model_time 1.1735 (1.0021) loss 0.8012 (1.0276) grad_norm 13.7978 (9.3338/2.9386) mem 68106MB [2022-12-19 09:30:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][270/1519] eta 0:20:54 lr 0.000033 time 0.9381 (1.0046) model_time 0.9379 (1.0025) loss 1.2861 (1.0289) grad_norm 9.3334 (9.3513/2.9058) mem 68106MB [2022-12-19 09:30:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][280/1519] eta 0:20:44 lr 0.000033 time 0.9444 (1.0045) model_time 0.9443 (1.0025) loss 1.0565 (1.0281) grad_norm 7.4799 (9.2801/2.8799) mem 68106MB [2022-12-19 09:30:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][290/1519] eta 0:20:34 lr 0.000033 time 0.9298 (1.0044) model_time 0.9296 (1.0024) loss 1.5024 (1.0298) grad_norm 6.6925 (9.2800/2.9288) mem 68106MB [2022-12-19 09:30:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][300/1519] eta 0:20:24 lr 0.000033 time 0.9343 (1.0043) model_time 0.9341 (1.0024) loss 1.1583 (1.0282) grad_norm 7.2955 (9.2562/2.8900) mem 68106MB [2022-12-19 09:30:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][310/1519] eta 0:20:14 lr 0.000033 time 0.9303 (1.0044) model_time 0.9301 (1.0025) loss 1.2967 (1.0286) grad_norm 9.1175 (9.2902/2.8688) mem 68106MB [2022-12-19 09:30:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][320/1519] eta 0:20:04 lr 0.000033 time 0.9298 (1.0042) model_time 0.9296 (1.0024) loss 0.7462 (1.0275) grad_norm 9.9704 (9.3066/2.8756) mem 68106MB [2022-12-19 09:31:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][330/1519] eta 0:19:53 lr 0.000033 time 0.9308 (1.0040) model_time 0.9304 (1.0022) loss 1.2243 (1.0283) grad_norm 11.4311 (9.3032/2.8412) mem 68106MB [2022-12-19 09:31:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][340/1519] eta 0:19:43 lr 0.000033 time 0.9342 (1.0038) model_time 0.9341 (1.0020) loss 0.9462 (1.0280) grad_norm 10.2858 (9.3226/2.8227) mem 68106MB [2022-12-19 09:31:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][350/1519] eta 0:19:33 lr 0.000033 time 0.9423 (1.0037) model_time 0.9421 (1.0020) loss 0.8273 (1.0279) grad_norm 12.2233 (9.3165/2.8354) mem 68106MB [2022-12-19 09:31:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][360/1519] eta 0:19:23 lr 0.000033 time 0.9294 (1.0035) model_time 0.9293 (1.0018) loss 0.7347 (1.0252) grad_norm 14.0459 (9.3007/2.8401) mem 68106MB [2022-12-19 09:31:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][370/1519] eta 0:19:12 lr 0.000033 time 0.9178 (1.0033) model_time 0.9176 (1.0017) loss 1.2903 (1.0278) grad_norm 7.9283 (9.3218/2.8266) mem 68106MB [2022-12-19 09:31:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][380/1519] eta 0:19:02 lr 0.000033 time 0.9296 (1.0034) model_time 0.9294 (1.0018) loss 1.2817 (1.0272) grad_norm 9.6073 (9.3019/2.7953) mem 68106MB [2022-12-19 09:32:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][390/1519] eta 0:18:52 lr 0.000033 time 0.9275 (1.0033) model_time 0.9273 (1.0017) loss 1.0954 (1.0279) grad_norm 8.8875 (9.3323/2.8018) mem 68106MB [2022-12-19 09:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][400/1519] eta 0:18:43 lr 0.000033 time 0.9321 (1.0037) model_time 0.9319 (1.0022) loss 0.8776 (1.0266) grad_norm 10.3124 (9.3909/2.8317) mem 68106MB [2022-12-19 09:32:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][410/1519] eta 0:18:32 lr 0.000033 time 0.9314 (1.0036) model_time 0.9312 (1.0021) loss 1.1163 (1.0282) grad_norm 9.4202 (9.3952/2.8061) mem 68106MB [2022-12-19 09:32:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][420/1519] eta 0:18:22 lr 0.000033 time 0.9301 (1.0035) model_time 0.9299 (1.0020) loss 0.7284 (1.0285) grad_norm 7.2383 (9.4298/2.8384) mem 68106MB [2022-12-19 09:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][430/1519] eta 0:18:12 lr 0.000033 time 0.9326 (1.0033) model_time 0.9324 (1.0019) loss 1.1533 (1.0264) grad_norm 9.6947 (9.4248/2.8351) mem 68106MB [2022-12-19 09:32:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][440/1519] eta 0:18:02 lr 0.000033 time 0.9289 (1.0033) model_time 0.9288 (1.0019) loss 1.1536 (1.0274) grad_norm 7.5037 (9.3857/2.8211) mem 68106MB [2022-12-19 09:33:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][450/1519] eta 0:17:53 lr 0.000033 time 0.9325 (1.0042) model_time 0.9323 (1.0028) loss 0.8018 (1.0275) grad_norm 9.0672 (9.3799/2.7920) mem 68106MB [2022-12-19 09:33:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][460/1519] eta 0:17:43 lr 0.000033 time 0.9385 (1.0041) model_time 0.9384 (1.0027) loss 0.8532 (1.0267) grad_norm 16.4773 (9.3969/2.8052) mem 68106MB [2022-12-19 09:33:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][470/1519] eta 0:17:33 lr 0.000033 time 0.9293 (1.0040) model_time 0.9292 (1.0026) loss 0.7795 (1.0293) grad_norm 14.2758 (9.4699/2.8651) mem 68106MB [2022-12-19 09:33:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][480/1519] eta 0:17:22 lr 0.000033 time 0.9296 (1.0038) model_time 0.9295 (1.0025) loss 1.0977 (1.0317) grad_norm 13.3941 (9.5052/2.8942) mem 68106MB [2022-12-19 09:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][490/1519] eta 0:17:12 lr 0.000033 time 0.9298 (1.0037) model_time 0.9296 (1.0024) loss 1.1708 (1.0337) grad_norm 8.3553 (9.5935/3.0550) mem 68106MB [2022-12-19 09:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][500/1519] eta 0:17:02 lr 0.000033 time 0.9317 (1.0036) model_time 0.9316 (1.0023) loss 0.8346 (1.0333) grad_norm 6.7292 (9.5839/3.0484) mem 68106MB [2022-12-19 09:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][510/1519] eta 0:16:52 lr 0.000033 time 0.9305 (1.0035) model_time 0.9304 (1.0023) loss 1.1166 (1.0326) grad_norm 11.0089 (9.5673/3.0283) mem 68106MB [2022-12-19 09:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][520/1519] eta 0:16:42 lr 0.000033 time 0.9619 (1.0035) model_time 0.9618 (1.0022) loss 0.9897 (1.0324) grad_norm 6.6607 (9.5107/3.0281) mem 68106MB [2022-12-19 09:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][530/1519] eta 0:16:32 lr 0.000033 time 0.9328 (1.0034) model_time 0.9327 (1.0022) loss 1.3971 (1.0337) grad_norm 6.9028 (9.5307/3.0324) mem 68106MB [2022-12-19 09:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][540/1519] eta 0:16:22 lr 0.000033 time 0.9327 (1.0034) model_time 0.9326 (1.0021) loss 1.0205 (1.0334) grad_norm 10.0629 (9.5105/3.0146) mem 68106MB [2022-12-19 09:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][550/1519] eta 0:16:12 lr 0.000033 time 0.9337 (1.0033) model_time 0.9336 (1.0021) loss 1.0902 (1.0320) grad_norm 7.6520 (9.5167/3.0134) mem 68106MB [2022-12-19 09:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][560/1519] eta 0:16:02 lr 0.000033 time 0.9314 (1.0032) model_time 0.9313 (1.0020) loss 1.1610 (1.0321) grad_norm 9.1079 (9.5216/2.9965) mem 68106MB [2022-12-19 09:35:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][570/1519] eta 0:15:52 lr 0.000033 time 0.9398 (1.0036) model_time 0.9396 (1.0024) loss 0.8686 (1.0320) grad_norm 9.7192 (9.4972/2.9816) mem 68106MB [2022-12-19 09:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][580/1519] eta 0:15:42 lr 0.000033 time 0.9217 (1.0037) model_time 0.9210 (1.0025) loss 0.9139 (1.0313) grad_norm 6.5442 (9.4884/2.9799) mem 68106MB [2022-12-19 09:35:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][590/1519] eta 0:15:32 lr 0.000033 time 0.9287 (1.0036) model_time 0.9286 (1.0024) loss 0.7857 (1.0303) grad_norm 7.6751 (9.4717/2.9765) mem 68106MB [2022-12-19 09:35:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][600/1519] eta 0:15:22 lr 0.000032 time 0.9266 (1.0036) model_time 0.9265 (1.0024) loss 1.1250 (1.0291) grad_norm 10.7139 (9.4544/2.9609) mem 68106MB [2022-12-19 09:35:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][610/1519] eta 0:15:12 lr 0.000032 time 0.9803 (1.0035) model_time 0.9801 (1.0024) loss 1.0916 (1.0288) grad_norm 6.8693 (9.4325/2.8324) mem 68106MB [2022-12-19 09:35:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][620/1519] eta 0:15:02 lr 0.000032 time 0.9274 (1.0035) model_time 0.9272 (1.0024) loss 0.9679 (1.0301) grad_norm 7.7880 (9.4251/2.9044) mem 68106MB [2022-12-19 09:36:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][630/1519] eta 0:14:52 lr 0.000032 time 0.9309 (1.0034) model_time 0.9307 (1.0023) loss 0.8043 (1.0296) grad_norm 9.5706 (9.3996/2.9051) mem 68106MB [2022-12-19 09:36:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][640/1519] eta 0:14:41 lr 0.000032 time 0.9349 (1.0034) model_time 0.9348 (1.0023) loss 1.1888 (1.0287) grad_norm 7.3908 (9.3837/2.8830) mem 68106MB [2022-12-19 09:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][650/1519] eta 0:14:31 lr 0.000032 time 0.9342 (1.0033) model_time 0.9338 (1.0022) loss 0.9275 (1.0287) grad_norm 7.9474 (9.3386/2.8592) mem 68106MB [2022-12-19 09:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][660/1519] eta 0:14:21 lr 0.000032 time 0.9279 (1.0033) model_time 0.9278 (1.0022) loss 0.9990 (1.0300) grad_norm 8.6820 (9.3423/2.8491) mem 68106MB [2022-12-19 09:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][670/1519] eta 0:14:11 lr 0.000032 time 0.9274 (1.0032) model_time 0.9272 (1.0021) loss 1.6000 (1.0306) grad_norm 6.8428 (9.2936/2.8295) mem 68106MB [2022-12-19 09:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][680/1519] eta 0:14:01 lr 0.000032 time 0.9314 (1.0031) model_time 0.9313 (1.0021) loss 0.9378 (1.0296) grad_norm 11.7258 (9.3037/2.8298) mem 68106MB [2022-12-19 09:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][690/1519] eta 0:13:51 lr 0.000032 time 0.9219 (1.0035) model_time 0.9218 (1.0025) loss 1.4336 (1.0303) grad_norm 7.8369 (9.3234/2.8352) mem 68106MB [2022-12-19 09:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][700/1519] eta 0:13:41 lr 0.000032 time 0.9331 (1.0035) model_time 0.9329 (1.0025) loss 1.2612 (1.0305) grad_norm 11.1616 (9.3265/2.8329) mem 68106MB [2022-12-19 09:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][710/1519] eta 0:13:32 lr 0.000032 time 0.9324 (1.0038) model_time 0.9271 (1.0028) loss 1.0070 (1.0304) grad_norm 9.4624 (9.3082/2.8272) mem 68106MB [2022-12-19 09:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][720/1519] eta 0:13:21 lr 0.000032 time 0.9369 (1.0038) model_time 0.9367 (1.0027) loss 0.8662 (1.0299) grad_norm 10.6537 (9.2922/2.8324) mem 68106MB [2022-12-19 09:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][730/1519] eta 0:13:11 lr 0.000032 time 0.9280 (1.0037) model_time 0.9279 (1.0027) loss 0.7848 (1.0287) grad_norm 6.5052 (9.2739/2.8277) mem 68106MB [2022-12-19 09:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][740/1519] eta 0:13:01 lr 0.000032 time 0.9279 (1.0036) model_time 0.9278 (1.0026) loss 1.2671 (1.0299) grad_norm 14.4102 (9.2948/2.8395) mem 68106MB [2022-12-19 09:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][750/1519] eta 0:12:51 lr 0.000032 time 0.9132 (1.0036) model_time 0.9130 (1.0026) loss 1.1760 (1.0296) grad_norm 11.2209 (9.3051/2.8429) mem 68106MB [2022-12-19 09:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][760/1519] eta 0:12:41 lr 0.000032 time 0.9498 (1.0037) model_time 0.9496 (1.0027) loss 0.8699 (1.0293) grad_norm 6.5662 (9.3042/2.8447) mem 68106MB [2022-12-19 09:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][770/1519] eta 0:12:31 lr 0.000032 time 0.9265 (1.0037) model_time 0.9264 (1.0027) loss 0.7404 (1.0293) grad_norm 13.0562 (9.2953/2.8317) mem 68106MB [2022-12-19 09:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][780/1519] eta 0:12:21 lr 0.000032 time 0.9367 (1.0037) model_time 0.9365 (1.0027) loss 0.9137 (1.0302) grad_norm 8.1337 (9.3398/2.8185) mem 68106MB [2022-12-19 09:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][790/1519] eta 0:12:11 lr 0.000032 time 1.0189 (1.0037) model_time 1.0187 (1.0027) loss 0.8843 (1.0302) grad_norm 4.5026 (9.3036/2.8282) mem 68106MB [2022-12-19 09:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][800/1519] eta 0:12:01 lr 0.000032 time 0.9307 (1.0036) model_time 0.9306 (1.0027) loss 1.0354 (1.0305) grad_norm 9.4347 (9.3061/2.8249) mem 68106MB [2022-12-19 09:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][810/1519] eta 0:11:51 lr 0.000032 time 0.9302 (1.0036) model_time 0.9301 (1.0026) loss 1.1491 (1.0319) grad_norm 7.5772 (9.3288/2.8393) mem 68106MB [2022-12-19 09:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][820/1519] eta 0:11:41 lr 0.000032 time 0.9354 (1.0035) model_time 0.9353 (1.0025) loss 0.9276 (1.0317) grad_norm 15.8344 (9.3249/2.8166) mem 68106MB [2022-12-19 09:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][830/1519] eta 0:11:31 lr 0.000032 time 0.9321 (1.0035) model_time 0.9320 (1.0025) loss 0.8337 (1.0315) grad_norm 13.1000 (9.3165/2.8231) mem 68106MB [2022-12-19 09:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][840/1519] eta 0:11:21 lr 0.000032 time 0.9317 (1.0034) model_time 0.9315 (1.0025) loss 1.0999 (1.0315) grad_norm 5.4960 (9.3337/2.8308) mem 68106MB [2022-12-19 09:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][850/1519] eta 0:11:11 lr 0.000032 time 0.9261 (1.0034) model_time 0.9260 (1.0024) loss 0.7843 (1.0305) grad_norm 12.6229 (9.3506/2.8468) mem 68106MB [2022-12-19 09:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][860/1519] eta 0:11:01 lr 0.000032 time 0.9374 (1.0034) model_time 0.9372 (1.0024) loss 0.8847 (1.0303) grad_norm 17.2772 (9.3811/2.9094) mem 68106MB [2022-12-19 09:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][870/1519] eta 0:10:51 lr 0.000032 time 0.9402 (1.0034) model_time 0.9401 (1.0024) loss 0.8372 (1.0298) grad_norm 6.5907 (9.3836/2.9848) mem 68106MB [2022-12-19 09:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][880/1519] eta 0:10:41 lr 0.000032 time 1.0224 (1.0035) model_time 1.0223 (1.0025) loss 0.8298 (1.0294) grad_norm 8.0476 (9.4116/2.9855) mem 68106MB [2022-12-19 09:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][890/1519] eta 0:10:31 lr 0.000032 time 0.9327 (1.0035) model_time 0.9325 (1.0026) loss 0.8363 (1.0303) grad_norm 7.3341 (9.4086/2.9552) mem 68106MB [2022-12-19 09:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][900/1519] eta 0:10:21 lr 0.000032 time 0.9327 (1.0034) model_time 0.9325 (1.0025) loss 0.8425 (1.0297) grad_norm 6.0690 (9.4009/2.9582) mem 68106MB [2022-12-19 09:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][910/1519] eta 0:10:11 lr 0.000032 time 0.9371 (1.0034) model_time 0.9370 (1.0025) loss 0.7839 (1.0320) grad_norm 5.9439 (9.4499/3.0324) mem 68106MB [2022-12-19 09:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][920/1519] eta 0:10:01 lr 0.000032 time 0.9304 (1.0034) model_time 0.9303 (1.0025) loss 0.7302 (1.0312) grad_norm 7.4717 (9.3945/3.0311) mem 68106MB [2022-12-19 09:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][930/1519] eta 0:09:51 lr 0.000032 time 0.8865 (1.0035) model_time 0.8863 (1.0026) loss 0.9061 (1.0312) grad_norm 7.4320 (9.4072/3.0798) mem 68106MB [2022-12-19 09:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][940/1519] eta 0:09:41 lr 0.000032 time 0.9452 (1.0035) model_time 0.9451 (1.0026) loss 0.8683 (1.0307) grad_norm 8.4476 (9.3785/3.0732) mem 68106MB [2022-12-19 09:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][950/1519] eta 0:09:30 lr 0.000032 time 0.9436 (1.0035) model_time 0.9434 (1.0026) loss 0.7233 (1.0304) grad_norm 14.2061 (9.4261/3.1105) mem 68106MB [2022-12-19 09:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][960/1519] eta 0:09:20 lr 0.000032 time 0.9283 (1.0035) model_time 0.9281 (1.0026) loss 0.9678 (1.0308) grad_norm 5.6740 (9.4535/3.1247) mem 68106MB [2022-12-19 09:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][970/1519] eta 0:09:10 lr 0.000032 time 0.9347 (1.0034) model_time 0.9345 (1.0025) loss 0.8509 (1.0303) grad_norm 5.0705 (9.4387/3.1457) mem 68106MB [2022-12-19 09:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][980/1519] eta 0:09:00 lr 0.000032 time 0.9318 (1.0034) model_time 0.9316 (1.0025) loss 1.3535 (1.0308) grad_norm 14.6693 (9.4642/3.1871) mem 68106MB [2022-12-19 09:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][990/1519] eta 0:08:50 lr 0.000032 time 0.9266 (1.0034) model_time 0.9265 (1.0025) loss 1.2063 (1.0305) grad_norm 8.7231 (9.4484/3.1860) mem 68106MB [2022-12-19 09:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1000/1519] eta 0:08:40 lr 0.000032 time 0.9226 (1.0034) model_time 0.9225 (1.0025) loss 0.9883 (1.0299) grad_norm 8.7154 (9.3824/3.1597) mem 68106MB [2022-12-19 09:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1010/1519] eta 0:08:30 lr 0.000032 time 0.9254 (1.0034) model_time 0.9253 (1.0025) loss 0.7315 (1.0298) grad_norm 7.7172 (9.3896/3.1598) mem 68106MB [2022-12-19 09:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1020/1519] eta 0:08:20 lr 0.000032 time 1.1741 (1.0036) model_time 1.1739 (1.0028) loss 0.8067 (1.0294) grad_norm 7.9135 (9.3527/3.1224) mem 68106MB [2022-12-19 09:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1030/1519] eta 0:08:10 lr 0.000032 time 0.9290 (1.0036) model_time 0.9288 (1.0028) loss 0.9217 (1.0291) grad_norm 9.5313 (9.3447/3.1065) mem 68106MB [2022-12-19 09:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1040/1519] eta 0:08:00 lr 0.000032 time 0.9348 (1.0036) model_time 0.9346 (1.0027) loss 0.9877 (1.0284) grad_norm 12.3975 (9.3599/3.1154) mem 68106MB [2022-12-19 09:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1050/1519] eta 0:07:50 lr 0.000032 time 0.9292 (1.0036) model_time 0.9290 (1.0027) loss 0.7755 (1.0284) grad_norm 7.8827 (9.3631/3.1311) mem 68106MB [2022-12-19 09:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1060/1519] eta 0:07:40 lr 0.000032 time 1.0408 (1.0036) model_time 1.0406 (1.0028) loss 0.9621 (1.0281) grad_norm 8.0126 (9.3392/3.1222) mem 68106MB [2022-12-19 09:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1070/1519] eta 0:07:30 lr 0.000032 time 0.9252 (1.0037) model_time 0.9250 (1.0029) loss 0.8647 (1.0275) grad_norm 8.1205 (9.2439/3.0705) mem 68106MB [2022-12-19 09:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1080/1519] eta 0:07:20 lr 0.000032 time 0.9255 (1.0037) model_time 0.9254 (1.0029) loss 1.2698 (1.0278) grad_norm 8.1855 (9.1964/3.0263) mem 68106MB [2022-12-19 09:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1090/1519] eta 0:07:10 lr 0.000032 time 0.9343 (1.0037) model_time 0.9341 (1.0029) loss 1.1754 (1.0277) grad_norm 14.4563 (9.1315/2.8821) mem 68106MB [2022-12-19 09:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1100/1519] eta 0:07:00 lr 0.000032 time 0.9383 (1.0037) model_time 0.9382 (1.0029) loss 0.8590 (1.0273) grad_norm 7.1939 (9.1631/2.8985) mem 68106MB [2022-12-19 09:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1110/1519] eta 0:06:50 lr 0.000032 time 1.0057 (1.0037) model_time 1.0055 (1.0029) loss 0.7602 (1.0279) grad_norm 6.7912 (9.2112/2.9440) mem 68106MB [2022-12-19 09:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1120/1519] eta 0:06:40 lr 0.000032 time 0.9249 (1.0037) model_time 0.9248 (1.0029) loss 0.9548 (1.0273) grad_norm 6.2530 (9.2663/2.9511) mem 68106MB [2022-12-19 09:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1130/1519] eta 0:06:30 lr 0.000032 time 0.9433 (1.0037) model_time 0.9431 (1.0029) loss 0.9823 (1.0268) grad_norm 9.8073 (9.2404/2.9252) mem 68106MB [2022-12-19 09:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1140/1519] eta 0:06:20 lr 0.000032 time 0.9396 (1.0037) model_time 0.9393 (1.0028) loss 0.8502 (1.0269) grad_norm 6.1721 (9.2269/2.9269) mem 68106MB [2022-12-19 09:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1150/1519] eta 0:06:10 lr 0.000032 time 0.9661 (1.0037) model_time 0.9660 (1.0029) loss 1.1469 (1.0265) grad_norm 11.9109 (9.1998/2.9189) mem 68106MB [2022-12-19 09:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1160/1519] eta 0:06:00 lr 0.000032 time 0.9325 (1.0036) model_time 0.9323 (1.0028) loss 0.8255 (1.0259) grad_norm 10.7731 (9.2289/2.9595) mem 68106MB [2022-12-19 09:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1170/1519] eta 0:05:50 lr 0.000032 time 0.9311 (1.0036) model_time 0.9309 (1.0028) loss 1.2017 (1.0249) grad_norm 6.9923 (9.2301/2.9587) mem 68106MB [2022-12-19 09:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1180/1519] eta 0:05:40 lr 0.000032 time 0.9374 (1.0036) model_time 0.9372 (1.0028) loss 0.8843 (1.0242) grad_norm 9.3304 (9.2249/2.9447) mem 68106MB [2022-12-19 09:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1190/1519] eta 0:05:30 lr 0.000032 time 0.9360 (1.0035) model_time 0.9358 (1.0027) loss 0.9817 (1.0240) grad_norm 9.9905 (9.2639/2.9503) mem 68106MB [2022-12-19 09:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1200/1519] eta 0:05:20 lr 0.000032 time 0.9799 (1.0036) model_time 0.9798 (1.0028) loss 0.7150 (1.0238) grad_norm 12.9599 (9.2763/2.9576) mem 68106MB [2022-12-19 09:45:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1210/1519] eta 0:05:10 lr 0.000032 time 0.9337 (1.0036) model_time 0.9336 (1.0028) loss 1.7715 (1.0242) grad_norm 6.9897 (9.2394/2.9332) mem 68106MB [2022-12-19 09:46:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1220/1519] eta 0:05:00 lr 0.000032 time 0.9378 (1.0036) model_time 0.9377 (1.0028) loss 1.1742 (1.0246) grad_norm 14.4695 (9.2396/2.8634) mem 68106MB [2022-12-19 09:46:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1230/1519] eta 0:04:50 lr 0.000032 time 0.9302 (1.0035) model_time 0.9300 (1.0027) loss 0.9628 (1.0249) grad_norm 9.3430 (9.2050/2.8437) mem 68106MB [2022-12-19 09:46:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1240/1519] eta 0:04:39 lr 0.000032 time 0.9315 (1.0035) model_time 0.9313 (1.0027) loss 1.5228 (1.0251) grad_norm 9.8451 (9.1807/2.8420) mem 68106MB [2022-12-19 09:46:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1250/1519] eta 0:04:29 lr 0.000032 time 0.9280 (1.0035) model_time 0.9279 (1.0027) loss 1.2658 (1.0248) grad_norm 7.3685 (9.1795/2.8424) mem 68106MB [2022-12-19 09:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1260/1519] eta 0:04:19 lr 0.000032 time 0.9217 (1.0035) model_time 0.9216 (1.0027) loss 1.5537 (1.0251) grad_norm 6.9509 (9.1528/2.8475) mem 68106MB [2022-12-19 09:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1270/1519] eta 0:04:09 lr 0.000032 time 0.9512 (1.0035) model_time 0.9511 (1.0027) loss 0.7742 (1.0259) grad_norm 6.0308 (9.1673/2.8568) mem 68106MB [2022-12-19 09:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1280/1519] eta 0:03:59 lr 0.000032 time 0.9340 (1.0035) model_time 0.9339 (1.0027) loss 1.7390 (1.0266) grad_norm 8.3443 (9.1573/2.8557) mem 68106MB [2022-12-19 09:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1290/1519] eta 0:03:49 lr 0.000032 time 0.9288 (1.0035) model_time 0.9286 (1.0027) loss 0.8135 (1.0266) grad_norm 8.3222 (9.1556/2.8630) mem 68106MB [2022-12-19 09:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1300/1519] eta 0:03:39 lr 0.000032 time 0.9429 (1.0035) model_time 0.9427 (1.0027) loss 0.9901 (1.0269) grad_norm 9.2060 (9.1652/2.8789) mem 68106MB [2022-12-19 09:47:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1310/1519] eta 0:03:29 lr 0.000032 time 0.9498 (1.0034) model_time 0.9496 (1.0027) loss 1.0579 (1.0273) grad_norm 11.8083 (9.1740/2.8805) mem 68106MB [2022-12-19 09:47:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1320/1519] eta 0:03:19 lr 0.000032 time 0.9274 (1.0035) model_time 0.9272 (1.0027) loss 0.9204 (1.0274) grad_norm 6.9872 (9.2058/2.8869) mem 68106MB [2022-12-19 09:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1330/1519] eta 0:03:09 lr 0.000032 time 0.9261 (1.0035) model_time 0.9259 (1.0028) loss 0.9568 (1.0272) grad_norm 22.6871 (9.2569/2.9853) mem 68106MB [2022-12-19 09:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1340/1519] eta 0:02:59 lr 0.000032 time 0.9324 (1.0036) model_time 0.9322 (1.0029) loss 0.9361 (1.0265) grad_norm 8.1422 (9.2710/3.0437) mem 68106MB [2022-12-19 09:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1350/1519] eta 0:02:49 lr 0.000032 time 0.9238 (1.0036) model_time 0.9236 (1.0029) loss 1.2113 (1.0265) grad_norm 10.4616 (9.2590/3.0561) mem 68106MB [2022-12-19 09:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1360/1519] eta 0:02:39 lr 0.000032 time 0.9367 (1.0036) model_time 0.9365 (1.0029) loss 1.4290 (1.0263) grad_norm 10.3020 (9.2865/3.0915) mem 68106MB [2022-12-19 09:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1370/1519] eta 0:02:29 lr 0.000032 time 0.9309 (1.0036) model_time 0.9307 (1.0028) loss 1.2382 (1.0260) grad_norm 6.7976 (9.2641/3.0802) mem 68106MB [2022-12-19 09:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1380/1519] eta 0:02:19 lr 0.000032 time 1.0291 (1.0038) model_time 1.0290 (1.0030) loss 1.4505 (1.0265) grad_norm 11.4508 (9.2562/3.0748) mem 68106MB [2022-12-19 09:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1390/1519] eta 0:02:09 lr 0.000032 time 0.9348 (1.0038) model_time 0.9346 (1.0030) loss 0.8875 (1.0263) grad_norm 8.7541 (9.3007/3.0913) mem 68106MB [2022-12-19 09:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1400/1519] eta 0:01:59 lr 0.000032 time 0.9326 (1.0037) model_time 0.9325 (1.0030) loss 0.9292 (1.0267) grad_norm 9.7546 (9.3132/3.1102) mem 68106MB [2022-12-19 09:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1410/1519] eta 0:01:49 lr 0.000032 time 0.9313 (1.0037) model_time 0.9312 (1.0030) loss 1.2567 (1.0270) grad_norm 8.8478 (9.2963/3.0947) mem 68106MB [2022-12-19 09:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1420/1519] eta 0:01:39 lr 0.000032 time 0.9276 (1.0037) model_time 0.9275 (1.0030) loss 1.3624 (1.0272) grad_norm 11.3593 (9.2533/3.0723) mem 68106MB [2022-12-19 09:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1430/1519] eta 0:01:29 lr 0.000032 time 0.9257 (1.0038) model_time 0.9256 (1.0030) loss 0.8662 (1.0268) grad_norm 5.9721 (9.2489/3.0683) mem 68106MB [2022-12-19 09:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1440/1519] eta 0:01:19 lr 0.000032 time 0.9304 (1.0037) model_time 0.9302 (1.0030) loss 1.0386 (1.0262) grad_norm 14.1611 (9.2561/3.0710) mem 68106MB [2022-12-19 09:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1450/1519] eta 0:01:09 lr 0.000032 time 0.9359 (1.0037) model_time 0.9357 (1.0030) loss 0.8930 (1.0260) grad_norm 18.6409 (9.2732/3.0970) mem 68106MB [2022-12-19 09:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1460/1519] eta 0:00:59 lr 0.000032 time 0.9239 (1.0037) model_time 0.9237 (1.0030) loss 1.0768 (1.0260) grad_norm 10.3991 (9.2300/3.0299) mem 68106MB [2022-12-19 09:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1470/1519] eta 0:00:49 lr 0.000032 time 0.9329 (1.0037) model_time 0.9327 (1.0030) loss 1.0227 (1.0256) grad_norm 10.7021 (9.2253/2.9620) mem 68106MB [2022-12-19 09:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1480/1519] eta 0:00:39 lr 0.000032 time 0.9285 (1.0036) model_time 0.9284 (1.0029) loss 0.9535 (1.0254) grad_norm 7.2995 (9.2289/2.9580) mem 68106MB [2022-12-19 09:50:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1490/1519] eta 0:00:29 lr 0.000032 time 0.9326 (1.0036) model_time 0.9325 (1.0029) loss 0.8391 (1.0258) grad_norm 9.8447 (9.2113/2.9597) mem 68106MB [2022-12-19 09:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1500/1519] eta 0:00:19 lr 0.000032 time 0.9235 (1.0036) model_time 0.9233 (1.0029) loss 1.1706 (1.0262) grad_norm 7.1349 (9.2222/2.9619) mem 68106MB [2022-12-19 09:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [12/100][1510/1519] eta 0:00:09 lr 0.000032 time 0.9219 (1.0036) model_time 0.9218 (1.0029) loss 1.0217 (1.0263) grad_norm 7.4449 (9.2111/3.0373) mem 68106MB [2022-12-19 09:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 12 training takes 0:25:24 [2022-12-19 09:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_12.pth saving...... [2022-12-19 09:51:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_12.pth saved !!! [2022-12-19 09:51:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.634 (0.634) Loss 0.6831 (0.6831) Acc@1 86.111 (86.111) Acc@5 96.875 (96.875) Mem 68106MB [2022-12-19 09:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.331) Loss 0.7135 (0.6855) Acc@1 87.500 (85.890) Acc@5 97.569 (97.412) Mem 68106MB [2022-12-19 09:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.293 (0.314) Loss 0.6753 (0.6847) Acc@1 87.153 (86.326) Acc@5 98.611 (97.388) Mem 68106MB [2022-12-19 09:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 0.7437 (0.6869) Acc@1 85.069 (86.313) Acc@5 96.528 (97.301) Mem 68106MB [2022-12-19 09:51:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.304 (0.307) Loss 0.6476 (0.6773) Acc@1 85.764 (86.501) Acc@5 97.917 (97.400) Mem 68106MB [2022-12-19 09:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 0.6759 (0.6715) Acc@1 84.028 (86.560) Acc@5 97.569 (97.461) Mem 68106MB [2022-12-19 09:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.303 (0.305) Loss 0.7059 (0.6720) Acc@1 86.111 (86.618) Acc@5 96.181 (97.467) Mem 68106MB [2022-12-19 09:51:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.7742 (0.6753) Acc@1 84.028 (86.595) Acc@5 96.528 (97.442) Mem 68106MB [2022-12-19 09:51:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.303) Loss 0.6414 (0.6751) Acc@1 87.500 (86.621) Acc@5 97.569 (97.432) Mem 68106MB [2022-12-19 09:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:12] * Acc@1 86.661 Acc@5 97.458 [2022-12-19 09:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 86.7% [2022-12-19 09:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 09:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 09:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 86.66% [2022-12-19 09:52:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][0/1519] eta 0:35:53 lr 0.000032 time 1.4174 (1.4174) model_time 0.9719 (0.9719) loss 1.2261 (1.2261) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 09:52:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][10/1519] eta 0:26:16 lr 0.000032 time 0.9435 (1.0446) model_time 0.9434 (1.0038) loss 1.1400 (1.0239) grad_norm 14.6283 (12.7441/3.0899) mem 68106MB [2022-12-19 09:52:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][20/1519] eta 0:25:35 lr 0.000032 time 0.9361 (1.0243) model_time 0.9360 (1.0028) loss 0.7956 (0.9990) grad_norm 11.5263 (10.9353/3.0596) mem 68106MB [2022-12-19 09:52:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][30/1519] eta 0:25:13 lr 0.000032 time 0.9287 (1.0166) model_time 0.9286 (1.0019) loss 0.8127 (0.9603) grad_norm 15.8911 (10.3639/3.4141) mem 68106MB [2022-12-19 09:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][40/1519] eta 0:24:58 lr 0.000032 time 0.9498 (1.0131) model_time 0.9496 (1.0019) loss 1.0447 (0.9694) grad_norm 7.6517 (9.6761/3.3114) mem 68106MB [2022-12-19 09:53:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][50/1519] eta 0:24:45 lr 0.000032 time 0.9225 (1.0110) model_time 0.9224 (1.0019) loss 1.2340 (0.9787) grad_norm 9.3245 (10.1229/3.9964) mem 68106MB [2022-12-19 09:53:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][60/1519] eta 0:24:33 lr 0.000032 time 0.9023 (1.0097) model_time 0.9022 (1.0021) loss 0.7951 (0.9809) grad_norm 12.5837 (10.0472/3.8199) mem 68106MB [2022-12-19 09:53:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][70/1519] eta 0:24:23 lr 0.000032 time 0.9284 (1.0098) model_time 0.9282 (1.0032) loss 1.1187 (0.9696) grad_norm 7.3383 (9.6445/3.6816) mem 68106MB [2022-12-19 09:53:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][80/1519] eta 0:24:11 lr 0.000032 time 0.9276 (1.0089) model_time 0.9275 (1.0031) loss 0.7651 (0.9686) grad_norm 8.2491 (9.6018/3.5660) mem 68106MB [2022-12-19 09:53:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][90/1519] eta 0:24:00 lr 0.000032 time 0.9207 (1.0077) model_time 0.9205 (1.0025) loss 0.9293 (0.9719) grad_norm 9.3974 (9.7171/3.5481) mem 68106MB [2022-12-19 09:53:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][100/1519] eta 0:23:48 lr 0.000032 time 0.9239 (1.0070) model_time 0.9238 (1.0022) loss 1.1068 (0.9841) grad_norm 7.6434 (9.4847/3.4460) mem 68106MB [2022-12-19 09:54:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][110/1519] eta 0:23:38 lr 0.000032 time 0.9306 (1.0065) model_time 0.9304 (1.0022) loss 0.7840 (0.9871) grad_norm 9.5494 (9.5946/3.4235) mem 68106MB [2022-12-19 09:54:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][120/1519] eta 0:23:28 lr 0.000032 time 0.9272 (1.0066) model_time 0.9271 (1.0026) loss 0.7973 (0.9810) grad_norm 10.8954 (9.7152/3.3803) mem 68106MB [2022-12-19 09:54:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][130/1519] eta 0:23:17 lr 0.000032 time 0.9370 (1.0063) model_time 0.9368 (1.0026) loss 0.9825 (0.9893) grad_norm 9.1663 (9.6408/3.2598) mem 68106MB [2022-12-19 09:54:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][140/1519] eta 0:23:07 lr 0.000032 time 0.9231 (1.0058) model_time 0.9228 (1.0024) loss 0.9576 (0.9914) grad_norm 6.1244 (9.5063/3.2149) mem 68106MB [2022-12-19 09:54:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][150/1519] eta 0:22:56 lr 0.000032 time 0.9238 (1.0056) model_time 0.9237 (1.0023) loss 0.9805 (0.9914) grad_norm 11.5256 (9.5163/3.2098) mem 68106MB [2022-12-19 09:54:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][160/1519] eta 0:22:46 lr 0.000032 time 0.9291 (1.0054) model_time 0.9290 (1.0023) loss 1.2785 (0.9918) grad_norm 5.5555 (9.4863/3.1945) mem 68106MB [2022-12-19 09:55:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][170/1519] eta 0:22:38 lr 0.000032 time 0.9333 (1.0067) model_time 0.9332 (1.0038) loss 1.3976 (1.0000) grad_norm 5.8674 (9.4184/3.1464) mem 68106MB [2022-12-19 09:55:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][180/1519] eta 0:22:28 lr 0.000032 time 0.9156 (1.0072) model_time 0.9154 (1.0044) loss 1.2875 (1.0014) grad_norm 7.8305 (9.4144/3.0808) mem 68106MB [2022-12-19 09:55:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][190/1519] eta 0:22:18 lr 0.000032 time 0.9364 (1.0069) model_time 0.9362 (1.0043) loss 1.2199 (1.0081) grad_norm 5.6304 (9.4075/3.0975) mem 68106MB [2022-12-19 09:55:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][200/1519] eta 0:22:07 lr 0.000032 time 0.9285 (1.0066) model_time 0.9282 (1.0041) loss 1.5747 (1.0167) grad_norm 7.3672 (9.3186/3.0871) mem 68106MB [2022-12-19 09:55:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][210/1519] eta 0:21:57 lr 0.000032 time 0.9188 (1.0064) model_time 0.9185 (1.0040) loss 1.3142 (1.0232) grad_norm 9.7090 (9.3200/3.0524) mem 68106MB [2022-12-19 09:55:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][220/1519] eta 0:21:47 lr 0.000032 time 0.9294 (1.0062) model_time 0.9287 (1.0038) loss 1.2833 (1.0233) grad_norm 8.5523 (9.2828/3.0083) mem 68106MB [2022-12-19 09:56:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][230/1519] eta 0:21:37 lr 0.000032 time 0.9444 (1.0064) model_time 0.9440 (1.0041) loss 1.0328 (1.0196) grad_norm 10.8808 (9.3228/2.9607) mem 68106MB [2022-12-19 09:56:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][240/1519] eta 0:21:26 lr 0.000032 time 0.9366 (1.0061) model_time 0.9363 (1.0039) loss 1.4776 (1.0221) grad_norm 8.8271 (9.3112/2.9225) mem 68106MB [2022-12-19 09:56:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][250/1519] eta 0:21:16 lr 0.000032 time 0.9279 (1.0059) model_time 0.9278 (1.0037) loss 0.8848 (1.0229) grad_norm 7.9184 (9.2784/2.9008) mem 68106MB [2022-12-19 09:56:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][260/1519] eta 0:21:06 lr 0.000032 time 0.9237 (1.0057) model_time 0.9236 (1.0037) loss 1.0615 (1.0203) grad_norm 7.8950 (9.3054/2.9100) mem 68106MB [2022-12-19 09:56:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][270/1519] eta 0:20:55 lr 0.000032 time 0.9195 (1.0055) model_time 0.9194 (1.0035) loss 0.7857 (1.0220) grad_norm 8.7748 (9.3128/2.8600) mem 68106MB [2022-12-19 09:56:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][280/1519] eta 0:20:45 lr 0.000032 time 0.9341 (1.0054) model_time 0.9340 (1.0034) loss 0.8361 (1.0213) grad_norm 6.4985 (9.3227/2.8512) mem 68106MB [2022-12-19 09:57:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][290/1519] eta 0:20:36 lr 0.000032 time 0.9196 (1.0057) model_time 0.9195 (1.0038) loss 1.0553 (1.0223) grad_norm 10.5659 (9.3774/2.8864) mem 68106MB [2022-12-19 09:57:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][300/1519] eta 0:20:26 lr 0.000032 time 1.0296 (1.0061) model_time 1.0295 (1.0043) loss 1.0830 (1.0243) grad_norm 7.5314 (9.3703/2.8701) mem 68106MB [2022-12-19 09:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][310/1519] eta 0:20:16 lr 0.000032 time 0.9239 (1.0061) model_time 0.9237 (1.0043) loss 1.3030 (1.0255) grad_norm 19.5598 (9.4519/2.9732) mem 68106MB [2022-12-19 09:57:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][320/1519] eta 0:20:06 lr 0.000032 time 0.9319 (1.0059) model_time 0.9318 (1.0042) loss 0.8537 (1.0236) grad_norm 9.9612 (9.4499/2.9431) mem 68106MB [2022-12-19 09:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][330/1519] eta 0:19:55 lr 0.000032 time 0.9241 (1.0057) model_time 0.9239 (1.0040) loss 1.0166 (1.0239) grad_norm 14.7714 (9.5483/3.0881) mem 68106MB [2022-12-19 09:57:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][340/1519] eta 0:19:45 lr 0.000032 time 0.9332 (1.0056) model_time 0.9331 (1.0039) loss 1.2164 (1.0223) grad_norm 7.3647 (9.4917/3.0669) mem 68106MB [2022-12-19 09:58:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][350/1519] eta 0:19:35 lr 0.000032 time 0.9234 (1.0058) model_time 0.9233 (1.0042) loss 1.0698 (1.0203) grad_norm 12.9064 (9.4829/3.0504) mem 68106MB [2022-12-19 09:58:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][360/1519] eta 0:19:26 lr 0.000032 time 0.9187 (1.0065) model_time 0.9186 (1.0049) loss 1.0158 (1.0207) grad_norm 8.2813 (9.4601/3.0172) mem 68106MB [2022-12-19 09:58:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][370/1519] eta 0:19:16 lr 0.000032 time 0.9220 (1.0065) model_time 0.9218 (1.0049) loss 1.3408 (1.0204) grad_norm 7.0198 (9.3953/3.0018) mem 68106MB [2022-12-19 09:58:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][380/1519] eta 0:19:06 lr 0.000032 time 0.9251 (1.0066) model_time 0.9250 (1.0051) loss 0.9862 (1.0204) grad_norm 6.8829 (9.4127/2.9980) mem 68106MB [2022-12-19 09:58:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][390/1519] eta 0:18:56 lr 0.000032 time 0.9298 (1.0064) model_time 0.9296 (1.0049) loss 1.0574 (1.0209) grad_norm 5.9321 (9.3545/2.9844) mem 68106MB [2022-12-19 09:58:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][400/1519] eta 0:18:46 lr 0.000032 time 0.9286 (1.0063) model_time 0.9284 (1.0048) loss 1.3329 (1.0197) grad_norm 7.5779 (9.3065/2.9827) mem 68106MB [2022-12-19 09:59:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][410/1519] eta 0:18:35 lr 0.000032 time 0.9260 (1.0061) model_time 0.9259 (1.0047) loss 0.9655 (1.0199) grad_norm 7.6052 (9.3296/2.9952) mem 68106MB [2022-12-19 09:59:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][420/1519] eta 0:18:25 lr 0.000032 time 0.9320 (1.0060) model_time 0.9319 (1.0046) loss 1.0234 (1.0199) grad_norm 7.7343 (9.3182/2.9780) mem 68106MB [2022-12-19 09:59:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][430/1519] eta 0:18:15 lr 0.000032 time 0.9313 (1.0058) model_time 0.9312 (1.0044) loss 1.0117 (1.0184) grad_norm 9.6715 (9.2932/2.9644) mem 68106MB [2022-12-19 09:59:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][440/1519] eta 0:18:05 lr 0.000032 time 0.9215 (1.0058) model_time 0.9214 (1.0044) loss 0.7103 (1.0190) grad_norm 10.7718 (9.2668/2.9453) mem 68106MB [2022-12-19 09:59:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][450/1519] eta 0:17:55 lr 0.000032 time 0.9297 (1.0057) model_time 0.9296 (1.0043) loss 0.8181 (1.0189) grad_norm 14.4978 (9.2655/2.9508) mem 68106MB [2022-12-19 09:59:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][460/1519] eta 0:17:44 lr 0.000032 time 0.9204 (1.0055) model_time 0.9203 (1.0042) loss 0.8149 (1.0163) grad_norm 7.1026 (9.2313/2.9341) mem 68106MB [2022-12-19 10:00:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][470/1519] eta 0:17:34 lr 0.000032 time 0.9194 (1.0053) model_time 0.9193 (1.0040) loss 0.8602 (1.0136) grad_norm 6.4470 (9.2115/2.9387) mem 68106MB [2022-12-19 10:00:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][480/1519] eta 0:17:24 lr 0.000032 time 0.9266 (1.0052) model_time 0.9264 (1.0039) loss 0.9499 (1.0142) grad_norm 7.3420 (9.1837/2.9163) mem 68106MB [2022-12-19 10:00:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][490/1519] eta 0:17:14 lr 0.000032 time 0.9322 (1.0054) model_time 0.9320 (1.0042) loss 0.6992 (1.0133) grad_norm 11.6896 (9.1986/2.9139) mem 68106MB [2022-12-19 10:00:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][500/1519] eta 0:17:04 lr 0.000032 time 0.9227 (1.0053) model_time 0.9225 (1.0040) loss 0.8076 (1.0123) grad_norm 7.7505 (9.1933/2.8970) mem 68106MB [2022-12-19 10:00:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][510/1519] eta 0:16:54 lr 0.000032 time 0.9332 (1.0052) model_time 0.9330 (1.0040) loss 0.7632 (1.0123) grad_norm 12.8194 (9.1809/2.8922) mem 68106MB [2022-12-19 10:00:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][520/1519] eta 0:16:44 lr 0.000032 time 0.9478 (1.0051) model_time 0.9477 (1.0039) loss 0.8520 (1.0123) grad_norm 14.5288 (9.1698/2.8971) mem 68106MB [2022-12-19 10:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][530/1519] eta 0:16:33 lr 0.000032 time 0.9543 (1.0050) model_time 0.9541 (1.0038) loss 0.9792 (1.0124) grad_norm 8.9837 (9.1567/2.8771) mem 68106MB [2022-12-19 10:01:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][540/1519] eta 0:16:23 lr 0.000032 time 0.9252 (1.0051) model_time 0.9250 (1.0039) loss 1.0859 (1.0130) grad_norm 8.0185 (9.1309/2.8616) mem 68106MB [2022-12-19 10:01:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][550/1519] eta 0:16:13 lr 0.000032 time 0.9284 (1.0050) model_time 0.9283 (1.0039) loss 1.0878 (1.0150) grad_norm 12.0060 (9.1487/2.8435) mem 68106MB [2022-12-19 10:01:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][560/1519] eta 0:16:03 lr 0.000032 time 0.9205 (1.0049) model_time 0.9204 (1.0038) loss 1.2572 (1.0177) grad_norm 6.4601 (9.1616/2.8400) mem 68106MB [2022-12-19 10:01:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][570/1519] eta 0:15:53 lr 0.000032 time 0.9370 (1.0049) model_time 0.9368 (1.0037) loss 0.8303 (1.0185) grad_norm 7.4778 (9.1330/2.8236) mem 68106MB [2022-12-19 10:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][580/1519] eta 0:15:43 lr 0.000032 time 0.9276 (1.0048) model_time 0.9275 (1.0037) loss 1.2348 (1.0210) grad_norm 13.0347 (9.1221/2.8185) mem 68106MB [2022-12-19 10:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][590/1519] eta 0:15:33 lr 0.000032 time 0.9206 (1.0047) model_time 0.9204 (1.0036) loss 0.9561 (1.0195) grad_norm 7.0315 (9.0978/2.8026) mem 68106MB [2022-12-19 10:02:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][600/1519] eta 0:15:23 lr 0.000032 time 1.0213 (1.0047) model_time 1.0211 (1.0037) loss 1.0232 (1.0200) grad_norm 10.2938 (9.0678/2.7959) mem 68106MB [2022-12-19 10:02:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][610/1519] eta 0:15:13 lr 0.000032 time 0.9146 (1.0047) model_time 0.9145 (1.0037) loss 0.7620 (1.0195) grad_norm 12.2782 (9.0214/2.7493) mem 68106MB [2022-12-19 10:02:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][620/1519] eta 0:15:03 lr 0.000032 time 0.9208 (1.0049) model_time 0.9207 (1.0038) loss 0.7941 (1.0198) grad_norm 7.4015 (9.0220/2.7885) mem 68106MB [2022-12-19 10:02:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][630/1519] eta 0:14:53 lr 0.000032 time 0.9359 (1.0048) model_time 0.9358 (1.0038) loss 0.8543 (1.0189) grad_norm 11.4107 (9.0403/2.7628) mem 68106MB [2022-12-19 10:03:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][640/1519] eta 0:14:43 lr 0.000032 time 0.9277 (1.0047) model_time 0.9276 (1.0037) loss 0.7107 (1.0195) grad_norm 8.8960 (9.0398/2.7608) mem 68106MB [2022-12-19 10:03:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][650/1519] eta 0:14:33 lr 0.000032 time 0.9373 (1.0047) model_time 0.9371 (1.0036) loss 0.7558 (1.0197) grad_norm 8.9031 (8.9672/2.6455) mem 68106MB [2022-12-19 10:03:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][660/1519] eta 0:14:23 lr 0.000032 time 0.9221 (1.0047) model_time 0.9220 (1.0037) loss 0.9802 (1.0186) grad_norm 6.9882 (8.9267/2.6343) mem 68106MB [2022-12-19 10:03:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][670/1519] eta 0:14:13 lr 0.000032 time 0.9310 (1.0049) model_time 0.9308 (1.0039) loss 1.0585 (1.0186) grad_norm 7.0686 (8.9931/2.6769) mem 68106MB [2022-12-19 10:03:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][680/1519] eta 0:14:03 lr 0.000032 time 0.9225 (1.0048) model_time 0.9224 (1.0038) loss 0.9171 (1.0188) grad_norm 5.8849 (8.9703/2.6647) mem 68106MB [2022-12-19 10:03:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][690/1519] eta 0:13:53 lr 0.000032 time 1.0211 (1.0049) model_time 1.0210 (1.0039) loss 1.0994 (1.0184) grad_norm 7.1548 (8.9341/2.6410) mem 68106MB [2022-12-19 10:04:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][700/1519] eta 0:13:42 lr 0.000032 time 0.9416 (1.0049) model_time 0.9414 (1.0039) loss 1.0901 (1.0177) grad_norm 5.5528 (8.9237/2.6541) mem 68106MB [2022-12-19 10:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][710/1519] eta 0:13:32 lr 0.000032 time 0.9329 (1.0048) model_time 0.9328 (1.0038) loss 1.0024 (1.0179) grad_norm 13.7767 (8.9095/2.6348) mem 68106MB [2022-12-19 10:04:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][720/1519] eta 0:13:22 lr 0.000032 time 0.9316 (1.0048) model_time 0.9314 (1.0038) loss 0.8580 (1.0165) grad_norm 10.7067 (8.8858/2.6034) mem 68106MB [2022-12-19 10:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][730/1519] eta 0:13:12 lr 0.000032 time 0.9325 (1.0047) model_time 0.9324 (1.0037) loss 1.1297 (1.0170) grad_norm 9.7492 (8.8599/2.6177) mem 68106MB [2022-12-19 10:04:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][740/1519] eta 0:13:02 lr 0.000032 time 0.9319 (1.0046) model_time 0.9314 (1.0037) loss 1.2672 (1.0169) grad_norm 12.6417 (8.8800/2.6241) mem 68106MB [2022-12-19 10:04:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][750/1519] eta 0:12:52 lr 0.000032 time 0.9337 (1.0046) model_time 0.9334 (1.0036) loss 1.1281 (1.0178) grad_norm 5.7073 (8.8373/2.6124) mem 68106MB [2022-12-19 10:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][760/1519] eta 0:12:42 lr 0.000032 time 0.9359 (1.0045) model_time 0.9357 (1.0036) loss 0.8063 (1.0178) grad_norm 11.2027 (8.8631/2.6043) mem 68106MB [2022-12-19 10:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][770/1519] eta 0:12:32 lr 0.000032 time 0.9328 (1.0045) model_time 0.9327 (1.0035) loss 0.7907 (1.0176) grad_norm 11.8802 (8.8737/2.5996) mem 68106MB [2022-12-19 10:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][780/1519] eta 0:12:22 lr 0.000032 time 1.1483 (1.0047) model_time 1.1481 (1.0038) loss 1.1269 (1.0197) grad_norm 10.0529 (8.8631/2.5989) mem 68106MB [2022-12-19 10:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][790/1519] eta 0:12:12 lr 0.000032 time 0.9314 (1.0047) model_time 0.9313 (1.0038) loss 1.0822 (1.0210) grad_norm 6.4778 (8.8513/2.5719) mem 68106MB [2022-12-19 10:05:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][800/1519] eta 0:12:02 lr 0.000032 time 0.9327 (1.0049) model_time 0.9323 (1.0040) loss 0.9411 (1.0217) grad_norm 10.0944 (8.8649/2.5530) mem 68106MB [2022-12-19 10:05:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][810/1519] eta 0:11:52 lr 0.000032 time 0.9293 (1.0048) model_time 0.9292 (1.0039) loss 1.2285 (1.0202) grad_norm 7.5000 (8.8244/2.5491) mem 68106MB [2022-12-19 10:06:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][820/1519] eta 0:11:42 lr 0.000032 time 0.9286 (1.0047) model_time 0.9284 (1.0038) loss 0.7419 (1.0213) grad_norm 9.0346 (8.8127/2.5476) mem 68106MB [2022-12-19 10:06:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][830/1519] eta 0:11:32 lr 0.000032 time 0.9317 (1.0046) model_time 0.9315 (1.0037) loss 1.2102 (1.0221) grad_norm 9.0108 (8.8262/2.5706) mem 68106MB [2022-12-19 10:06:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][840/1519] eta 0:11:22 lr 0.000032 time 0.9323 (1.0046) model_time 0.9321 (1.0037) loss 1.0433 (1.0224) grad_norm 7.3402 (8.8200/2.5699) mem 68106MB [2022-12-19 10:06:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][850/1519] eta 0:11:12 lr 0.000032 time 0.9275 (1.0045) model_time 0.9274 (1.0037) loss 0.9791 (1.0217) grad_norm 7.4891 (8.8348/2.5636) mem 68106MB [2022-12-19 10:06:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][860/1519] eta 0:11:01 lr 0.000032 time 0.9226 (1.0045) model_time 0.9225 (1.0037) loss 1.0824 (1.0215) grad_norm 6.3965 (8.8250/2.5915) mem 68106MB [2022-12-19 10:06:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][870/1519] eta 0:10:52 lr 0.000032 time 1.0624 (1.0047) model_time 1.0590 (1.0038) loss 0.7737 (1.0217) grad_norm 10.6172 (8.8057/2.5944) mem 68106MB [2022-12-19 10:07:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][880/1519] eta 0:10:41 lr 0.000032 time 0.9298 (1.0047) model_time 0.9297 (1.0038) loss 0.8476 (1.0225) grad_norm 9.4693 (8.7983/2.5725) mem 68106MB [2022-12-19 10:07:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][890/1519] eta 0:10:31 lr 0.000032 time 0.9288 (1.0047) model_time 0.9287 (1.0038) loss 0.7950 (1.0213) grad_norm 7.1377 (8.7420/2.5313) mem 68106MB [2022-12-19 10:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][900/1519] eta 0:10:21 lr 0.000032 time 0.9255 (1.0046) model_time 0.9253 (1.0038) loss 0.9511 (1.0213) grad_norm 7.3166 (8.7480/2.5241) mem 68106MB [2022-12-19 10:07:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][910/1519] eta 0:10:11 lr 0.000032 time 0.9338 (1.0046) model_time 0.9337 (1.0037) loss 0.9541 (1.0209) grad_norm 8.7015 (8.6816/2.4306) mem 68106MB [2022-12-19 10:07:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][920/1519] eta 0:10:01 lr 0.000032 time 0.9284 (1.0045) model_time 0.9282 (1.0037) loss 1.3008 (1.0214) grad_norm 11.1588 (8.6523/2.4365) mem 68106MB [2022-12-19 10:07:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][930/1519] eta 0:09:51 lr 0.000032 time 0.9407 (1.0045) model_time 0.9406 (1.0037) loss 1.0417 (1.0216) grad_norm 9.9228 (8.5677/2.2910) mem 68106MB [2022-12-19 10:08:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][940/1519] eta 0:09:41 lr 0.000032 time 0.9251 (1.0045) model_time 0.9249 (1.0036) loss 1.1394 (1.0228) grad_norm 9.7494 (8.6087/2.2937) mem 68106MB [2022-12-19 10:08:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][950/1519] eta 0:09:31 lr 0.000032 time 0.9425 (1.0044) model_time 0.9424 (1.0036) loss 1.2174 (1.0235) grad_norm 8.0376 (8.5894/2.2890) mem 68106MB [2022-12-19 10:08:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][960/1519] eta 0:09:21 lr 0.000032 time 0.9241 (1.0044) model_time 0.9236 (1.0035) loss 1.0191 (1.0229) grad_norm 7.5755 (8.5838/2.3243) mem 68106MB [2022-12-19 10:08:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][970/1519] eta 0:09:11 lr 0.000032 time 0.9264 (1.0043) model_time 0.9262 (1.0035) loss 0.7303 (1.0221) grad_norm 6.7076 (8.6150/2.3400) mem 68106MB [2022-12-19 10:08:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][980/1519] eta 0:09:01 lr 0.000032 time 0.9069 (1.0046) model_time 0.9067 (1.0038) loss 0.9264 (1.0209) grad_norm 7.3116 (8.6394/2.3805) mem 68106MB [2022-12-19 10:08:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][990/1519] eta 0:08:51 lr 0.000032 time 0.9218 (1.0046) model_time 0.9215 (1.0038) loss 1.0138 (1.0220) grad_norm 8.4265 (8.6743/2.4027) mem 68106MB [2022-12-19 10:09:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1000/1519] eta 0:08:41 lr 0.000032 time 0.9330 (1.0045) model_time 0.9329 (1.0037) loss 0.9573 (1.0226) grad_norm 9.1676 (8.7035/2.4072) mem 68106MB [2022-12-19 10:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1010/1519] eta 0:08:31 lr 0.000032 time 0.9223 (1.0045) model_time 0.9221 (1.0037) loss 0.8903 (1.0230) grad_norm 5.3318 (8.6510/2.3716) mem 68106MB [2022-12-19 10:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1020/1519] eta 0:08:21 lr 0.000032 time 0.9226 (1.0045) model_time 0.9224 (1.0037) loss 1.1166 (1.0222) grad_norm 10.9255 (8.6841/2.4071) mem 68106MB [2022-12-19 10:09:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1030/1519] eta 0:08:11 lr 0.000032 time 0.9287 (1.0044) model_time 0.9286 (1.0036) loss 1.0416 (1.0212) grad_norm 9.2523 (8.6883/2.3957) mem 68106MB [2022-12-19 10:09:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1040/1519] eta 0:08:01 lr 0.000032 time 0.9339 (1.0043) model_time 0.9334 (1.0036) loss 0.9207 (1.0222) grad_norm 13.3103 (8.7119/2.4081) mem 68106MB [2022-12-19 10:09:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1050/1519] eta 0:07:51 lr 0.000032 time 0.9355 (1.0043) model_time 0.9354 (1.0035) loss 1.0986 (1.0220) grad_norm 6.3994 (8.6927/2.3872) mem 68106MB [2022-12-19 10:10:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1060/1519] eta 0:07:40 lr 0.000032 time 0.9313 (1.0043) model_time 0.9311 (1.0035) loss 1.0697 (1.0216) grad_norm 7.9768 (8.7004/2.3895) mem 68106MB [2022-12-19 10:10:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1070/1519] eta 0:07:30 lr 0.000032 time 0.9284 (1.0042) model_time 0.9282 (1.0035) loss 1.7022 (1.0226) grad_norm 10.2593 (8.7033/2.3632) mem 68106MB [2022-12-19 10:10:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1080/1519] eta 0:07:20 lr 0.000032 time 0.9299 (1.0042) model_time 0.9298 (1.0034) loss 0.9479 (1.0229) grad_norm 8.3906 (8.7136/2.3616) mem 68106MB [2022-12-19 10:10:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1090/1519] eta 0:07:10 lr 0.000032 time 0.9307 (1.0041) model_time 0.9306 (1.0034) loss 1.2396 (1.0234) grad_norm 9.8119 (8.7054/2.3396) mem 68106MB [2022-12-19 10:10:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1100/1519] eta 0:07:00 lr 0.000032 time 0.9225 (1.0042) model_time 0.9224 (1.0034) loss 1.1402 (1.0238) grad_norm 9.4147 (8.6939/2.3408) mem 68106MB [2022-12-19 10:10:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1110/1519] eta 0:06:50 lr 0.000032 time 0.9369 (1.0043) model_time 0.9367 (1.0036) loss 1.0934 (1.0237) grad_norm 8.6977 (8.6766/2.3263) mem 68106MB [2022-12-19 10:11:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1120/1519] eta 0:06:40 lr 0.000032 time 0.9207 (1.0043) model_time 0.9205 (1.0035) loss 1.0587 (1.0233) grad_norm 10.0239 (8.6576/2.3044) mem 68106MB [2022-12-19 10:11:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1130/1519] eta 0:06:30 lr 0.000032 time 0.9302 (1.0043) model_time 0.9300 (1.0035) loss 1.1151 (1.0236) grad_norm 9.3523 (8.7220/2.4142) mem 68106MB [2022-12-19 10:11:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1140/1519] eta 0:06:20 lr 0.000032 time 0.9279 (1.0042) model_time 0.9278 (1.0035) loss 1.2984 (1.0240) grad_norm 6.7123 (8.7346/2.4226) mem 68106MB [2022-12-19 10:11:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1150/1519] eta 0:06:10 lr 0.000032 time 0.9326 (1.0042) model_time 0.9325 (1.0034) loss 0.8851 (1.0246) grad_norm 10.4196 (8.7452/2.4330) mem 68106MB [2022-12-19 10:11:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1160/1519] eta 0:06:00 lr 0.000032 time 0.9336 (1.0042) model_time 0.9334 (1.0034) loss 1.1427 (1.0248) grad_norm 8.1678 (8.7361/2.4256) mem 68106MB [2022-12-19 10:11:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1170/1519] eta 0:05:50 lr 0.000032 time 0.9274 (1.0044) model_time 0.9272 (1.0036) loss 1.0464 (1.0246) grad_norm 11.0253 (8.7636/2.4393) mem 68106MB [2022-12-19 10:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1180/1519] eta 0:05:40 lr 0.000032 time 0.9292 (1.0044) model_time 0.9290 (1.0036) loss 1.2273 (1.0243) grad_norm 7.4722 (8.7625/2.4363) mem 68106MB [2022-12-19 10:12:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1190/1519] eta 0:05:30 lr 0.000032 time 0.9288 (1.0045) model_time 0.9286 (1.0037) loss 0.9127 (1.0238) grad_norm 12.5298 (8.7759/2.4471) mem 68106MB [2022-12-19 10:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1200/1519] eta 0:05:20 lr 0.000032 time 0.9195 (1.0044) model_time 0.9194 (1.0037) loss 1.0252 (1.0239) grad_norm 7.9818 (8.7813/2.4403) mem 68106MB [2022-12-19 10:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1210/1519] eta 0:05:10 lr 0.000032 time 0.9310 (1.0044) model_time 0.9309 (1.0037) loss 1.0557 (1.0232) grad_norm 6.0159 (8.7407/2.4255) mem 68106MB [2022-12-19 10:12:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1220/1519] eta 0:05:00 lr 0.000032 time 0.9340 (1.0044) model_time 0.9339 (1.0036) loss 0.8620 (1.0227) grad_norm 6.5773 (8.6999/2.3883) mem 68106MB [2022-12-19 10:12:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1230/1519] eta 0:04:50 lr 0.000032 time 0.9191 (1.0044) model_time 0.9189 (1.0036) loss 1.1656 (1.0233) grad_norm 6.0720 (8.6492/2.3757) mem 68106MB [2022-12-19 10:13:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1240/1519] eta 0:04:40 lr 0.000032 time 0.9238 (1.0044) model_time 0.9236 (1.0036) loss 1.1349 (1.0237) grad_norm 8.7088 (8.6944/2.4009) mem 68106MB [2022-12-19 10:13:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1250/1519] eta 0:04:30 lr 0.000032 time 0.9310 (1.0043) model_time 0.9309 (1.0036) loss 1.2200 (1.0232) grad_norm 6.5904 (8.6946/2.4055) mem 68106MB [2022-12-19 10:13:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1260/1519] eta 0:04:20 lr 0.000032 time 0.9191 (1.0043) model_time 0.9189 (1.0036) loss 0.7623 (1.0234) grad_norm 7.5419 (8.7139/2.4031) mem 68106MB [2022-12-19 10:13:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1270/1519] eta 0:04:10 lr 0.000032 time 0.9374 (1.0042) model_time 0.9372 (1.0035) loss 1.2123 (1.0235) grad_norm 4.8543 (8.6929/2.3692) mem 68106MB [2022-12-19 10:13:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1280/1519] eta 0:04:00 lr 0.000032 time 0.9298 (1.0042) model_time 0.9297 (1.0035) loss 1.2244 (1.0236) grad_norm 7.0289 (8.6896/2.3778) mem 68106MB [2022-12-19 10:13:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1290/1519] eta 0:03:50 lr 0.000032 time 0.9548 (1.0044) model_time 0.9546 (1.0037) loss 0.8009 (1.0235) grad_norm 9.8458 (8.7166/2.3874) mem 68106MB [2022-12-19 10:14:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1300/1519] eta 0:03:39 lr 0.000032 time 0.9293 (1.0044) model_time 0.9292 (1.0037) loss 0.9134 (1.0236) grad_norm 21.8973 (8.7894/2.4884) mem 68106MB [2022-12-19 10:14:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1310/1519] eta 0:03:29 lr 0.000032 time 0.9213 (1.0043) model_time 0.9212 (1.0036) loss 1.2717 (1.0232) grad_norm 8.6285 (8.7883/2.5441) mem 68106MB [2022-12-19 10:14:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1320/1519] eta 0:03:19 lr 0.000032 time 0.9187 (1.0043) model_time 0.9185 (1.0036) loss 1.0650 (1.0233) grad_norm 7.2736 (8.7691/2.5450) mem 68106MB [2022-12-19 10:14:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1330/1519] eta 0:03:09 lr 0.000032 time 0.9310 (1.0043) model_time 0.9309 (1.0036) loss 0.8077 (1.0231) grad_norm 11.8399 (8.7825/2.5520) mem 68106MB [2022-12-19 10:14:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1340/1519] eta 0:02:59 lr 0.000032 time 0.9331 (1.0042) model_time 0.9329 (1.0035) loss 1.0657 (1.0231) grad_norm 13.7607 (8.8103/2.5569) mem 68106MB [2022-12-19 10:14:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1350/1519] eta 0:02:49 lr 0.000032 time 0.9216 (1.0042) model_time 0.9215 (1.0035) loss 1.0567 (1.0232) grad_norm 8.0552 (8.8400/2.5388) mem 68106MB [2022-12-19 10:15:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1360/1519] eta 0:02:39 lr 0.000032 time 0.9301 (1.0042) model_time 0.9300 (1.0035) loss 1.1981 (1.0234) grad_norm 8.8221 (8.7920/2.5246) mem 68106MB [2022-12-19 10:15:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1370/1519] eta 0:02:29 lr 0.000032 time 0.9265 (1.0042) model_time 0.9264 (1.0035) loss 0.9940 (1.0231) grad_norm 7.7081 (8.7612/2.5265) mem 68106MB [2022-12-19 10:15:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1380/1519] eta 0:02:19 lr 0.000032 time 0.9383 (1.0041) model_time 0.9382 (1.0034) loss 0.7585 (1.0230) grad_norm 8.6921 (8.8230/2.6076) mem 68106MB [2022-12-19 10:15:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1390/1519] eta 0:02:09 lr 0.000032 time 0.9374 (1.0041) model_time 0.9373 (1.0034) loss 0.8897 (1.0232) grad_norm 6.1468 (8.7885/2.6162) mem 68106MB [2022-12-19 10:15:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1400/1519] eta 0:01:59 lr 0.000032 time 0.9450 (1.0042) model_time 0.9448 (1.0035) loss 1.1223 (1.0231) grad_norm 6.5357 (8.7916/2.6333) mem 68106MB [2022-12-19 10:15:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1410/1519] eta 0:01:49 lr 0.000032 time 0.9286 (1.0044) model_time 0.9284 (1.0037) loss 0.8709 (1.0224) grad_norm 6.8163 (8.8229/2.6498) mem 68106MB [2022-12-19 10:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1420/1519] eta 0:01:39 lr 0.000032 time 0.9246 (1.0045) model_time 0.9245 (1.0038) loss 1.3305 (1.0224) grad_norm 10.8185 (8.8337/2.6523) mem 68106MB [2022-12-19 10:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1430/1519] eta 0:01:29 lr 0.000032 time 0.9198 (1.0044) model_time 0.9196 (1.0038) loss 0.9107 (1.0218) grad_norm 9.6850 (8.7807/2.6323) mem 68106MB [2022-12-19 10:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1440/1519] eta 0:01:19 lr 0.000032 time 0.9310 (1.0044) model_time 0.9308 (1.0038) loss 0.9773 (1.0221) grad_norm 6.1094 (8.7674/2.6360) mem 68106MB [2022-12-19 10:16:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1450/1519] eta 0:01:09 lr 0.000032 time 0.9364 (1.0044) model_time 0.9362 (1.0037) loss 1.0897 (1.0224) grad_norm 11.1101 (8.8016/2.6638) mem 68106MB [2022-12-19 10:16:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1460/1519] eta 0:00:59 lr 0.000032 time 0.9170 (1.0044) model_time 0.9168 (1.0037) loss 0.8438 (1.0219) grad_norm 6.0202 (8.7717/2.6112) mem 68106MB [2022-12-19 10:16:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1470/1519] eta 0:00:49 lr 0.000032 time 0.9667 (1.0045) model_time 0.9665 (1.0038) loss 1.2083 (1.0229) grad_norm 7.6318 (8.7785/2.6206) mem 68106MB [2022-12-19 10:17:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1480/1519] eta 0:00:39 lr 0.000032 time 0.9240 (1.0045) model_time 0.9239 (1.0038) loss 1.1767 (1.0231) grad_norm 9.2309 (8.7744/2.6202) mem 68106MB [2022-12-19 10:17:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1490/1519] eta 0:00:29 lr 0.000032 time 0.9605 (1.0045) model_time 0.9604 (1.0038) loss 1.0796 (1.0229) grad_norm 5.5164 (8.7703/2.6221) mem 68106MB [2022-12-19 10:17:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1500/1519] eta 0:00:19 lr 0.000032 time 0.9193 (1.0046) model_time 0.9191 (1.0039) loss 0.9431 (1.0224) grad_norm 7.2084 (8.7583/2.6191) mem 68106MB [2022-12-19 10:17:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [13/100][1510/1519] eta 0:00:09 lr 0.000032 time 0.9217 (1.0045) model_time 0.9216 (1.0039) loss 0.8567 (1.0220) grad_norm 7.0123 (8.7446/2.6193) mem 68106MB [2022-12-19 10:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 13 training takes 0:25:25 [2022-12-19 10:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_13.pth saving...... [2022-12-19 10:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_13.pth saved !!! [2022-12-19 10:18:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.643 (0.643) Loss 0.6609 (0.6609) Acc@1 86.806 (86.806) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 10:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.329) Loss 0.6993 (0.6552) Acc@1 87.153 (87.216) Acc@5 97.569 (97.538) Mem 68106MB [2022-12-19 10:18:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.5977 (0.6531) Acc@1 87.500 (86.822) Acc@5 98.611 (97.652) Mem 68106MB [2022-12-19 10:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.7480 (0.6551) Acc@1 84.375 (86.761) Acc@5 96.181 (97.547) Mem 68106MB [2022-12-19 10:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.306) Loss 0.6386 (0.6471) Acc@1 86.111 (86.950) Acc@5 97.569 (97.654) Mem 68106MB [2022-12-19 10:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.305) Loss 0.6735 (0.6415) Acc@1 83.681 (87.166) Acc@5 97.222 (97.706) Mem 68106MB [2022-12-19 10:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.304) Loss 0.6514 (0.6405) Acc@1 86.111 (87.210) Acc@5 97.222 (97.661) Mem 68106MB [2022-12-19 10:18:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.299 (0.303) Loss 0.7169 (0.6429) Acc@1 86.458 (87.231) Acc@5 96.875 (97.643) Mem 68106MB [2022-12-19 10:18:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.302) Loss 0.5765 (0.6422) Acc@1 87.847 (87.251) Acc@5 97.569 (97.668) Mem 68106MB [2022-12-19 10:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:13] * Acc@1 87.242 Acc@5 97.679 [2022-12-19 10:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 87.2% [2022-12-19 10:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 10:18:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 10:18:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 87.24% [2022-12-19 10:18:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][0/1519] eta 0:36:09 lr 0.000032 time 1.4280 (1.4280) model_time 0.9736 (0.9736) loss 0.8747 (0.8747) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 10:19:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][10/1519] eta 0:26:23 lr 0.000032 time 0.9371 (1.0495) model_time 0.9369 (1.0079) loss 0.9577 (1.0529) grad_norm 8.9750 (8.9574/1.9079) mem 68106MB [2022-12-19 10:19:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][20/1519] eta 0:25:42 lr 0.000032 time 0.9316 (1.0288) model_time 0.9314 (1.0068) loss 0.8839 (1.0115) grad_norm 8.1833 (9.0768/1.9396) mem 68106MB [2022-12-19 10:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][30/1519] eta 0:25:20 lr 0.000032 time 0.9563 (1.0210) model_time 0.9562 (1.0061) loss 0.8440 (1.0233) grad_norm 10.4425 (8.9888/2.2557) mem 68106MB [2022-12-19 10:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][40/1519] eta 0:25:03 lr 0.000032 time 0.9444 (1.0169) model_time 0.9443 (1.0055) loss 0.9373 (1.0246) grad_norm 7.8857 (8.7820/2.0663) mem 68106MB [2022-12-19 10:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][50/1519] eta 0:24:50 lr 0.000032 time 0.9296 (1.0149) model_time 0.9294 (1.0057) loss 0.7859 (1.0170) grad_norm 7.0194 (9.1094/2.6627) mem 68106MB [2022-12-19 10:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][60/1519] eta 0:24:38 lr 0.000032 time 0.9189 (1.0130) model_time 0.9188 (1.0053) loss 1.1359 (1.0204) grad_norm 6.8041 (9.3596/2.6487) mem 68106MB [2022-12-19 10:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][70/1519] eta 0:24:25 lr 0.000032 time 0.9287 (1.0115) model_time 0.9286 (1.0048) loss 1.0950 (1.0157) grad_norm 8.3379 (9.2204/2.6078) mem 68106MB [2022-12-19 10:20:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][80/1519] eta 0:24:13 lr 0.000032 time 0.9272 (1.0104) model_time 0.9270 (1.0044) loss 0.8968 (1.0119) grad_norm 8.0660 (9.1196/2.4901) mem 68106MB [2022-12-19 10:20:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][90/1519] eta 0:24:02 lr 0.000032 time 0.9224 (1.0095) model_time 0.9222 (1.0042) loss 0.9814 (1.0085) grad_norm 10.2114 (8.9835/2.4278) mem 68106MB [2022-12-19 10:20:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][100/1519] eta 0:23:51 lr 0.000032 time 0.9361 (1.0089) model_time 0.9360 (1.0041) loss 1.1853 (1.0183) grad_norm 17.6752 (9.0767/2.7122) mem 68106MB [2022-12-19 10:20:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][110/1519] eta 0:23:40 lr 0.000032 time 0.9189 (1.0080) model_time 0.9187 (1.0036) loss 0.8508 (1.0221) grad_norm 8.1625 (9.0283/2.6141) mem 68106MB [2022-12-19 10:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][120/1519] eta 0:23:33 lr 0.000032 time 1.1441 (1.0101) model_time 1.1439 (1.0061) loss 1.0166 (1.0258) grad_norm 7.7601 (9.1105/2.5591) mem 68106MB [2022-12-19 10:21:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][130/1519] eta 0:23:21 lr 0.000032 time 0.9226 (1.0092) model_time 0.9224 (1.0054) loss 1.0416 (1.0186) grad_norm 6.1764 (8.9752/2.5252) mem 68106MB [2022-12-19 10:21:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][140/1519] eta 0:23:10 lr 0.000032 time 0.9183 (1.0084) model_time 0.9180 (1.0049) loss 1.2379 (1.0196) grad_norm 5.9118 (8.8997/2.5090) mem 68106MB [2022-12-19 10:21:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][150/1519] eta 0:23:00 lr 0.000032 time 0.9303 (1.0084) model_time 0.9301 (1.0050) loss 0.9493 (1.0187) grad_norm 9.8528 (8.8026/2.4763) mem 68106MB [2022-12-19 10:21:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][160/1519] eta 0:22:49 lr 0.000032 time 0.9235 (1.0078) model_time 0.9233 (1.0047) loss 0.9248 (1.0191) grad_norm 6.6263 (8.7015/2.4324) mem 68106MB [2022-12-19 10:21:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][170/1519] eta 0:22:40 lr 0.000032 time 0.9369 (1.0083) model_time 0.9367 (1.0053) loss 0.8548 (1.0185) grad_norm 9.4956 (8.7309/2.4290) mem 68106MB [2022-12-19 10:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][180/1519] eta 0:22:29 lr 0.000032 time 0.9203 (1.0080) model_time 0.9201 (1.0052) loss 0.9592 (1.0166) grad_norm 6.7639 (8.6899/2.3944) mem 68106MB [2022-12-19 10:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][190/1519] eta 0:22:19 lr 0.000032 time 0.9286 (1.0076) model_time 0.9284 (1.0049) loss 1.0877 (1.0160) grad_norm 8.2032 (8.6981/2.3445) mem 68106MB [2022-12-19 10:22:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][200/1519] eta 0:22:08 lr 0.000032 time 0.9173 (1.0071) model_time 0.9171 (1.0046) loss 0.8088 (1.0170) grad_norm 10.3171 (8.7241/2.3007) mem 68106MB [2022-12-19 10:22:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][210/1519] eta 0:21:58 lr 0.000032 time 0.9548 (1.0069) model_time 0.9546 (1.0044) loss 1.0156 (1.0118) grad_norm 10.4793 (8.6946/2.2858) mem 68106MB [2022-12-19 10:22:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][220/1519] eta 0:21:47 lr 0.000032 time 0.9303 (1.0067) model_time 0.9302 (1.0043) loss 0.7937 (1.0138) grad_norm 8.6652 (8.7432/2.2606) mem 68106MB [2022-12-19 10:22:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][230/1519] eta 0:21:37 lr 0.000032 time 0.9365 (1.0066) model_time 0.9363 (1.0043) loss 0.8797 (1.0157) grad_norm 14.7460 (8.7293/2.3121) mem 68106MB [2022-12-19 10:23:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][240/1519] eta 0:21:27 lr 0.000032 time 0.9047 (1.0068) model_time 0.9046 (1.0046) loss 0.8798 (1.0166) grad_norm 7.0808 (8.6874/2.2999) mem 68106MB [2022-12-19 10:23:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][250/1519] eta 0:21:17 lr 0.000032 time 0.9206 (1.0065) model_time 0.9204 (1.0044) loss 0.8603 (1.0166) grad_norm 9.2991 (8.7389/2.2815) mem 68106MB [2022-12-19 10:23:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][260/1519] eta 0:21:07 lr 0.000032 time 0.9311 (1.0066) model_time 0.9309 (1.0046) loss 0.8997 (1.0165) grad_norm 9.1421 (8.7041/2.2657) mem 68106MB [2022-12-19 10:23:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][270/1519] eta 0:20:56 lr 0.000032 time 0.9253 (1.0064) model_time 0.9251 (1.0044) loss 0.9299 (1.0150) grad_norm 6.8206 (8.6795/2.2482) mem 68106MB [2022-12-19 10:23:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][280/1519] eta 0:20:46 lr 0.000032 time 0.9312 (1.0062) model_time 0.9310 (1.0042) loss 1.0875 (1.0150) grad_norm 6.1881 (8.6628/2.2487) mem 68106MB [2022-12-19 10:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][290/1519] eta 0:20:36 lr 0.000032 time 0.9223 (1.0059) model_time 0.9222 (1.0040) loss 1.1135 (1.0148) grad_norm 13.9190 (8.7110/2.2944) mem 68106MB [2022-12-19 10:24:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][300/1519] eta 0:20:25 lr 0.000032 time 0.9308 (1.0057) model_time 0.9306 (1.0038) loss 0.8734 (1.0119) grad_norm 10.0676 (8.7058/2.2808) mem 68106MB [2022-12-19 10:24:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][310/1519] eta 0:20:15 lr 0.000032 time 0.9176 (1.0055) model_time 0.9174 (1.0037) loss 1.0626 (1.0107) grad_norm 12.7894 (8.7508/2.2792) mem 68106MB [2022-12-19 10:24:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][320/1519] eta 0:20:05 lr 0.000032 time 0.9312 (1.0052) model_time 0.9310 (1.0035) loss 0.9402 (1.0083) grad_norm 8.3154 (8.7677/2.2740) mem 68106MB [2022-12-19 10:24:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][330/1519] eta 0:19:55 lr 0.000032 time 0.9270 (1.0051) model_time 0.9268 (1.0034) loss 0.8720 (1.0069) grad_norm 12.3398 (8.8186/2.3104) mem 68106MB [2022-12-19 10:24:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][340/1519] eta 0:19:44 lr 0.000032 time 0.9320 (1.0049) model_time 0.9318 (1.0032) loss 0.9858 (1.0080) grad_norm 7.5313 (8.8079/2.2884) mem 68106MB [2022-12-19 10:24:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][350/1519] eta 0:19:34 lr 0.000032 time 0.9220 (1.0049) model_time 0.9219 (1.0032) loss 1.1857 (1.0052) grad_norm 8.6051 (8.8491/2.3669) mem 68106MB [2022-12-19 10:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][360/1519] eta 0:19:24 lr 0.000032 time 0.9098 (1.0049) model_time 0.9096 (1.0033) loss 0.8845 (1.0050) grad_norm 8.8083 (8.8515/2.3384) mem 68106MB [2022-12-19 10:25:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][370/1519] eta 0:19:14 lr 0.000032 time 0.9253 (1.0049) model_time 0.9252 (1.0033) loss 1.2691 (1.0043) grad_norm 6.1277 (8.8767/2.3748) mem 68106MB [2022-12-19 10:25:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][380/1519] eta 0:19:04 lr 0.000032 time 0.9317 (1.0047) model_time 0.9314 (1.0031) loss 0.7325 (1.0027) grad_norm 6.2716 (8.8186/2.3761) mem 68106MB [2022-12-19 10:25:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][390/1519] eta 0:18:54 lr 0.000032 time 0.9223 (1.0045) model_time 0.9222 (1.0030) loss 1.3676 (1.0045) grad_norm 6.5531 (8.8380/2.4129) mem 68106MB [2022-12-19 10:25:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][400/1519] eta 0:18:43 lr 0.000032 time 0.9232 (1.0044) model_time 0.9231 (1.0029) loss 0.8301 (1.0042) grad_norm 6.3769 (8.8161/2.3948) mem 68106MB [2022-12-19 10:25:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][410/1519] eta 0:18:33 lr 0.000032 time 0.9353 (1.0043) model_time 0.9352 (1.0029) loss 0.7821 (1.0028) grad_norm 7.3505 (8.7960/2.3719) mem 68106MB [2022-12-19 10:26:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][420/1519] eta 0:18:23 lr 0.000032 time 0.9198 (1.0042) model_time 0.9196 (1.0028) loss 1.4857 (1.0039) grad_norm 13.9464 (8.8203/2.3801) mem 68106MB [2022-12-19 10:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][430/1519] eta 0:18:14 lr 0.000032 time 0.9293 (1.0048) model_time 0.9292 (1.0034) loss 0.9222 (1.0048) grad_norm 6.6717 (8.8209/2.3958) mem 68106MB [2022-12-19 10:26:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][440/1519] eta 0:18:04 lr 0.000032 time 0.9210 (1.0049) model_time 0.9209 (1.0035) loss 0.8054 (1.0025) grad_norm 8.1236 (8.7893/2.3796) mem 68106MB [2022-12-19 10:26:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][450/1519] eta 0:17:54 lr 0.000032 time 0.9280 (1.0048) model_time 0.9279 (1.0034) loss 0.9434 (1.0042) grad_norm 7.2068 (8.7936/2.3729) mem 68106MB [2022-12-19 10:26:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][460/1519] eta 0:17:43 lr 0.000032 time 0.9246 (1.0047) model_time 0.9244 (1.0033) loss 1.0041 (1.0060) grad_norm 8.6491 (8.7768/2.3585) mem 68106MB [2022-12-19 10:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][470/1519] eta 0:17:33 lr 0.000032 time 0.9248 (1.0046) model_time 0.9247 (1.0033) loss 0.8473 (1.0049) grad_norm 14.3070 (8.8458/2.4079) mem 68106MB [2022-12-19 10:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][480/1519] eta 0:17:23 lr 0.000032 time 0.9363 (1.0046) model_time 0.9362 (1.0033) loss 0.9168 (1.0036) grad_norm 6.1986 (8.8247/2.3939) mem 68106MB [2022-12-19 10:27:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][490/1519] eta 0:17:13 lr 0.000032 time 0.9193 (1.0048) model_time 0.9192 (1.0035) loss 0.8466 (1.0039) grad_norm 9.6503 (8.8280/2.3849) mem 68106MB [2022-12-19 10:27:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][500/1519] eta 0:17:03 lr 0.000032 time 0.9248 (1.0047) model_time 0.9247 (1.0034) loss 0.7612 (1.0033) grad_norm 6.6079 (8.8124/2.3931) mem 68106MB [2022-12-19 10:27:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][510/1519] eta 0:16:53 lr 0.000032 time 0.9342 (1.0046) model_time 0.9340 (1.0034) loss 0.9812 (1.0032) grad_norm 10.1831 (8.8382/2.4175) mem 68106MB [2022-12-19 10:27:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][520/1519] eta 0:16:43 lr 0.000032 time 0.9209 (1.0045) model_time 0.9208 (1.0033) loss 1.4861 (1.0060) grad_norm 9.7386 (8.8360/2.4081) mem 68106MB [2022-12-19 10:27:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][530/1519] eta 0:16:33 lr 0.000032 time 0.9460 (1.0045) model_time 0.9458 (1.0033) loss 0.7858 (1.0062) grad_norm 7.4106 (8.9048/2.6152) mem 68106MB [2022-12-19 10:28:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][540/1519] eta 0:16:23 lr 0.000032 time 0.9535 (1.0045) model_time 0.9533 (1.0033) loss 1.2024 (1.0082) grad_norm 7.0539 (8.8940/2.6013) mem 68106MB [2022-12-19 10:28:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][550/1519] eta 0:16:13 lr 0.000032 time 0.9198 (1.0044) model_time 0.9197 (1.0032) loss 0.9958 (1.0069) grad_norm 8.7771 (8.9032/2.6259) mem 68106MB [2022-12-19 10:28:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][560/1519] eta 0:16:03 lr 0.000032 time 0.9315 (1.0044) model_time 0.9312 (1.0032) loss 0.8727 (1.0068) grad_norm 9.4624 (8.9013/2.6084) mem 68106MB [2022-12-19 10:28:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][570/1519] eta 0:15:53 lr 0.000032 time 1.1811 (1.0047) model_time 1.1809 (1.0036) loss 1.0816 (1.0077) grad_norm 10.7300 (8.8737/2.6047) mem 68106MB [2022-12-19 10:28:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][580/1519] eta 0:15:43 lr 0.000032 time 0.9319 (1.0046) model_time 0.9317 (1.0035) loss 0.9024 (1.0062) grad_norm 7.6162 (8.8781/2.5853) mem 68106MB [2022-12-19 10:28:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][590/1519] eta 0:15:33 lr 0.000032 time 0.9374 (1.0045) model_time 0.9372 (1.0034) loss 0.9199 (1.0056) grad_norm 6.6135 (8.8611/2.5740) mem 68106MB [2022-12-19 10:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][600/1519] eta 0:15:23 lr 0.000032 time 0.9313 (1.0045) model_time 0.9311 (1.0033) loss 1.3752 (1.0059) grad_norm 8.4269 (8.8882/2.5914) mem 68106MB [2022-12-19 10:29:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][610/1519] eta 0:15:13 lr 0.000032 time 0.9352 (1.0044) model_time 0.9351 (1.0033) loss 1.0781 (1.0056) grad_norm 7.6564 (8.8828/2.5986) mem 68106MB [2022-12-19 10:29:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][620/1519] eta 0:15:02 lr 0.000032 time 0.9338 (1.0044) model_time 0.9337 (1.0033) loss 1.0234 (1.0072) grad_norm 5.2772 (8.8674/2.6220) mem 68106MB [2022-12-19 10:29:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][630/1519] eta 0:14:52 lr 0.000032 time 0.9311 (1.0043) model_time 0.9309 (1.0032) loss 0.8961 (1.0081) grad_norm 8.0685 (8.9175/2.6415) mem 68106MB [2022-12-19 10:29:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][640/1519] eta 0:14:42 lr 0.000032 time 0.9254 (1.0042) model_time 0.9253 (1.0032) loss 0.7411 (1.0068) grad_norm 5.5859 (8.9125/2.6575) mem 68106MB [2022-12-19 10:29:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][650/1519] eta 0:14:32 lr 0.000032 time 0.9448 (1.0042) model_time 0.9447 (1.0031) loss 0.7471 (1.0060) grad_norm 8.1778 (8.8852/2.6051) mem 68106MB [2022-12-19 10:30:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][660/1519] eta 0:14:22 lr 0.000032 time 0.9258 (1.0041) model_time 0.9257 (1.0030) loss 0.8053 (1.0049) grad_norm 10.7466 (8.8339/2.5960) mem 68106MB [2022-12-19 10:30:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][670/1519] eta 0:14:12 lr 0.000032 time 0.9320 (1.0041) model_time 0.9315 (1.0031) loss 1.0203 (1.0032) grad_norm 12.4458 (8.8439/2.5995) mem 68106MB [2022-12-19 10:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][680/1519] eta 0:14:02 lr 0.000032 time 0.9613 (1.0042) model_time 0.9611 (1.0031) loss 1.0240 (1.0011) grad_norm 12.9343 (8.8458/2.6103) mem 68106MB [2022-12-19 10:30:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][690/1519] eta 0:13:52 lr 0.000032 time 0.9354 (1.0041) model_time 0.9352 (1.0031) loss 1.1425 (1.0005) grad_norm 7.4110 (8.8762/2.6395) mem 68106MB [2022-12-19 10:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][700/1519] eta 0:13:42 lr 0.000032 time 0.9348 (1.0041) model_time 0.9347 (1.0030) loss 0.7969 (0.9990) grad_norm 8.7211 (8.8451/2.5797) mem 68106MB [2022-12-19 10:30:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][710/1519] eta 0:13:32 lr 0.000032 time 0.9345 (1.0041) model_time 0.9343 (1.0030) loss 0.9975 (0.9991) grad_norm 5.9010 (8.8319/2.5914) mem 68106MB [2022-12-19 10:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][720/1519] eta 0:13:22 lr 0.000032 time 0.9273 (1.0040) model_time 0.9272 (1.0030) loss 0.7229 (0.9998) grad_norm 7.5904 (8.8024/2.5864) mem 68106MB [2022-12-19 10:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][730/1519] eta 0:13:12 lr 0.000032 time 0.9414 (1.0040) model_time 0.9412 (1.0030) loss 1.0368 (0.9992) grad_norm 7.9527 (8.8136/2.5795) mem 68106MB [2022-12-19 10:31:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][740/1519] eta 0:13:02 lr 0.000032 time 1.0173 (1.0040) model_time 1.0172 (1.0030) loss 0.8101 (0.9996) grad_norm 11.2286 (8.8138/2.5833) mem 68106MB [2022-12-19 10:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][750/1519] eta 0:12:52 lr 0.000032 time 1.0441 (1.0042) model_time 1.0440 (1.0032) loss 0.7409 (0.9996) grad_norm 9.7106 (8.8655/2.6080) mem 68106MB [2022-12-19 10:31:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][760/1519] eta 0:12:42 lr 0.000032 time 0.9278 (1.0041) model_time 0.9277 (1.0031) loss 1.3873 (0.9990) grad_norm 7.8135 (8.8887/2.6069) mem 68106MB [2022-12-19 10:31:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][770/1519] eta 0:12:32 lr 0.000032 time 0.9329 (1.0041) model_time 0.9326 (1.0031) loss 0.9771 (0.9994) grad_norm 6.1959 (8.8816/2.6154) mem 68106MB [2022-12-19 10:32:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][780/1519] eta 0:12:21 lr 0.000032 time 0.9350 (1.0040) model_time 0.9349 (1.0031) loss 0.7215 (0.9993) grad_norm 6.8784 (8.9471/2.7102) mem 68106MB [2022-12-19 10:32:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][790/1519] eta 0:12:11 lr 0.000032 time 0.9333 (1.0040) model_time 0.9332 (1.0030) loss 1.1243 (1.0002) grad_norm 7.9732 (8.9497/2.7127) mem 68106MB [2022-12-19 10:32:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][800/1519] eta 0:12:01 lr 0.000032 time 0.9340 (1.0041) model_time 0.9338 (1.0032) loss 1.0683 (0.9994) grad_norm 5.4783 (8.9646/2.7654) mem 68106MB [2022-12-19 10:32:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][810/1519] eta 0:11:51 lr 0.000032 time 0.9287 (1.0040) model_time 0.9286 (1.0031) loss 1.0772 (0.9998) grad_norm 6.7617 (8.9761/2.7763) mem 68106MB [2022-12-19 10:32:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][820/1519] eta 0:11:41 lr 0.000032 time 0.9332 (1.0040) model_time 0.9330 (1.0031) loss 0.8217 (1.0008) grad_norm 7.5490 (8.9428/2.7740) mem 68106MB [2022-12-19 10:32:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][830/1519] eta 0:11:31 lr 0.000032 time 0.9218 (1.0040) model_time 0.9216 (1.0031) loss 1.1505 (1.0013) grad_norm 6.7320 (8.9330/2.7542) mem 68106MB [2022-12-19 10:33:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][840/1519] eta 0:11:21 lr 0.000032 time 0.9368 (1.0040) model_time 0.9366 (1.0031) loss 0.8882 (1.0019) grad_norm 11.4497 (8.9499/2.7481) mem 68106MB [2022-12-19 10:33:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][850/1519] eta 0:11:11 lr 0.000032 time 0.9349 (1.0040) model_time 0.9348 (1.0031) loss 0.7584 (1.0015) grad_norm 6.3085 (8.8913/2.7593) mem 68106MB [2022-12-19 10:33:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][860/1519] eta 0:11:01 lr 0.000032 time 0.9379 (1.0040) model_time 0.9376 (1.0031) loss 0.8499 (1.0016) grad_norm 12.5267 (8.9057/2.7680) mem 68106MB [2022-12-19 10:33:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][870/1519] eta 0:10:51 lr 0.000032 time 0.9298 (1.0040) model_time 0.9296 (1.0031) loss 0.7492 (1.0020) grad_norm 7.7888 (8.9170/2.7638) mem 68106MB [2022-12-19 10:33:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][880/1519] eta 0:10:41 lr 0.000032 time 0.9336 (1.0040) model_time 0.9335 (1.0031) loss 0.9437 (1.0023) grad_norm 7.2608 (8.9260/2.7670) mem 68106MB [2022-12-19 10:33:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][890/1519] eta 0:10:31 lr 0.000032 time 0.9256 (1.0039) model_time 0.9254 (1.0030) loss 0.7659 (1.0024) grad_norm 6.3559 (8.8856/2.7424) mem 68106MB [2022-12-19 10:34:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][900/1519] eta 0:10:21 lr 0.000032 time 0.9285 (1.0038) model_time 0.9283 (1.0030) loss 1.1365 (1.0034) grad_norm 8.2092 (8.9393/2.8095) mem 68106MB [2022-12-19 10:34:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][910/1519] eta 0:10:11 lr 0.000032 time 0.9355 (1.0038) model_time 0.9353 (1.0029) loss 0.8808 (1.0034) grad_norm 7.0327 (8.9363/2.8174) mem 68106MB [2022-12-19 10:34:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][920/1519] eta 0:10:01 lr 0.000032 time 0.9271 (1.0038) model_time 0.9270 (1.0029) loss 1.1332 (1.0032) grad_norm 8.8801 (8.9423/2.8164) mem 68106MB [2022-12-19 10:34:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][930/1519] eta 0:09:51 lr 0.000032 time 0.9261 (1.0038) model_time 0.9260 (1.0029) loss 1.0776 (1.0039) grad_norm 7.4262 (8.8994/2.7919) mem 68106MB [2022-12-19 10:34:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][940/1519] eta 0:09:41 lr 0.000032 time 0.9301 (1.0037) model_time 0.9300 (1.0029) loss 0.9549 (1.0041) grad_norm 12.2764 (8.9217/2.7949) mem 68106MB [2022-12-19 10:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][950/1519] eta 0:09:31 lr 0.000032 time 0.9291 (1.0037) model_time 0.9289 (1.0028) loss 1.0316 (1.0036) grad_norm 6.7206 (8.8992/2.7527) mem 68106MB [2022-12-19 10:35:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][960/1519] eta 0:09:21 lr 0.000032 time 0.9477 (1.0037) model_time 0.9476 (1.0028) loss 0.7931 (1.0037) grad_norm 6.3849 (8.8652/2.7633) mem 68106MB [2022-12-19 10:35:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][970/1519] eta 0:09:10 lr 0.000032 time 0.9327 (1.0036) model_time 0.9325 (1.0028) loss 1.0754 (1.0034) grad_norm 14.6118 (8.9024/2.7925) mem 68106MB [2022-12-19 10:35:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][980/1519] eta 0:09:00 lr 0.000032 time 0.9305 (1.0036) model_time 0.9304 (1.0027) loss 1.2043 (1.0036) grad_norm 7.4142 (8.9142/2.7890) mem 68106MB [2022-12-19 10:35:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][990/1519] eta 0:08:50 lr 0.000032 time 0.9394 (1.0037) model_time 0.9393 (1.0029) loss 1.3913 (1.0038) grad_norm 7.3705 (8.9143/2.7654) mem 68106MB [2022-12-19 10:35:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1000/1519] eta 0:08:40 lr 0.000032 time 0.9194 (1.0037) model_time 0.9192 (1.0029) loss 1.0410 (1.0038) grad_norm 9.4369 (8.9424/2.7738) mem 68106MB [2022-12-19 10:35:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1010/1519] eta 0:08:30 lr 0.000032 time 0.9499 (1.0038) model_time 0.9497 (1.0030) loss 0.8422 (1.0031) grad_norm 8.3553 (8.9607/2.7732) mem 68106MB [2022-12-19 10:36:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1020/1519] eta 0:08:20 lr 0.000032 time 0.9199 (1.0038) model_time 0.9197 (1.0030) loss 0.9206 (1.0026) grad_norm 7.8550 (8.9419/2.7613) mem 68106MB [2022-12-19 10:36:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1030/1519] eta 0:08:10 lr 0.000032 time 0.9319 (1.0038) model_time 0.9317 (1.0029) loss 1.0996 (1.0026) grad_norm 7.3916 (8.9350/2.7432) mem 68106MB [2022-12-19 10:36:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1040/1519] eta 0:08:00 lr 0.000032 time 0.9363 (1.0038) model_time 0.9362 (1.0029) loss 0.7940 (1.0027) grad_norm 10.3692 (8.9599/2.7398) mem 68106MB [2022-12-19 10:36:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1050/1519] eta 0:07:50 lr 0.000032 time 0.9421 (1.0038) model_time 0.9419 (1.0030) loss 0.9410 (1.0029) grad_norm 8.7117 (8.9890/2.7498) mem 68106MB [2022-12-19 10:36:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1060/1519] eta 0:07:40 lr 0.000032 time 0.9243 (1.0042) model_time 0.9242 (1.0034) loss 0.8250 (1.0030) grad_norm 10.5848 (9.0128/2.7526) mem 68106MB [2022-12-19 10:36:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1070/1519] eta 0:07:30 lr 0.000032 time 0.9219 (1.0045) model_time 0.9218 (1.0036) loss 0.9169 (1.0035) grad_norm 5.7971 (8.9346/2.7173) mem 68106MB [2022-12-19 10:37:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1080/1519] eta 0:07:20 lr 0.000032 time 0.9268 (1.0044) model_time 0.9267 (1.0036) loss 0.8373 (1.0031) grad_norm 7.9041 (8.9262/2.7175) mem 68106MB [2022-12-19 10:37:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1090/1519] eta 0:07:10 lr 0.000032 time 0.9265 (1.0044) model_time 0.9263 (1.0036) loss 0.8166 (1.0026) grad_norm 9.3796 (8.9384/2.7172) mem 68106MB [2022-12-19 10:37:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1100/1519] eta 0:07:00 lr 0.000032 time 0.9743 (1.0044) model_time 0.9741 (1.0036) loss 1.2738 (1.0030) grad_norm 6.2233 (8.9362/2.7049) mem 68106MB [2022-12-19 10:37:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1110/1519] eta 0:06:50 lr 0.000032 time 0.9294 (1.0045) model_time 0.9292 (1.0037) loss 0.8328 (1.0033) grad_norm 10.5068 (8.9107/2.6744) mem 68106MB [2022-12-19 10:37:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1120/1519] eta 0:06:40 lr 0.000032 time 0.9177 (1.0044) model_time 0.9175 (1.0036) loss 1.0291 (1.0036) grad_norm 9.6451 (8.9091/2.6648) mem 68106MB [2022-12-19 10:37:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1130/1519] eta 0:06:30 lr 0.000032 time 0.9369 (1.0044) model_time 0.9367 (1.0036) loss 0.7231 (1.0032) grad_norm 8.2366 (8.8342/2.4722) mem 68106MB [2022-12-19 10:38:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1140/1519] eta 0:06:20 lr 0.000032 time 0.9205 (1.0043) model_time 0.9204 (1.0035) loss 1.1118 (1.0034) grad_norm 7.2082 (8.8275/2.4710) mem 68106MB [2022-12-19 10:38:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1150/1519] eta 0:06:10 lr 0.000032 time 0.9238 (1.0043) model_time 0.9237 (1.0035) loss 0.7674 (1.0030) grad_norm 6.7303 (8.8204/2.4319) mem 68106MB [2022-12-19 10:38:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1160/1519] eta 0:06:00 lr 0.000032 time 0.9180 (1.0042) model_time 0.9178 (1.0035) loss 0.7246 (1.0029) grad_norm 6.6340 (8.8181/2.4352) mem 68106MB [2022-12-19 10:38:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1170/1519] eta 0:05:50 lr 0.000032 time 0.9342 (1.0043) model_time 0.9340 (1.0035) loss 0.8915 (1.0026) grad_norm 7.6879 (8.8237/2.4213) mem 68106MB [2022-12-19 10:38:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1180/1519] eta 0:05:40 lr 0.000032 time 0.9309 (1.0043) model_time 0.9307 (1.0035) loss 1.2713 (1.0021) grad_norm 10.7073 (8.8115/2.4246) mem 68106MB [2022-12-19 10:38:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1190/1519] eta 0:05:30 lr 0.000032 time 0.9235 (1.0043) model_time 0.9233 (1.0035) loss 1.1029 (1.0021) grad_norm 10.3798 (8.8376/2.4218) mem 68106MB [2022-12-19 10:39:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1200/1519] eta 0:05:20 lr 0.000032 time 0.9219 (1.0043) model_time 0.9218 (1.0035) loss 1.1511 (1.0031) grad_norm 7.3732 (8.8199/2.3996) mem 68106MB [2022-12-19 10:39:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1210/1519] eta 0:05:10 lr 0.000032 time 0.9217 (1.0042) model_time 0.9216 (1.0035) loss 0.7929 (1.0036) grad_norm 7.6204 (8.7966/2.3908) mem 68106MB [2022-12-19 10:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1220/1519] eta 0:05:00 lr 0.000032 time 0.9335 (1.0042) model_time 0.9333 (1.0035) loss 1.1789 (1.0029) grad_norm 6.7525 (8.8066/2.3671) mem 68106MB [2022-12-19 10:39:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1230/1519] eta 0:04:50 lr 0.000032 time 0.9293 (1.0042) model_time 0.9291 (1.0035) loss 0.8300 (1.0027) grad_norm 7.1822 (8.7564/2.3204) mem 68106MB [2022-12-19 10:39:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1240/1519] eta 0:04:40 lr 0.000032 time 0.9295 (1.0043) model_time 0.9294 (1.0036) loss 1.0575 (1.0030) grad_norm 18.2761 (8.8031/2.3642) mem 68106MB [2022-12-19 10:39:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1250/1519] eta 0:04:30 lr 0.000032 time 0.9209 (1.0043) model_time 0.9208 (1.0036) loss 1.1887 (1.0027) grad_norm 10.7984 (8.8005/2.3723) mem 68106MB [2022-12-19 10:40:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1260/1519] eta 0:04:20 lr 0.000032 time 0.9180 (1.0043) model_time 0.9179 (1.0036) loss 1.3172 (1.0026) grad_norm 7.5347 (8.7973/2.3688) mem 68106MB [2022-12-19 10:40:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1270/1519] eta 0:04:10 lr 0.000032 time 0.9247 (1.0043) model_time 0.9246 (1.0036) loss 0.8325 (1.0017) grad_norm 6.0128 (8.7864/2.3589) mem 68106MB [2022-12-19 10:40:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1280/1519] eta 0:04:00 lr 0.000032 time 0.9881 (1.0043) model_time 0.9880 (1.0036) loss 0.8870 (1.0022) grad_norm 11.7693 (8.8198/2.3726) mem 68106MB [2022-12-19 10:40:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1290/1519] eta 0:03:50 lr 0.000032 time 0.9319 (1.0044) model_time 0.9317 (1.0037) loss 0.9599 (1.0027) grad_norm 7.9634 (8.8142/2.3759) mem 68106MB [2022-12-19 10:40:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1300/1519] eta 0:03:39 lr 0.000032 time 0.9270 (1.0043) model_time 0.9269 (1.0036) loss 1.1012 (1.0032) grad_norm 10.6003 (8.8670/2.4004) mem 68106MB [2022-12-19 10:40:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1310/1519] eta 0:03:29 lr 0.000032 time 0.9292 (1.0043) model_time 0.9291 (1.0036) loss 0.7684 (1.0029) grad_norm 7.3774 (8.8795/2.4072) mem 68106MB [2022-12-19 10:41:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1320/1519] eta 0:03:19 lr 0.000032 time 0.9265 (1.0043) model_time 0.9264 (1.0036) loss 0.7313 (1.0026) grad_norm 8.4621 (8.9062/2.4077) mem 68106MB [2022-12-19 10:41:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1330/1519] eta 0:03:09 lr 0.000032 time 0.9407 (1.0043) model_time 0.9405 (1.0036) loss 0.9464 (1.0023) grad_norm 7.6298 (8.9036/2.4118) mem 68106MB [2022-12-19 10:41:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1340/1519] eta 0:02:59 lr 0.000032 time 0.9073 (1.0043) model_time 0.9072 (1.0036) loss 1.0615 (1.0022) grad_norm 8.8670 (8.9188/2.3995) mem 68106MB [2022-12-19 10:41:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1350/1519] eta 0:02:49 lr 0.000032 time 0.9420 (1.0042) model_time 0.9417 (1.0035) loss 0.8247 (1.0020) grad_norm 7.9489 (8.8957/2.3821) mem 68106MB [2022-12-19 10:41:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1360/1519] eta 0:02:39 lr 0.000032 time 0.9223 (1.0043) model_time 0.9221 (1.0036) loss 0.8191 (1.0022) grad_norm 7.5688 (8.9118/2.3933) mem 68106MB [2022-12-19 10:41:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1370/1519] eta 0:02:29 lr 0.000032 time 0.9323 (1.0043) model_time 0.9322 (1.0036) loss 0.9306 (1.0027) grad_norm 7.8766 (8.9013/2.3818) mem 68106MB [2022-12-19 10:42:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1380/1519] eta 0:02:19 lr 0.000032 time 0.9231 (1.0043) model_time 0.9229 (1.0036) loss 1.0300 (1.0027) grad_norm 6.9855 (8.8287/2.2748) mem 68106MB [2022-12-19 10:42:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1390/1519] eta 0:02:09 lr 0.000032 time 0.9227 (1.0043) model_time 0.9225 (1.0036) loss 1.2693 (1.0035) grad_norm 6.2703 (8.8272/2.2826) mem 68106MB [2022-12-19 10:42:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1400/1519] eta 0:01:59 lr 0.000032 time 0.9336 (1.0043) model_time 0.9334 (1.0036) loss 0.9898 (1.0034) grad_norm 6.8247 (8.7900/2.2287) mem 68106MB [2022-12-19 10:42:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1410/1519] eta 0:01:49 lr 0.000032 time 0.9341 (1.0043) model_time 0.9339 (1.0036) loss 0.9096 (1.0028) grad_norm 6.9264 (8.7715/2.2052) mem 68106MB [2022-12-19 10:42:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1420/1519] eta 0:01:39 lr 0.000032 time 0.9254 (1.0043) model_time 0.9250 (1.0036) loss 1.0295 (1.0027) grad_norm 12.0386 (8.8303/2.2487) mem 68106MB [2022-12-19 10:42:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1430/1519] eta 0:01:29 lr 0.000032 time 0.9314 (1.0043) model_time 0.9313 (1.0036) loss 0.8599 (1.0027) grad_norm 9.8546 (8.8434/2.2455) mem 68106MB [2022-12-19 10:43:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1440/1519] eta 0:01:19 lr 0.000032 time 0.9348 (1.0043) model_time 0.9346 (1.0036) loss 1.0829 (1.0024) grad_norm 8.8853 (8.8361/2.2447) mem 68106MB [2022-12-19 10:43:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1450/1519] eta 0:01:09 lr 0.000032 time 0.9344 (1.0042) model_time 0.9342 (1.0036) loss 0.7573 (1.0023) grad_norm 9.6432 (8.8687/2.2303) mem 68106MB [2022-12-19 10:43:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1460/1519] eta 0:00:59 lr 0.000032 time 0.9976 (1.0043) model_time 0.9974 (1.0036) loss 0.7710 (1.0023) grad_norm 8.7484 (8.8733/2.2201) mem 68106MB [2022-12-19 10:43:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1470/1519] eta 0:00:49 lr 0.000032 time 0.9395 (1.0043) model_time 0.9394 (1.0036) loss 0.9005 (1.0024) grad_norm 6.7584 (8.8614/2.2242) mem 68106MB [2022-12-19 10:43:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1480/1519] eta 0:00:39 lr 0.000032 time 0.9311 (1.0043) model_time 0.9308 (1.0036) loss 0.7661 (1.0024) grad_norm 8.3686 (8.8517/2.2105) mem 68106MB [2022-12-19 10:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1490/1519] eta 0:00:29 lr 0.000032 time 0.9250 (1.0043) model_time 0.9249 (1.0036) loss 0.9382 (1.0024) grad_norm 8.0803 (8.8706/2.2202) mem 68106MB [2022-12-19 10:44:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1500/1519] eta 0:00:19 lr 0.000032 time 0.9181 (1.0043) model_time 0.9180 (1.0036) loss 0.8881 (1.0028) grad_norm 12.7351 (8.8389/2.1430) mem 68106MB [2022-12-19 10:44:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [14/100][1510/1519] eta 0:00:09 lr 0.000032 time 0.9239 (1.0042) model_time 0.9238 (1.0036) loss 0.9649 (1.0027) grad_norm 10.3915 (8.8042/2.1256) mem 68106MB [2022-12-19 10:44:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 14 training takes 0:25:25 [2022-12-19 10:44:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_14.pth saving...... [2022-12-19 10:44:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_14.pth saved !!! [2022-12-19 10:44:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.673 (0.673) Loss 0.6348 (0.6348) Acc@1 87.847 (87.847) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 10:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.304 (0.332) Loss 0.6394 (0.6207) Acc@1 89.583 (88.258) Acc@5 97.569 (97.664) Mem 68106MB [2022-12-19 10:44:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.316) Loss 0.5808 (0.6186) Acc@1 89.236 (87.996) Acc@5 98.264 (97.652) Mem 68106MB [2022-12-19 10:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6911 (0.6229) Acc@1 87.500 (87.679) Acc@5 96.875 (97.603) Mem 68106MB [2022-12-19 10:45:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.294 (0.307) Loss 0.6307 (0.6145) Acc@1 86.458 (87.890) Acc@5 97.222 (97.747) Mem 68106MB [2022-12-19 10:45:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 0.6249 (0.6090) Acc@1 87.153 (88.092) Acc@5 97.917 (97.787) Mem 68106MB [2022-12-19 10:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.6338 (0.6090) Acc@1 89.236 (88.109) Acc@5 97.222 (97.774) Mem 68106MB [2022-12-19 10:45:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.6826 (0.6103) Acc@1 86.806 (88.038) Acc@5 98.264 (97.804) Mem 68106MB [2022-12-19 10:45:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.298 (0.302) Loss 0.5395 (0.6092) Acc@1 89.583 (88.049) Acc@5 97.917 (97.822) Mem 68106MB [2022-12-19 10:45:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:14] * Acc@1 88.069 Acc@5 97.827 [2022-12-19 10:45:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 88.1% [2022-12-19 10:45:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 10:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 10:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 88.07% [2022-12-19 10:45:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][0/1519] eta 0:34:32 lr 0.000032 time 1.3646 (1.3646) model_time 0.9787 (0.9787) loss 1.0695 (1.0695) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 10:45:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][10/1519] eta 0:25:56 lr 0.000032 time 0.9218 (1.0315) model_time 0.9217 (0.9961) loss 1.0627 (1.0431) grad_norm 11.2235 (8.0556/1.7742) mem 68106MB [2022-12-19 10:45:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][20/1519] eta 0:25:23 lr 0.000032 time 0.9292 (1.0163) model_time 0.9291 (0.9977) loss 0.8415 (0.9845) grad_norm 6.2273 (7.3441/1.5700) mem 68106MB [2022-12-19 10:46:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][30/1519] eta 0:25:11 lr 0.000032 time 0.9224 (1.0152) model_time 0.9222 (1.0025) loss 0.9981 (0.9895) grad_norm 12.4846 (8.2553/2.0939) mem 68106MB [2022-12-19 10:46:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][40/1519] eta 0:25:09 lr 0.000032 time 0.9514 (1.0206) model_time 0.9512 (1.0109) loss 1.2329 (0.9842) grad_norm 8.2902 (8.3371/1.9316) mem 68106MB [2022-12-19 10:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][50/1519] eta 0:24:53 lr 0.000032 time 0.9262 (1.0169) model_time 0.9260 (1.0090) loss 0.8123 (0.9853) grad_norm 7.3387 (8.2067/1.8157) mem 68106MB [2022-12-19 10:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][60/1519] eta 0:24:39 lr 0.000032 time 0.9192 (1.0139) model_time 0.9191 (1.0073) loss 1.0417 (1.0042) grad_norm 9.4951 (8.4280/1.8461) mem 68106MB [2022-12-19 10:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][70/1519] eta 0:24:26 lr 0.000032 time 0.9375 (1.0123) model_time 0.9374 (1.0065) loss 0.8784 (0.9946) grad_norm 12.4961 (8.5538/1.8569) mem 68106MB [2022-12-19 10:47:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][80/1519] eta 0:24:16 lr 0.000032 time 1.0009 (1.0123) model_time 1.0007 (1.0072) loss 0.9555 (0.9946) grad_norm 9.5896 (8.7725/2.2825) mem 68106MB [2022-12-19 10:47:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][90/1519] eta 0:24:04 lr 0.000032 time 0.9183 (1.0109) model_time 0.9181 (1.0064) loss 0.9007 (0.9979) grad_norm 6.7150 (8.6806/2.1858) mem 68106MB [2022-12-19 10:47:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][100/1519] eta 0:23:54 lr 0.000032 time 0.9407 (1.0109) model_time 0.9405 (1.0068) loss 0.8762 (0.9905) grad_norm 11.6521 (8.7978/2.3037) mem 68106MB [2022-12-19 10:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][110/1519] eta 0:23:43 lr 0.000032 time 0.9294 (1.0101) model_time 0.9292 (1.0063) loss 0.8612 (0.9882) grad_norm 7.1655 (8.7187/2.2591) mem 68106MB [2022-12-19 10:47:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][120/1519] eta 0:23:32 lr 0.000032 time 0.9545 (1.0096) model_time 0.9543 (1.0061) loss 0.8026 (0.9920) grad_norm 9.0451 (8.7080/2.2459) mem 68106MB [2022-12-19 10:47:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][130/1519] eta 0:23:21 lr 0.000032 time 0.9343 (1.0093) model_time 0.9341 (1.0061) loss 0.7368 (0.9883) grad_norm 9.0253 (8.7036/2.1625) mem 68106MB [2022-12-19 10:48:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][140/1519] eta 0:23:11 lr 0.000032 time 0.9257 (1.0088) model_time 0.9254 (1.0057) loss 1.1179 (0.9908) grad_norm 6.4318 (8.6541/2.1368) mem 68106MB [2022-12-19 10:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][150/1519] eta 0:23:00 lr 0.000032 time 0.9961 (1.0086) model_time 0.9960 (1.0057) loss 1.1265 (0.9912) grad_norm 7.1864 (8.5938/2.0932) mem 68106MB [2022-12-19 10:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][160/1519] eta 0:22:50 lr 0.000032 time 0.9331 (1.0087) model_time 0.9329 (1.0060) loss 0.9393 (0.9881) grad_norm 14.1059 (8.6653/2.1392) mem 68106MB [2022-12-19 10:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][170/1519] eta 0:22:40 lr 0.000032 time 0.9303 (1.0084) model_time 0.9302 (1.0058) loss 1.0626 (0.9935) grad_norm 10.0904 (8.6579/2.1201) mem 68106MB [2022-12-19 10:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][180/1519] eta 0:22:29 lr 0.000032 time 0.9322 (1.0080) model_time 0.9321 (1.0055) loss 0.9463 (0.9946) grad_norm 5.8896 (8.5928/2.0957) mem 68106MB [2022-12-19 10:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][190/1519] eta 0:22:19 lr 0.000032 time 0.9342 (1.0076) model_time 0.9341 (1.0053) loss 1.5525 (0.9966) grad_norm 9.5316 (8.5947/2.0742) mem 68106MB [2022-12-19 10:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][200/1519] eta 0:22:08 lr 0.000032 time 0.9299 (1.0073) model_time 0.9298 (1.0050) loss 0.7946 (0.9903) grad_norm 6.5831 (8.5967/2.0531) mem 68106MB [2022-12-19 10:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][210/1519] eta 0:21:57 lr 0.000032 time 0.9240 (1.0068) model_time 0.9238 (1.0047) loss 0.8633 (0.9901) grad_norm 14.0230 (8.7268/2.1604) mem 68106MB [2022-12-19 10:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][220/1519] eta 0:21:47 lr 0.000032 time 0.9285 (1.0068) model_time 0.9284 (1.0047) loss 0.9571 (0.9926) grad_norm 7.4804 (8.7287/2.1303) mem 68106MB [2022-12-19 10:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][230/1519] eta 0:21:37 lr 0.000032 time 0.9205 (1.0064) model_time 0.9204 (1.0044) loss 0.9496 (0.9936) grad_norm 14.0731 (8.7739/2.1805) mem 68106MB [2022-12-19 10:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][240/1519] eta 0:21:26 lr 0.000032 time 0.9284 (1.0062) model_time 0.9283 (1.0042) loss 0.8145 (0.9939) grad_norm 9.1520 (8.8442/2.2338) mem 68106MB [2022-12-19 10:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][250/1519] eta 0:21:16 lr 0.000032 time 0.9262 (1.0059) model_time 0.9261 (1.0040) loss 0.8588 (0.9942) grad_norm 7.8433 (8.8189/2.1970) mem 68106MB [2022-12-19 10:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][260/1519] eta 0:21:06 lr 0.000032 time 0.9651 (1.0061) model_time 0.9650 (1.0043) loss 1.0908 (0.9940) grad_norm 7.7535 (8.8095/2.1809) mem 68106MB [2022-12-19 10:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][270/1519] eta 0:20:56 lr 0.000032 time 0.9222 (1.0059) model_time 0.9221 (1.0041) loss 1.1449 (0.9927) grad_norm 7.4558 (8.7824/2.1457) mem 68106MB [2022-12-19 10:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][280/1519] eta 0:20:46 lr 0.000032 time 0.9387 (1.0060) model_time 0.9385 (1.0043) loss 1.6468 (0.9962) grad_norm 9.8881 (8.7462/2.1301) mem 68106MB [2022-12-19 10:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][290/1519] eta 0:20:36 lr 0.000032 time 0.9309 (1.0058) model_time 0.9308 (1.0042) loss 0.9494 (0.9969) grad_norm 13.6359 (8.8172/2.2176) mem 68106MB [2022-12-19 10:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][300/1519] eta 0:20:26 lr 0.000032 time 0.9354 (1.0058) model_time 0.9353 (1.0041) loss 1.0666 (0.9968) grad_norm 9.3971 (8.8403/2.2148) mem 68106MB [2022-12-19 10:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][310/1519] eta 0:20:15 lr 0.000032 time 0.8976 (1.0056) model_time 0.8974 (1.0040) loss 0.9381 (0.9973) grad_norm 9.0185 (8.8630/2.2477) mem 68106MB [2022-12-19 10:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][320/1519] eta 0:20:05 lr 0.000032 time 0.9262 (1.0056) model_time 0.9260 (1.0041) loss 0.9200 (0.9981) grad_norm 10.3880 (8.8306/2.2358) mem 68106MB [2022-12-19 10:51:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][330/1519] eta 0:19:56 lr 0.000032 time 1.2356 (1.0064) model_time 1.2355 (1.0049) loss 0.8324 (0.9984) grad_norm 13.1233 (8.9142/2.3193) mem 68106MB [2022-12-19 10:51:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][340/1519] eta 0:19:47 lr 0.000032 time 0.9275 (1.0071) model_time 0.9274 (1.0056) loss 0.9994 (0.9968) grad_norm 8.4462 (8.9122/2.3315) mem 68106MB [2022-12-19 10:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][350/1519] eta 0:19:37 lr 0.000032 time 0.9248 (1.0072) model_time 0.9247 (1.0057) loss 1.1376 (0.9986) grad_norm 6.3887 (8.8409/2.3432) mem 68106MB [2022-12-19 10:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][360/1519] eta 0:19:27 lr 0.000032 time 0.9201 (1.0071) model_time 0.9200 (1.0057) loss 1.0188 (1.0020) grad_norm 9.5112 (8.8099/2.3249) mem 68106MB [2022-12-19 10:51:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][370/1519] eta 0:19:16 lr 0.000032 time 0.9261 (1.0069) model_time 0.9259 (1.0055) loss 0.7766 (0.9990) grad_norm 8.3271 (8.8032/2.3518) mem 68106MB [2022-12-19 10:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][380/1519] eta 0:19:06 lr 0.000032 time 0.9264 (1.0068) model_time 0.9263 (1.0055) loss 0.8930 (0.9989) grad_norm 7.1075 (8.7546/2.3451) mem 68106MB [2022-12-19 10:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][390/1519] eta 0:18:56 lr 0.000032 time 0.9222 (1.0069) model_time 0.9220 (1.0056) loss 0.7532 (0.9996) grad_norm 12.0436 (8.7857/2.3543) mem 68106MB [2022-12-19 10:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][400/1519] eta 0:18:46 lr 0.000032 time 0.9215 (1.0069) model_time 0.9214 (1.0056) loss 0.9114 (0.9991) grad_norm 8.5157 (8.8154/2.3752) mem 68106MB [2022-12-19 10:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][410/1519] eta 0:18:37 lr 0.000032 time 0.9083 (1.0074) model_time 0.9082 (1.0062) loss 0.9548 (0.9999) grad_norm 9.2705 (8.8004/2.3637) mem 68106MB [2022-12-19 10:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][420/1519] eta 0:18:27 lr 0.000032 time 0.9260 (1.0074) model_time 0.9259 (1.0062) loss 0.8735 (0.9987) grad_norm 6.5745 (8.7865/2.3556) mem 68106MB [2022-12-19 10:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][430/1519] eta 0:18:17 lr 0.000032 time 0.9283 (1.0074) model_time 0.9282 (1.0061) loss 0.7558 (0.9994) grad_norm 5.2967 (8.7735/2.3781) mem 68106MB [2022-12-19 10:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][440/1519] eta 0:18:06 lr 0.000032 time 0.9362 (1.0072) model_time 0.9361 (1.0060) loss 1.0546 (0.9985) grad_norm 5.3668 (8.7847/2.3744) mem 68106MB [2022-12-19 10:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][450/1519] eta 0:17:56 lr 0.000032 time 0.9286 (1.0072) model_time 0.9285 (1.0060) loss 1.5262 (1.0000) grad_norm 8.5618 (8.7796/2.3555) mem 68106MB [2022-12-19 10:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][460/1519] eta 0:17:46 lr 0.000032 time 0.9381 (1.0072) model_time 0.9380 (1.0060) loss 0.8943 (1.0011) grad_norm 6.7060 (8.7972/2.3507) mem 68106MB [2022-12-19 10:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][470/1519] eta 0:17:36 lr 0.000032 time 0.9046 (1.0071) model_time 0.9045 (1.0060) loss 0.8083 (1.0003) grad_norm 9.6350 (8.8191/2.3680) mem 68106MB [2022-12-19 10:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][480/1519] eta 0:17:26 lr 0.000032 time 0.9277 (1.0071) model_time 0.9276 (1.0060) loss 1.0687 (0.9989) grad_norm 8.9377 (8.8196/2.3467) mem 68106MB [2022-12-19 10:53:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][490/1519] eta 0:17:16 lr 0.000032 time 0.9251 (1.0070) model_time 0.9250 (1.0059) loss 1.0046 (0.9971) grad_norm 8.8889 (8.8204/2.3399) mem 68106MB [2022-12-19 10:54:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][500/1519] eta 0:17:05 lr 0.000032 time 0.9157 (1.0068) model_time 0.9155 (1.0057) loss 1.1745 (0.9974) grad_norm 11.0077 (8.8282/2.3228) mem 68106MB [2022-12-19 10:54:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][510/1519] eta 0:16:55 lr 0.000032 time 0.9377 (1.0069) model_time 0.9376 (1.0058) loss 0.8336 (0.9957) grad_norm 17.9972 (8.9301/2.4706) mem 68106MB [2022-12-19 10:54:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][520/1519] eta 0:16:45 lr 0.000032 time 0.9323 (1.0069) model_time 0.9322 (1.0058) loss 1.4768 (0.9967) grad_norm 9.5169 (8.9175/2.4513) mem 68106MB [2022-12-19 10:54:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][530/1519] eta 0:16:35 lr 0.000032 time 0.9315 (1.0070) model_time 0.9313 (1.0059) loss 0.8354 (0.9967) grad_norm 6.6287 (8.8912/2.4420) mem 68106MB [2022-12-19 10:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][540/1519] eta 0:16:25 lr 0.000032 time 0.9274 (1.0069) model_time 0.9272 (1.0058) loss 0.9105 (0.9966) grad_norm 9.8988 (8.9265/2.4365) mem 68106MB [2022-12-19 10:54:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][550/1519] eta 0:16:15 lr 0.000032 time 0.9440 (1.0068) model_time 0.9438 (1.0058) loss 1.0055 (0.9958) grad_norm 7.0662 (8.9104/2.4247) mem 68106MB [2022-12-19 10:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][560/1519] eta 0:16:05 lr 0.000032 time 0.9242 (1.0067) model_time 0.9241 (1.0057) loss 0.7589 (0.9948) grad_norm 12.2722 (8.9263/2.4162) mem 68106MB [2022-12-19 10:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][570/1519] eta 0:15:55 lr 0.000032 time 0.9304 (1.0067) model_time 0.9303 (1.0057) loss 1.0517 (0.9935) grad_norm 21.0234 (8.9668/2.5168) mem 68106MB [2022-12-19 10:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][580/1519] eta 0:15:45 lr 0.000032 time 0.9222 (1.0069) model_time 0.9220 (1.0059) loss 0.7984 (0.9932) grad_norm 7.2812 (8.9258/2.5178) mem 68106MB [2022-12-19 10:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][590/1519] eta 0:15:35 lr 0.000032 time 0.9345 (1.0068) model_time 0.9344 (1.0058) loss 0.8494 (0.9925) grad_norm 8.0037 (8.9515/2.5291) mem 68106MB [2022-12-19 10:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][600/1519] eta 0:15:25 lr 0.000032 time 0.9362 (1.0069) model_time 0.9361 (1.0059) loss 1.0704 (0.9919) grad_norm 7.3194 (8.9482/2.5160) mem 68106MB [2022-12-19 10:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][610/1519] eta 0:15:15 lr 0.000032 time 0.9720 (1.0068) model_time 0.9719 (1.0059) loss 0.8627 (0.9915) grad_norm 11.7155 (8.9765/2.5171) mem 68106MB [2022-12-19 10:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][620/1519] eta 0:15:05 lr 0.000032 time 0.9323 (1.0067) model_time 0.9322 (1.0057) loss 0.9297 (0.9895) grad_norm 9.6057 (9.0204/2.5153) mem 68106MB [2022-12-19 10:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][630/1519] eta 0:14:54 lr 0.000032 time 0.9320 (1.0067) model_time 0.9319 (1.0057) loss 0.9438 (0.9895) grad_norm 9.1512 (9.0028/2.5156) mem 68106MB [2022-12-19 10:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][640/1519] eta 0:14:44 lr 0.000032 time 0.9232 (1.0066) model_time 0.9231 (1.0056) loss 1.0050 (0.9885) grad_norm 7.2198 (9.0207/2.5588) mem 68106MB [2022-12-19 10:56:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][650/1519] eta 0:14:34 lr 0.000032 time 0.9295 (1.0066) model_time 0.9294 (1.0057) loss 1.0739 (0.9887) grad_norm 12.2963 (9.1196/2.7502) mem 68106MB [2022-12-19 10:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][660/1519] eta 0:14:24 lr 0.000032 time 0.9164 (1.0065) model_time 0.9163 (1.0056) loss 1.4050 (0.9909) grad_norm 8.1432 (9.0889/2.7515) mem 68106MB [2022-12-19 10:56:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][670/1519] eta 0:14:14 lr 0.000032 time 0.9212 (1.0064) model_time 0.9211 (1.0055) loss 0.8170 (0.9906) grad_norm 8.1811 (9.0696/2.7523) mem 68106MB [2022-12-19 10:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][680/1519] eta 0:14:04 lr 0.000032 time 0.9284 (1.0063) model_time 0.9283 (1.0054) loss 0.7720 (0.9905) grad_norm 6.2739 (9.0314/2.7247) mem 68106MB [2022-12-19 10:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][690/1519] eta 0:13:54 lr 0.000032 time 0.9275 (1.0062) model_time 0.9274 (1.0054) loss 0.9685 (0.9888) grad_norm 21.1477 (9.0876/2.8177) mem 68106MB [2022-12-19 10:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][700/1519] eta 0:13:44 lr 0.000032 time 0.9285 (1.0062) model_time 0.9283 (1.0053) loss 0.9444 (0.9888) grad_norm 7.0376 (9.0931/2.8140) mem 68106MB [2022-12-19 10:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][710/1519] eta 0:13:33 lr 0.000032 time 0.9206 (1.0061) model_time 0.9205 (1.0052) loss 1.3291 (0.9906) grad_norm 6.9098 (9.1613/2.8722) mem 68106MB [2022-12-19 10:57:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][720/1519] eta 0:13:23 lr 0.000032 time 0.9256 (1.0061) model_time 0.9255 (1.0052) loss 0.8277 (0.9895) grad_norm 7.0380 (9.1434/2.8670) mem 68106MB [2022-12-19 10:57:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][730/1519] eta 0:13:13 lr 0.000032 time 0.9216 (1.0063) model_time 0.9214 (1.0054) loss 1.0377 (0.9895) grad_norm 8.5580 (9.1356/2.8721) mem 68106MB [2022-12-19 10:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][740/1519] eta 0:13:03 lr 0.000032 time 0.9203 (1.0062) model_time 0.9202 (1.0053) loss 1.2282 (0.9880) grad_norm 15.3415 (9.1571/2.8874) mem 68106MB [2022-12-19 10:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][750/1519] eta 0:12:53 lr 0.000032 time 0.9211 (1.0061) model_time 0.9210 (1.0053) loss 1.0663 (0.9890) grad_norm 10.9608 (9.1952/2.9017) mem 68106MB [2022-12-19 10:58:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][760/1519] eta 0:12:43 lr 0.000032 time 0.9333 (1.0060) model_time 0.9332 (1.0052) loss 1.0810 (0.9893) grad_norm 8.6036 (9.1821/2.8901) mem 68106MB [2022-12-19 10:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][770/1519] eta 0:12:33 lr 0.000032 time 0.9117 (1.0061) model_time 0.9115 (1.0053) loss 0.8053 (0.9889) grad_norm 10.2307 (9.2040/2.8881) mem 68106MB [2022-12-19 10:58:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][780/1519] eta 0:12:23 lr 0.000032 time 0.9722 (1.0060) model_time 0.9720 (1.0052) loss 0.9586 (0.9882) grad_norm 6.7781 (9.2136/2.8935) mem 68106MB [2022-12-19 10:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][790/1519] eta 0:12:13 lr 0.000032 time 0.9326 (1.0061) model_time 0.9324 (1.0053) loss 1.0714 (0.9882) grad_norm 10.8798 (9.2401/2.8963) mem 68106MB [2022-12-19 10:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][800/1519] eta 0:12:03 lr 0.000032 time 0.9403 (1.0060) model_time 0.9401 (1.0052) loss 1.0327 (0.9880) grad_norm 9.8703 (9.2270/2.8969) mem 68106MB [2022-12-19 10:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][810/1519] eta 0:11:53 lr 0.000032 time 0.9362 (1.0060) model_time 0.9359 (1.0052) loss 0.8921 (0.9875) grad_norm 9.3137 (9.1895/2.8734) mem 68106MB [2022-12-19 10:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][820/1519] eta 0:11:43 lr 0.000032 time 0.9372 (1.0059) model_time 0.9371 (1.0051) loss 0.7953 (0.9876) grad_norm 7.7196 (9.1764/2.8821) mem 68106MB [2022-12-19 10:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][830/1519] eta 0:11:33 lr 0.000032 time 0.9179 (1.0059) model_time 0.9178 (1.0051) loss 0.9460 (0.9884) grad_norm 7.5054 (9.1284/2.8732) mem 68106MB [2022-12-19 10:59:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][840/1519] eta 0:11:23 lr 0.000032 time 0.9321 (1.0061) model_time 0.9319 (1.0053) loss 0.7930 (0.9882) grad_norm 8.5937 (9.0902/2.8561) mem 68106MB [2022-12-19 10:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][850/1519] eta 0:11:13 lr 0.000032 time 0.9258 (1.0061) model_time 0.9257 (1.0053) loss 0.9305 (0.9879) grad_norm 8.1444 (9.1066/2.8716) mem 68106MB [2022-12-19 11:00:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][860/1519] eta 0:11:02 lr 0.000032 time 0.9318 (1.0060) model_time 0.9316 (1.0052) loss 1.1307 (0.9882) grad_norm 10.0204 (9.1162/2.8686) mem 68106MB [2022-12-19 11:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][870/1519] eta 0:10:52 lr 0.000032 time 0.9284 (1.0059) model_time 0.9283 (1.0052) loss 1.1943 (0.9875) grad_norm 21.4769 (9.1576/2.9731) mem 68106MB [2022-12-19 11:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][880/1519] eta 0:10:42 lr 0.000032 time 0.9153 (1.0059) model_time 0.9151 (1.0051) loss 1.1782 (0.9879) grad_norm 8.7220 (9.1815/2.9724) mem 68106MB [2022-12-19 11:00:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][890/1519] eta 0:10:32 lr 0.000032 time 0.9209 (1.0060) model_time 0.9208 (1.0052) loss 0.7372 (0.9873) grad_norm 8.5560 (9.1566/2.9424) mem 68106MB [2022-12-19 11:00:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][900/1519] eta 0:10:22 lr 0.000032 time 0.9263 (1.0059) model_time 0.9262 (1.0051) loss 0.7342 (0.9867) grad_norm 16.0843 (9.1854/2.9646) mem 68106MB [2022-12-19 11:00:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][910/1519] eta 0:10:12 lr 0.000032 time 0.8929 (1.0059) model_time 0.8927 (1.0052) loss 1.1871 (0.9875) grad_norm 9.3926 (9.1842/2.9562) mem 68106MB [2022-12-19 11:01:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][920/1519] eta 0:10:02 lr 0.000032 time 0.9404 (1.0059) model_time 0.9402 (1.0051) loss 1.1220 (0.9870) grad_norm 8.5062 (9.2012/2.9612) mem 68106MB [2022-12-19 11:01:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][930/1519] eta 0:09:52 lr 0.000032 time 0.9307 (1.0058) model_time 0.9305 (1.0050) loss 1.0876 (0.9876) grad_norm 7.4444 (9.1574/2.9345) mem 68106MB [2022-12-19 11:01:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][940/1519] eta 0:09:42 lr 0.000032 time 0.9259 (1.0057) model_time 0.9258 (1.0050) loss 0.8255 (0.9874) grad_norm 6.6875 (9.1385/2.9212) mem 68106MB [2022-12-19 11:01:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][950/1519] eta 0:09:32 lr 0.000032 time 0.9261 (1.0057) model_time 0.9260 (1.0049) loss 1.1109 (0.9883) grad_norm 9.5697 (9.2314/2.9434) mem 68106MB [2022-12-19 11:01:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][960/1519] eta 0:09:22 lr 0.000032 time 0.9246 (1.0059) model_time 0.9244 (1.0051) loss 1.0733 (0.9881) grad_norm 9.8524 (9.2622/2.9378) mem 68106MB [2022-12-19 11:01:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][970/1519] eta 0:09:12 lr 0.000032 time 0.9356 (1.0058) model_time 0.9355 (1.0051) loss 1.1938 (0.9890) grad_norm 11.7120 (9.2917/2.9522) mem 68106MB [2022-12-19 11:02:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][980/1519] eta 0:09:02 lr 0.000032 time 0.9395 (1.0058) model_time 0.9394 (1.0051) loss 0.8456 (0.9890) grad_norm 7.4095 (9.3012/2.9488) mem 68106MB [2022-12-19 11:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][990/1519] eta 0:08:52 lr 0.000032 time 0.9290 (1.0058) model_time 0.9288 (1.0050) loss 0.8008 (0.9879) grad_norm 7.5546 (9.3158/2.9934) mem 68106MB [2022-12-19 11:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1000/1519] eta 0:08:41 lr 0.000032 time 0.9267 (1.0057) model_time 0.9266 (1.0050) loss 1.0963 (0.9878) grad_norm 10.7199 (9.3142/2.9777) mem 68106MB [2022-12-19 11:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1010/1519] eta 0:08:31 lr 0.000032 time 0.9984 (1.0057) model_time 0.9982 (1.0050) loss 0.8452 (0.9879) grad_norm 8.5785 (9.3259/2.9747) mem 68106MB [2022-12-19 11:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1020/1519] eta 0:08:21 lr 0.000032 time 0.9249 (1.0057) model_time 0.9248 (1.0050) loss 0.7534 (0.9873) grad_norm 7.6507 (9.3272/2.9711) mem 68106MB [2022-12-19 11:02:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1030/1519] eta 0:08:11 lr 0.000032 time 0.9243 (1.0057) model_time 0.9235 (1.0050) loss 1.1193 (0.9883) grad_norm 7.5392 (9.3353/2.9591) mem 68106MB [2022-12-19 11:03:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1040/1519] eta 0:08:01 lr 0.000032 time 0.9283 (1.0059) model_time 0.9282 (1.0052) loss 1.1869 (0.9886) grad_norm 10.7740 (9.3534/2.9628) mem 68106MB [2022-12-19 11:03:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1050/1519] eta 0:07:51 lr 0.000032 time 0.9345 (1.0058) model_time 0.9344 (1.0051) loss 1.1223 (0.9890) grad_norm 6.3809 (9.3431/2.9670) mem 68106MB [2022-12-19 11:03:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1060/1519] eta 0:07:41 lr 0.000032 time 0.9301 (1.0058) model_time 0.9300 (1.0051) loss 1.1128 (0.9894) grad_norm 14.2623 (9.3557/2.9843) mem 68106MB [2022-12-19 11:03:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1070/1519] eta 0:07:31 lr 0.000032 time 0.9310 (1.0057) model_time 0.9309 (1.0050) loss 0.8091 (0.9901) grad_norm 9.8784 (9.3562/3.0105) mem 68106MB [2022-12-19 11:03:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1080/1519] eta 0:07:21 lr 0.000032 time 0.9202 (1.0056) model_time 0.9200 (1.0049) loss 0.8041 (0.9899) grad_norm 9.2017 (9.3546/3.0195) mem 68106MB [2022-12-19 11:03:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1090/1519] eta 0:07:11 lr 0.000032 time 0.9293 (1.0056) model_time 0.9292 (1.0049) loss 0.8293 (0.9891) grad_norm 7.6541 (9.3358/3.0173) mem 68106MB [2022-12-19 11:04:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1100/1519] eta 0:07:01 lr 0.000032 time 0.9234 (1.0056) model_time 0.9232 (1.0049) loss 0.8795 (0.9895) grad_norm 8.7936 (9.3176/3.0204) mem 68106MB [2022-12-19 11:04:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1110/1519] eta 0:06:51 lr 0.000032 time 0.9298 (1.0055) model_time 0.9297 (1.0048) loss 1.0125 (0.9894) grad_norm 6.9124 (9.2146/2.9272) mem 68106MB [2022-12-19 11:04:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1120/1519] eta 0:06:41 lr 0.000032 time 0.9247 (1.0055) model_time 0.9245 (1.0048) loss 0.9224 (0.9893) grad_norm 11.9658 (9.2120/2.9371) mem 68106MB [2022-12-19 11:04:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1130/1519] eta 0:06:31 lr 0.000032 time 0.9336 (1.0054) model_time 0.9333 (1.0048) loss 0.9540 (0.9898) grad_norm 10.2929 (9.2676/2.9469) mem 68106MB [2022-12-19 11:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1140/1519] eta 0:06:21 lr 0.000032 time 0.9274 (1.0054) model_time 0.9273 (1.0047) loss 1.0165 (0.9893) grad_norm 6.7212 (9.2118/2.9515) mem 68106MB [2022-12-19 11:04:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1150/1519] eta 0:06:11 lr 0.000032 time 1.2101 (1.0056) model_time 1.2100 (1.0049) loss 0.9049 (0.9891) grad_norm 7.7828 (9.2163/2.9465) mem 68106MB [2022-12-19 11:05:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1160/1519] eta 0:06:00 lr 0.000032 time 0.9300 (1.0055) model_time 0.9299 (1.0049) loss 0.8809 (0.9902) grad_norm 7.8061 (9.2003/2.9425) mem 68106MB [2022-12-19 11:05:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1170/1519] eta 0:05:50 lr 0.000032 time 0.9318 (1.0055) model_time 0.9316 (1.0048) loss 1.2016 (0.9910) grad_norm 10.7948 (9.1633/2.8533) mem 68106MB [2022-12-19 11:05:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1180/1519] eta 0:05:40 lr 0.000032 time 0.9397 (1.0055) model_time 0.9396 (1.0048) loss 1.0799 (0.9911) grad_norm 9.0586 (9.1909/2.8355) mem 68106MB [2022-12-19 11:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1190/1519] eta 0:05:30 lr 0.000032 time 0.9954 (1.0055) model_time 0.9953 (1.0048) loss 0.9273 (0.9910) grad_norm 5.4370 (9.1509/2.8218) mem 68106MB [2022-12-19 11:05:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1200/1519] eta 0:05:20 lr 0.000032 time 0.9197 (1.0055) model_time 0.9195 (1.0048) loss 0.8416 (0.9908) grad_norm 9.1822 (9.1333/2.8233) mem 68106MB [2022-12-19 11:05:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1210/1519] eta 0:05:10 lr 0.000032 time 0.9279 (1.0055) model_time 0.9278 (1.0048) loss 1.1164 (0.9911) grad_norm 5.7606 (9.1047/2.8200) mem 68106MB [2022-12-19 11:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1220/1519] eta 0:05:00 lr 0.000032 time 0.9498 (1.0055) model_time 0.9497 (1.0049) loss 0.7594 (0.9911) grad_norm 6.5624 (9.0658/2.8191) mem 68106MB [2022-12-19 11:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1230/1519] eta 0:04:50 lr 0.000032 time 0.9323 (1.0055) model_time 0.9321 (1.0048) loss 1.0648 (0.9915) grad_norm 10.3582 (9.0780/2.8194) mem 68106MB [2022-12-19 11:06:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1240/1519] eta 0:04:40 lr 0.000032 time 0.9346 (1.0055) model_time 0.9344 (1.0048) loss 0.8293 (0.9911) grad_norm 9.5145 (9.0495/2.7818) mem 68106MB [2022-12-19 11:06:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1250/1519] eta 0:04:30 lr 0.000032 time 0.9681 (1.0055) model_time 0.9680 (1.0048) loss 1.0745 (0.9915) grad_norm 7.5827 (8.9878/2.6099) mem 68106MB [2022-12-19 11:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1260/1519] eta 0:04:20 lr 0.000032 time 0.9219 (1.0054) model_time 0.9217 (1.0048) loss 1.3504 (0.9913) grad_norm 8.4422 (9.0052/2.6103) mem 68106MB [2022-12-19 11:06:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1270/1519] eta 0:04:10 lr 0.000032 time 0.9331 (1.0054) model_time 0.9329 (1.0047) loss 0.8594 (0.9911) grad_norm 6.0281 (8.9903/2.6158) mem 68106MB [2022-12-19 11:07:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1280/1519] eta 0:04:00 lr 0.000032 time 0.9334 (1.0054) model_time 0.9329 (1.0048) loss 0.7582 (0.9911) grad_norm 8.5891 (9.0134/2.6090) mem 68106MB [2022-12-19 11:07:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1290/1519] eta 0:03:50 lr 0.000032 time 0.9322 (1.0054) model_time 0.9320 (1.0048) loss 0.9105 (0.9909) grad_norm 11.9791 (8.9627/2.5152) mem 68106MB [2022-12-19 11:07:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1300/1519] eta 0:03:40 lr 0.000032 time 0.9301 (1.0054) model_time 0.9299 (1.0047) loss 1.1971 (0.9914) grad_norm 8.5366 (8.9292/2.5021) mem 68106MB [2022-12-19 11:07:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1310/1519] eta 0:03:30 lr 0.000032 time 0.9269 (1.0053) model_time 0.9263 (1.0047) loss 0.9249 (0.9914) grad_norm 12.1198 (8.8718/2.4308) mem 68106MB [2022-12-19 11:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1320/1519] eta 0:03:20 lr 0.000032 time 0.9367 (1.0053) model_time 0.9366 (1.0047) loss 1.4681 (0.9927) grad_norm 6.3208 (8.8793/2.4311) mem 68106MB [2022-12-19 11:07:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1330/1519] eta 0:03:09 lr 0.000032 time 0.9102 (1.0053) model_time 0.9100 (1.0046) loss 1.0297 (0.9929) grad_norm 6.1771 (8.9099/2.4500) mem 68106MB [2022-12-19 11:08:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1340/1519] eta 0:02:59 lr 0.000032 time 0.9266 (1.0053) model_time 0.9264 (1.0046) loss 0.9531 (0.9919) grad_norm 8.8997 (8.9417/2.4561) mem 68106MB [2022-12-19 11:08:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1350/1519] eta 0:02:49 lr 0.000032 time 0.9297 (1.0052) model_time 0.9296 (1.0046) loss 1.1263 (0.9919) grad_norm 20.1653 (8.9567/2.5153) mem 68106MB [2022-12-19 11:08:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1360/1519] eta 0:02:39 lr 0.000032 time 0.9178 (1.0052) model_time 0.9177 (1.0046) loss 1.1065 (0.9921) grad_norm 6.4314 (8.9604/2.5214) mem 68106MB [2022-12-19 11:08:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1370/1519] eta 0:02:29 lr 0.000032 time 0.9233 (1.0052) model_time 0.9232 (1.0045) loss 1.0549 (0.9924) grad_norm 7.5537 (8.9517/2.5201) mem 68106MB [2022-12-19 11:08:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1380/1519] eta 0:02:19 lr 0.000032 time 0.9296 (1.0051) model_time 0.9291 (1.0045) loss 0.8952 (0.9925) grad_norm 8.3187 (8.9434/2.5111) mem 68106MB [2022-12-19 11:08:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1390/1519] eta 0:02:09 lr 0.000032 time 0.9307 (1.0051) model_time 0.9305 (1.0044) loss 1.0608 (0.9925) grad_norm 10.1094 (8.9521/2.5367) mem 68106MB [2022-12-19 11:09:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1400/1519] eta 0:01:59 lr 0.000032 time 0.9326 (1.0051) model_time 0.9325 (1.0044) loss 0.9869 (0.9929) grad_norm 9.1085 (8.9457/2.5443) mem 68106MB [2022-12-19 11:09:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1410/1519] eta 0:01:49 lr 0.000032 time 0.9278 (1.0050) model_time 0.9277 (1.0044) loss 1.1135 (0.9931) grad_norm 13.6017 (8.9353/2.5596) mem 68106MB [2022-12-19 11:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1420/1519] eta 0:01:39 lr 0.000032 time 0.9363 (1.0050) model_time 0.9361 (1.0044) loss 0.9424 (0.9930) grad_norm 8.5537 (8.9294/2.5551) mem 68106MB [2022-12-19 11:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1430/1519] eta 0:01:29 lr 0.000032 time 0.9862 (1.0050) model_time 0.9861 (1.0044) loss 1.3003 (0.9930) grad_norm 7.1710 (8.9589/2.5518) mem 68106MB [2022-12-19 11:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1440/1519] eta 0:01:19 lr 0.000032 time 0.9309 (1.0050) model_time 0.9308 (1.0043) loss 1.0612 (0.9930) grad_norm 7.1182 (8.9567/2.5526) mem 68106MB [2022-12-19 11:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1450/1519] eta 0:01:09 lr 0.000032 time 0.9243 (1.0050) model_time 0.9242 (1.0043) loss 0.8458 (0.9931) grad_norm 12.4093 (8.9843/2.5678) mem 68106MB [2022-12-19 11:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1460/1519] eta 0:00:59 lr 0.000032 time 0.9517 (1.0050) model_time 0.9516 (1.0044) loss 1.3513 (0.9930) grad_norm 16.9684 (9.0176/2.6082) mem 68106MB [2022-12-19 11:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1470/1519] eta 0:00:49 lr 0.000032 time 0.9262 (1.0050) model_time 0.9261 (1.0044) loss 0.7062 (0.9927) grad_norm 18.1956 (9.0559/2.6117) mem 68106MB [2022-12-19 11:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1480/1519] eta 0:00:39 lr 0.000032 time 0.9371 (1.0050) model_time 0.9369 (1.0044) loss 1.0652 (0.9929) grad_norm 11.2263 (9.0936/2.6410) mem 68106MB [2022-12-19 11:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1490/1519] eta 0:00:29 lr 0.000032 time 0.9246 (1.0050) model_time 0.9244 (1.0044) loss 0.7265 (0.9929) grad_norm 13.7887 (9.1153/2.6536) mem 68106MB [2022-12-19 11:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1500/1519] eta 0:00:19 lr 0.000032 time 0.9207 (1.0050) model_time 0.9206 (1.0044) loss 1.0846 (0.9929) grad_norm 7.3417 (9.0588/2.6245) mem 68106MB [2022-12-19 11:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [15/100][1510/1519] eta 0:00:09 lr 0.000032 time 1.0168 (1.0051) model_time 1.0166 (1.0044) loss 1.1364 (0.9932) grad_norm 5.8715 (9.0462/2.6183) mem 68106MB [2022-12-19 11:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 15 training takes 0:25:26 [2022-12-19 11:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_15.pth saving...... [2022-12-19 11:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_15.pth saved !!! [2022-12-19 11:11:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.667 (0.667) Loss 0.5981 (0.5981) Acc@1 89.236 (89.236) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 11:11:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.331) Loss 0.6190 (0.5907) Acc@1 88.889 (89.173) Acc@5 97.569 (98.011) Mem 68106MB [2022-12-19 11:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.316) Loss 0.5641 (0.5880) Acc@1 90.278 (88.806) Acc@5 97.917 (97.867) Mem 68106MB [2022-12-19 11:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.303 (0.311) Loss 0.6602 (0.5920) Acc@1 87.500 (88.766) Acc@5 96.875 (97.816) Mem 68106MB [2022-12-19 11:11:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.308) Loss 0.5710 (0.5846) Acc@1 85.764 (88.855) Acc@5 97.569 (97.866) Mem 68106MB [2022-12-19 11:11:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.307) Loss 0.5893 (0.5803) Acc@1 86.111 (88.841) Acc@5 98.611 (97.923) Mem 68106MB [2022-12-19 11:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.306) Loss 0.5930 (0.5817) Acc@1 89.236 (88.758) Acc@5 97.222 (97.900) Mem 68106MB [2022-12-19 11:11:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.304 (0.305) Loss 0.6652 (0.5838) Acc@1 88.194 (88.757) Acc@5 97.917 (97.922) Mem 68106MB [2022-12-19 11:11:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.304) Loss 0.5172 (0.5840) Acc@1 90.972 (88.795) Acc@5 97.917 (97.908) Mem 68106MB [2022-12-19 11:11:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:15] * Acc@1 88.826 Acc@5 97.921 [2022-12-19 11:11:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 88.8% [2022-12-19 11:11:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 11:12:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 11:12:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 88.83% [2022-12-19 11:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][0/1519] eta 0:35:00 lr 0.000032 time 1.3826 (1.3826) model_time 0.9713 (0.9713) loss 0.9968 (0.9968) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 11:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][10/1519] eta 0:26:35 lr 0.000032 time 0.9336 (1.0574) model_time 0.9335 (1.0198) loss 0.7980 (0.9371) grad_norm 8.0992 (8.7832/2.5441) mem 68106MB [2022-12-19 11:12:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][20/1519] eta 0:25:46 lr 0.000032 time 0.9307 (1.0316) model_time 0.9305 (1.0117) loss 1.2427 (0.9632) grad_norm 8.2919 (8.7884/2.0783) mem 68106MB [2022-12-19 11:12:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][30/1519] eta 0:25:23 lr 0.000032 time 0.9249 (1.0229) model_time 0.9247 (1.0093) loss 0.9322 (0.9553) grad_norm 11.5961 (8.4830/2.0782) mem 68106MB [2022-12-19 11:13:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][40/1519] eta 0:25:06 lr 0.000032 time 0.9210 (1.0183) model_time 0.9209 (1.0080) loss 1.3607 (0.9534) grad_norm 8.9189 (8.6093/1.8865) mem 68106MB [2022-12-19 11:13:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][50/1519] eta 0:24:50 lr 0.000032 time 0.9268 (1.0145) model_time 0.9267 (1.0062) loss 0.8356 (0.9632) grad_norm 7.0302 (8.5343/1.7982) mem 68106MB [2022-12-19 11:13:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][60/1519] eta 0:24:36 lr 0.000032 time 0.9292 (1.0123) model_time 0.9290 (1.0052) loss 0.9880 (0.9643) grad_norm 8.7106 (8.8989/2.1830) mem 68106MB [2022-12-19 11:13:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][70/1519] eta 0:24:25 lr 0.000032 time 0.9671 (1.0114) model_time 0.9670 (1.0053) loss 0.7246 (0.9681) grad_norm 10.7197 (8.9033/2.1021) mem 68106MB [2022-12-19 11:13:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][80/1519] eta 0:24:13 lr 0.000032 time 0.9198 (1.0099) model_time 0.9197 (1.0045) loss 1.1983 (0.9714) grad_norm 6.3674 (8.9445/2.1401) mem 68106MB [2022-12-19 11:13:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][90/1519] eta 0:24:01 lr 0.000032 time 0.9195 (1.0089) model_time 0.9194 (1.0041) loss 0.7783 (0.9782) grad_norm 8.7621 (8.8390/2.0831) mem 68106MB [2022-12-19 11:14:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][100/1519] eta 0:23:50 lr 0.000032 time 0.9210 (1.0083) model_time 0.9209 (1.0040) loss 0.9246 (0.9814) grad_norm 9.3324 (8.9993/2.2328) mem 68106MB [2022-12-19 11:14:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][110/1519] eta 0:23:39 lr 0.000032 time 0.9216 (1.0075) model_time 0.9215 (1.0035) loss 0.8869 (0.9799) grad_norm 8.4274 (8.9433/2.1581) mem 68106MB [2022-12-19 11:14:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][120/1519] eta 0:23:28 lr 0.000032 time 0.9301 (1.0070) model_time 0.9299 (1.0033) loss 0.9648 (0.9806) grad_norm 6.1018 (8.8379/2.1320) mem 68106MB [2022-12-19 11:14:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][130/1519] eta 0:23:18 lr 0.000032 time 0.9427 (1.0071) model_time 0.9425 (1.0037) loss 1.1509 (0.9817) grad_norm 9.4069 (8.7863/2.0875) mem 68106MB [2022-12-19 11:14:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][140/1519] eta 0:23:08 lr 0.000032 time 0.9430 (1.0068) model_time 0.9427 (1.0036) loss 1.3299 (0.9851) grad_norm 9.9090 (8.9226/2.1533) mem 68106MB [2022-12-19 11:14:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][150/1519] eta 0:22:57 lr 0.000032 time 0.9167 (1.0065) model_time 0.9165 (1.0035) loss 0.9679 (0.9828) grad_norm 10.1720 (9.0804/2.2051) mem 68106MB [2022-12-19 11:15:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][160/1519] eta 0:22:47 lr 0.000032 time 0.9199 (1.0061) model_time 0.9197 (1.0032) loss 0.9362 (0.9809) grad_norm 9.7526 (9.0403/2.1635) mem 68106MB [2022-12-19 11:15:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][170/1519] eta 0:22:36 lr 0.000032 time 0.9255 (1.0058) model_time 0.9253 (1.0031) loss 0.8578 (0.9761) grad_norm 7.2818 (8.9651/2.2171) mem 68106MB [2022-12-19 11:15:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][180/1519] eta 0:22:26 lr 0.000032 time 0.9139 (1.0056) model_time 0.9137 (1.0030) loss 0.8074 (0.9775) grad_norm 7.2511 (8.9671/2.3416) mem 68106MB [2022-12-19 11:15:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][190/1519] eta 0:22:15 lr 0.000032 time 0.9224 (1.0052) model_time 0.9222 (1.0028) loss 0.8302 (0.9816) grad_norm 15.5153 (9.0771/2.3999) mem 68106MB [2022-12-19 11:15:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][200/1519] eta 0:22:05 lr 0.000032 time 0.9431 (1.0052) model_time 0.9430 (1.0029) loss 0.8453 (0.9823) grad_norm 14.1582 (9.1613/2.4091) mem 68106MB [2022-12-19 11:15:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][210/1519] eta 0:21:56 lr 0.000032 time 0.9243 (1.0056) model_time 0.9242 (1.0033) loss 1.0837 (0.9786) grad_norm 13.0635 (9.1547/2.4453) mem 68106MB [2022-12-19 11:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][220/1519] eta 0:21:47 lr 0.000032 time 0.9247 (1.0066) model_time 0.9245 (1.0044) loss 1.1514 (0.9801) grad_norm 10.5156 (9.1852/2.4577) mem 68106MB [2022-12-19 11:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][230/1519] eta 0:21:37 lr 0.000032 time 0.9201 (1.0064) model_time 0.9199 (1.0043) loss 1.0154 (0.9825) grad_norm 7.6543 (9.2362/2.5728) mem 68106MB [2022-12-19 11:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][240/1519] eta 0:21:26 lr 0.000032 time 0.9264 (1.0062) model_time 0.9263 (1.0042) loss 0.8137 (0.9802) grad_norm 6.6153 (9.1544/2.5573) mem 68106MB [2022-12-19 11:16:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][250/1519] eta 0:21:16 lr 0.000032 time 0.9308 (1.0061) model_time 0.9307 (1.0041) loss 0.8541 (0.9787) grad_norm 11.4147 (9.1265/2.5332) mem 68106MB [2022-12-19 11:16:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][260/1519] eta 0:21:06 lr 0.000032 time 0.9395 (1.0059) model_time 0.9393 (1.0040) loss 0.8509 (0.9743) grad_norm 10.4157 (9.0880/2.5194) mem 68106MB [2022-12-19 11:16:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][270/1519] eta 0:20:56 lr 0.000032 time 0.9371 (1.0056) model_time 0.9370 (1.0038) loss 1.4337 (0.9729) grad_norm 7.4766 (9.0493/2.4901) mem 68106MB [2022-12-19 11:17:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][280/1519] eta 0:20:45 lr 0.000032 time 0.9297 (1.0054) model_time 0.9295 (1.0036) loss 0.9622 (0.9735) grad_norm 11.8420 (9.0861/2.4799) mem 68106MB [2022-12-19 11:17:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][290/1519] eta 0:20:35 lr 0.000032 time 0.9253 (1.0051) model_time 0.9252 (1.0034) loss 0.8624 (0.9771) grad_norm 14.5915 (9.1205/2.5029) mem 68106MB [2022-12-19 11:17:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][300/1519] eta 0:20:25 lr 0.000032 time 0.9182 (1.0056) model_time 0.9181 (1.0039) loss 1.1322 (0.9755) grad_norm 9.7541 (9.1677/2.5198) mem 68106MB [2022-12-19 11:17:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][310/1519] eta 0:20:15 lr 0.000032 time 0.9210 (1.0056) model_time 0.9208 (1.0039) loss 1.0095 (0.9768) grad_norm 10.5735 (9.1836/2.4852) mem 68106MB [2022-12-19 11:17:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][320/1519] eta 0:20:05 lr 0.000032 time 1.0377 (1.0057) model_time 1.0376 (1.0041) loss 0.8679 (0.9794) grad_norm 9.3178 (9.1342/2.4707) mem 68106MB [2022-12-19 11:17:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][330/1519] eta 0:19:55 lr 0.000032 time 0.9210 (1.0055) model_time 0.9209 (1.0039) loss 0.8383 (0.9812) grad_norm 6.6654 (9.1201/2.4573) mem 68106MB [2022-12-19 11:18:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][340/1519] eta 0:19:45 lr 0.000032 time 0.9195 (1.0054) model_time 0.9193 (1.0038) loss 0.9744 (0.9798) grad_norm 10.9377 (9.1706/2.4909) mem 68106MB [2022-12-19 11:18:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][350/1519] eta 0:19:35 lr 0.000032 time 0.9207 (1.0053) model_time 0.9204 (1.0038) loss 1.0850 (0.9828) grad_norm 9.2605 (9.1592/2.4739) mem 68106MB [2022-12-19 11:18:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][360/1519] eta 0:19:24 lr 0.000032 time 0.9252 (1.0052) model_time 0.9250 (1.0037) loss 1.0307 (0.9836) grad_norm 7.5598 (9.1719/2.5158) mem 68106MB [2022-12-19 11:18:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][370/1519] eta 0:19:14 lr 0.000032 time 0.9310 (1.0050) model_time 0.9308 (1.0036) loss 0.7784 (0.9836) grad_norm 13.6863 (9.1888/2.5058) mem 68106MB [2022-12-19 11:18:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][380/1519] eta 0:19:04 lr 0.000032 time 0.9846 (1.0050) model_time 0.9845 (1.0036) loss 1.7803 (0.9859) grad_norm 6.7236 (9.1648/2.4856) mem 68106MB [2022-12-19 11:18:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][390/1519] eta 0:18:54 lr 0.000032 time 0.9219 (1.0049) model_time 0.9218 (1.0035) loss 0.8631 (0.9851) grad_norm 8.6764 (9.1390/2.4670) mem 68106MB [2022-12-19 11:19:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][400/1519] eta 0:18:44 lr 0.000032 time 0.9305 (1.0048) model_time 0.9304 (1.0034) loss 0.7070 (0.9856) grad_norm 14.2099 (9.1184/2.4902) mem 68106MB [2022-12-19 11:19:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][410/1519] eta 0:18:34 lr 0.000032 time 0.9580 (1.0047) model_time 0.9578 (1.0034) loss 0.7136 (0.9820) grad_norm 7.1027 (9.1550/2.5785) mem 68106MB [2022-12-19 11:19:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][420/1519] eta 0:18:24 lr 0.000032 time 0.9218 (1.0047) model_time 0.9216 (1.0033) loss 1.1101 (0.9811) grad_norm 9.1896 (9.1392/2.5569) mem 68106MB [2022-12-19 11:19:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][430/1519] eta 0:18:13 lr 0.000032 time 0.9227 (1.0046) model_time 0.9226 (1.0033) loss 0.8106 (0.9802) grad_norm 7.6146 (9.2200/2.8812) mem 68106MB [2022-12-19 11:19:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][440/1519] eta 0:18:03 lr 0.000032 time 0.9224 (1.0044) model_time 0.9223 (1.0032) loss 1.3808 (0.9813) grad_norm 8.4119 (9.1959/2.8699) mem 68106MB [2022-12-19 11:19:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][450/1519] eta 0:17:53 lr 0.000032 time 0.9240 (1.0043) model_time 0.9238 (1.0031) loss 0.7692 (0.9788) grad_norm 16.6557 (9.2549/2.8880) mem 68106MB [2022-12-19 11:20:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][460/1519] eta 0:17:43 lr 0.000032 time 0.9320 (1.0042) model_time 0.9317 (1.0030) loss 1.3428 (0.9810) grad_norm 7.2388 (9.2231/2.8654) mem 68106MB [2022-12-19 11:20:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][470/1519] eta 0:17:33 lr 0.000032 time 0.9245 (1.0041) model_time 0.9243 (1.0029) loss 1.2068 (0.9827) grad_norm 14.4346 (9.2904/2.9055) mem 68106MB [2022-12-19 11:20:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][480/1519] eta 0:17:23 lr 0.000032 time 0.9252 (1.0042) model_time 0.9251 (1.0030) loss 1.0030 (0.9809) grad_norm 8.2982 (9.2666/2.8942) mem 68106MB [2022-12-19 11:20:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][490/1519] eta 0:17:13 lr 0.000032 time 0.9310 (1.0040) model_time 0.9308 (1.0028) loss 0.7819 (0.9806) grad_norm 10.8785 (9.2493/2.8817) mem 68106MB [2022-12-19 11:20:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][500/1519] eta 0:17:03 lr 0.000032 time 1.2044 (1.0045) model_time 1.2042 (1.0033) loss 0.9762 (0.9811) grad_norm 6.6627 (9.2535/2.8767) mem 68106MB [2022-12-19 11:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][510/1519] eta 0:16:53 lr 0.000032 time 0.9205 (1.0044) model_time 0.9203 (1.0032) loss 0.8841 (0.9819) grad_norm 11.0148 (9.2318/2.8624) mem 68106MB [2022-12-19 11:21:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][520/1519] eta 0:16:43 lr 0.000032 time 0.9261 (1.0044) model_time 0.9259 (1.0032) loss 0.9926 (0.9806) grad_norm 7.7842 (9.2272/2.8561) mem 68106MB [2022-12-19 11:21:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][530/1519] eta 0:16:33 lr 0.000032 time 0.9225 (1.0045) model_time 0.9223 (1.0034) loss 0.7915 (0.9797) grad_norm 7.9214 (9.2030/2.8365) mem 68106MB [2022-12-19 11:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][540/1519] eta 0:16:23 lr 0.000032 time 0.9383 (1.0045) model_time 0.9381 (1.0034) loss 1.0514 (0.9802) grad_norm 9.7212 (9.2056/2.8164) mem 68106MB [2022-12-19 11:21:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][550/1519] eta 0:16:13 lr 0.000032 time 0.9269 (1.0044) model_time 0.9267 (1.0033) loss 1.8240 (0.9815) grad_norm 7.2784 (9.1967/2.7988) mem 68106MB [2022-12-19 11:21:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][560/1519] eta 0:16:03 lr 0.000032 time 0.9788 (1.0045) model_time 0.9786 (1.0034) loss 1.0621 (0.9816) grad_norm 8.5610 (9.2171/2.7881) mem 68106MB [2022-12-19 11:21:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][570/1519] eta 0:15:53 lr 0.000032 time 0.9327 (1.0045) model_time 0.9303 (1.0034) loss 1.2371 (0.9806) grad_norm 6.9861 (9.2039/2.7725) mem 68106MB [2022-12-19 11:22:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][580/1519] eta 0:15:43 lr 0.000032 time 0.9273 (1.0044) model_time 0.9272 (1.0034) loss 1.2178 (0.9802) grad_norm 6.5686 (9.1867/2.7588) mem 68106MB [2022-12-19 11:22:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][590/1519] eta 0:15:32 lr 0.000032 time 0.9252 (1.0043) model_time 0.9251 (1.0033) loss 1.4753 (0.9812) grad_norm 11.5959 (9.1846/2.7526) mem 68106MB [2022-12-19 11:22:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][600/1519] eta 0:15:22 lr 0.000032 time 0.9275 (1.0043) model_time 0.9272 (1.0032) loss 0.9183 (0.9809) grad_norm 12.9070 (9.1771/2.7442) mem 68106MB [2022-12-19 11:22:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][610/1519] eta 0:15:13 lr 0.000032 time 1.2000 (1.0046) model_time 1.1998 (1.0036) loss 0.8065 (0.9815) grad_norm 12.0197 (9.2319/2.7748) mem 68106MB [2022-12-19 11:22:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][620/1519] eta 0:15:03 lr 0.000032 time 0.9167 (1.0046) model_time 0.9165 (1.0036) loss 0.9699 (0.9836) grad_norm 14.6379 (9.2438/2.7885) mem 68106MB [2022-12-19 11:22:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][630/1519] eta 0:14:53 lr 0.000032 time 0.9297 (1.0045) model_time 0.9295 (1.0035) loss 0.9715 (0.9821) grad_norm 7.1967 (9.2388/2.7879) mem 68106MB [2022-12-19 11:23:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][640/1519] eta 0:14:42 lr 0.000032 time 0.9256 (1.0045) model_time 0.9254 (1.0035) loss 0.8506 (0.9813) grad_norm 6.8558 (9.2229/2.7958) mem 68106MB [2022-12-19 11:23:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][650/1519] eta 0:14:32 lr 0.000032 time 0.9270 (1.0044) model_time 0.9267 (1.0034) loss 0.7668 (0.9794) grad_norm 8.8640 (9.2380/2.8049) mem 68106MB [2022-12-19 11:23:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][660/1519] eta 0:14:22 lr 0.000032 time 0.9254 (1.0044) model_time 0.9253 (1.0034) loss 1.0292 (0.9808) grad_norm 11.0427 (9.2200/2.7828) mem 68106MB [2022-12-19 11:23:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][670/1519] eta 0:14:12 lr 0.000032 time 0.9280 (1.0043) model_time 0.9278 (1.0033) loss 0.9107 (0.9817) grad_norm 8.1034 (9.2141/2.7872) mem 68106MB [2022-12-19 11:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][680/1519] eta 0:14:02 lr 0.000032 time 0.9385 (1.0043) model_time 0.9384 (1.0033) loss 1.2166 (0.9831) grad_norm 11.4755 (9.2544/2.7993) mem 68106MB [2022-12-19 11:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][690/1519] eta 0:13:52 lr 0.000032 time 0.9321 (1.0042) model_time 0.9319 (1.0033) loss 0.9186 (0.9833) grad_norm 11.0047 (9.2737/2.8092) mem 68106MB [2022-12-19 11:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][700/1519] eta 0:13:42 lr 0.000032 time 0.9282 (1.0042) model_time 0.9281 (1.0032) loss 0.7897 (0.9827) grad_norm 8.0679 (9.2440/2.7828) mem 68106MB [2022-12-19 11:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][710/1519] eta 0:13:32 lr 0.000032 time 0.9201 (1.0041) model_time 0.9200 (1.0032) loss 0.9296 (0.9814) grad_norm 7.6679 (9.2565/2.7818) mem 68106MB [2022-12-19 11:24:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][720/1519] eta 0:13:22 lr 0.000032 time 0.9332 (1.0041) model_time 0.9330 (1.0032) loss 0.8466 (0.9806) grad_norm 12.8152 (9.2875/2.7798) mem 68106MB [2022-12-19 11:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][730/1519] eta 0:13:12 lr 0.000032 time 0.9319 (1.0041) model_time 0.9318 (1.0032) loss 0.9577 (0.9806) grad_norm 9.4021 (9.3118/2.7770) mem 68106MB [2022-12-19 11:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][740/1519] eta 0:13:02 lr 0.000032 time 0.9336 (1.0041) model_time 0.9332 (1.0032) loss 0.7229 (0.9798) grad_norm 9.5355 (9.2751/2.7654) mem 68106MB [2022-12-19 11:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][750/1519] eta 0:12:52 lr 0.000032 time 0.9272 (1.0041) model_time 0.9271 (1.0032) loss 1.0560 (0.9792) grad_norm 8.0151 (9.2377/2.7529) mem 68106MB [2022-12-19 11:25:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][760/1519] eta 0:12:42 lr 0.000032 time 0.9220 (1.0040) model_time 0.9218 (1.0031) loss 1.3333 (0.9795) grad_norm 9.8223 (9.2789/2.7664) mem 68106MB [2022-12-19 11:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][770/1519] eta 0:12:31 lr 0.000032 time 0.9270 (1.0039) model_time 0.9269 (1.0031) loss 0.8096 (0.9784) grad_norm 10.8968 (9.3193/2.7505) mem 68106MB [2022-12-19 11:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][780/1519] eta 0:12:21 lr 0.000032 time 0.9240 (1.0039) model_time 0.9239 (1.0031) loss 1.1006 (0.9787) grad_norm 7.2441 (9.3293/2.7160) mem 68106MB [2022-12-19 11:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][790/1519] eta 0:12:11 lr 0.000032 time 0.9113 (1.0040) model_time 0.9112 (1.0031) loss 0.8405 (0.9786) grad_norm 9.5279 (9.3011/2.6903) mem 68106MB [2022-12-19 11:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][800/1519] eta 0:12:01 lr 0.000032 time 0.9310 (1.0039) model_time 0.9309 (1.0030) loss 1.1432 (0.9778) grad_norm 7.5051 (9.2627/2.6776) mem 68106MB [2022-12-19 11:25:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][810/1519] eta 0:11:51 lr 0.000032 time 0.9287 (1.0038) model_time 0.9286 (1.0030) loss 1.0576 (0.9782) grad_norm 8.6980 (9.2537/2.6584) mem 68106MB [2022-12-19 11:26:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][820/1519] eta 0:11:41 lr 0.000032 time 0.9292 (1.0039) model_time 0.9291 (1.0030) loss 0.8942 (0.9781) grad_norm 10.8621 (9.2271/2.6456) mem 68106MB [2022-12-19 11:26:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][830/1519] eta 0:11:31 lr 0.000032 time 0.9277 (1.0039) model_time 0.9276 (1.0030) loss 0.8734 (0.9773) grad_norm 5.4848 (9.1990/2.5943) mem 68106MB [2022-12-19 11:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][840/1519] eta 0:11:21 lr 0.000032 time 0.9267 (1.0042) model_time 0.9265 (1.0033) loss 0.9624 (0.9768) grad_norm 6.5689 (9.2310/2.5954) mem 68106MB [2022-12-19 11:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][850/1519] eta 0:11:11 lr 0.000032 time 0.9350 (1.0041) model_time 0.9348 (1.0032) loss 1.1051 (0.9774) grad_norm 7.2389 (9.2773/2.6776) mem 68106MB [2022-12-19 11:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][860/1519] eta 0:11:01 lr 0.000032 time 0.9298 (1.0040) model_time 0.9296 (1.0032) loss 1.0648 (0.9784) grad_norm 7.3730 (9.2799/2.6664) mem 68106MB [2022-12-19 11:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][870/1519] eta 0:10:51 lr 0.000032 time 0.9303 (1.0040) model_time 0.9301 (1.0032) loss 0.8204 (0.9788) grad_norm 10.0199 (9.2728/2.6732) mem 68106MB [2022-12-19 11:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][880/1519] eta 0:10:41 lr 0.000032 time 0.9302 (1.0041) model_time 0.9300 (1.0032) loss 0.8677 (0.9798) grad_norm 10.3241 (9.2717/2.6688) mem 68106MB [2022-12-19 11:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][890/1519] eta 0:10:31 lr 0.000032 time 0.9307 (1.0040) model_time 0.9306 (1.0032) loss 1.0544 (0.9812) grad_norm 7.4931 (9.2391/2.6525) mem 68106MB [2022-12-19 11:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][900/1519] eta 0:10:21 lr 0.000032 time 0.9287 (1.0039) model_time 0.9286 (1.0031) loss 1.1633 (0.9810) grad_norm 6.8015 (9.1955/2.6407) mem 68106MB [2022-12-19 11:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][910/1519] eta 0:10:11 lr 0.000032 time 0.9324 (1.0039) model_time 0.9322 (1.0031) loss 1.0040 (0.9811) grad_norm 9.7988 (9.1927/2.6422) mem 68106MB [2022-12-19 11:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][920/1519] eta 0:10:01 lr 0.000032 time 0.9327 (1.0039) model_time 0.9326 (1.0030) loss 1.3346 (0.9821) grad_norm 8.5925 (9.2152/2.6336) mem 68106MB [2022-12-19 11:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][930/1519] eta 0:09:51 lr 0.000032 time 0.9322 (1.0040) model_time 0.9321 (1.0032) loss 1.2663 (0.9826) grad_norm 12.6158 (9.2213/2.6330) mem 68106MB [2022-12-19 11:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][940/1519] eta 0:09:41 lr 0.000032 time 0.9250 (1.0039) model_time 0.9249 (1.0031) loss 0.7895 (0.9822) grad_norm 18.4434 (9.2525/2.6869) mem 68106MB [2022-12-19 11:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][950/1519] eta 0:09:31 lr 0.000032 time 0.9282 (1.0038) model_time 0.9281 (1.0030) loss 0.7587 (0.9812) grad_norm 13.3300 (9.2929/2.7146) mem 68106MB [2022-12-19 11:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][960/1519] eta 0:09:21 lr 0.000032 time 0.9310 (1.0038) model_time 0.9308 (1.0030) loss 0.6945 (0.9814) grad_norm 8.8806 (9.2778/2.6760) mem 68106MB [2022-12-19 11:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][970/1519] eta 0:09:11 lr 0.000032 time 0.9354 (1.0038) model_time 0.9353 (1.0030) loss 0.8360 (0.9820) grad_norm 11.2792 (9.2717/2.6741) mem 68106MB [2022-12-19 11:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][980/1519] eta 0:09:01 lr 0.000032 time 0.9235 (1.0038) model_time 0.9233 (1.0030) loss 1.2491 (0.9818) grad_norm 8.3436 (9.2809/2.6747) mem 68106MB [2022-12-19 11:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][990/1519] eta 0:08:50 lr 0.000032 time 0.9443 (1.0037) model_time 0.9441 (1.0029) loss 1.2858 (0.9817) grad_norm 6.2888 (9.3075/2.6853) mem 68106MB [2022-12-19 11:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1000/1519] eta 0:08:40 lr 0.000032 time 0.9342 (1.0037) model_time 0.9340 (1.0029) loss 0.9744 (0.9819) grad_norm 14.7074 (9.3923/2.7281) mem 68106MB [2022-12-19 11:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1010/1519] eta 0:08:30 lr 0.000032 time 0.9284 (1.0037) model_time 0.9283 (1.0029) loss 1.0903 (0.9822) grad_norm 9.4729 (9.3828/2.6611) mem 68106MB [2022-12-19 11:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1020/1519] eta 0:08:20 lr 0.000032 time 0.9152 (1.0038) model_time 0.9150 (1.0030) loss 0.9034 (0.9820) grad_norm 8.2554 (9.4034/2.6563) mem 68106MB [2022-12-19 11:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1030/1519] eta 0:08:10 lr 0.000032 time 0.9116 (1.0038) model_time 0.9114 (1.0030) loss 1.1036 (0.9833) grad_norm 5.9229 (9.3602/2.4453) mem 68106MB [2022-12-19 11:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1040/1519] eta 0:08:00 lr 0.000032 time 0.9188 (1.0037) model_time 0.9186 (1.0029) loss 0.9891 (0.9849) grad_norm 6.1647 (9.3539/2.4398) mem 68106MB [2022-12-19 11:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1050/1519] eta 0:07:50 lr 0.000032 time 0.9300 (1.0037) model_time 0.9298 (1.0030) loss 0.7833 (0.9843) grad_norm 8.8526 (9.2874/2.4057) mem 68106MB [2022-12-19 11:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1060/1519] eta 0:07:40 lr 0.000032 time 0.9277 (1.0037) model_time 0.9276 (1.0029) loss 1.1113 (0.9839) grad_norm 7.6136 (9.2972/2.4114) mem 68106MB [2022-12-19 11:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1070/1519] eta 0:07:30 lr 0.000032 time 0.9306 (1.0036) model_time 0.9304 (1.0028) loss 0.9681 (0.9842) grad_norm 9.0372 (9.2180/2.3662) mem 68106MB [2022-12-19 11:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1080/1519] eta 0:07:20 lr 0.000032 time 0.9260 (1.0036) model_time 0.9258 (1.0028) loss 1.0509 (0.9843) grad_norm 7.8851 (9.2834/2.4054) mem 68106MB [2022-12-19 11:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1090/1519] eta 0:07:10 lr 0.000031 time 0.9439 (1.0035) model_time 0.9437 (1.0028) loss 1.1823 (0.9846) grad_norm 6.6551 (9.2856/2.3955) mem 68106MB [2022-12-19 11:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1100/1519] eta 0:07:00 lr 0.000031 time 0.9340 (1.0036) model_time 0.9338 (1.0028) loss 0.9436 (0.9858) grad_norm 6.8821 (9.2759/2.3813) mem 68106MB [2022-12-19 11:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1110/1519] eta 0:06:50 lr 0.000031 time 0.9301 (1.0036) model_time 0.9299 (1.0028) loss 0.8077 (0.9854) grad_norm 8.4634 (9.2918/2.3781) mem 68106MB [2022-12-19 11:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1120/1519] eta 0:06:40 lr 0.000031 time 0.9295 (1.0035) model_time 0.9290 (1.0028) loss 0.9921 (0.9855) grad_norm 7.9158 (9.2764/2.3730) mem 68106MB [2022-12-19 11:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1130/1519] eta 0:06:30 lr 0.000031 time 0.9261 (1.0036) model_time 0.9259 (1.0028) loss 0.9069 (0.9869) grad_norm 8.2356 (9.2662/2.3817) mem 68106MB [2022-12-19 11:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1140/1519] eta 0:06:20 lr 0.000031 time 0.9270 (1.0035) model_time 0.9268 (1.0028) loss 1.0785 (0.9869) grad_norm 9.2298 (9.2796/2.3862) mem 68106MB [2022-12-19 11:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1150/1519] eta 0:06:10 lr 0.000031 time 0.9021 (1.0036) model_time 0.9019 (1.0028) loss 0.7948 (0.9865) grad_norm 6.9829 (9.2819/2.4134) mem 68106MB [2022-12-19 11:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1160/1519] eta 0:06:00 lr 0.000031 time 0.9301 (1.0035) model_time 0.9299 (1.0028) loss 1.1295 (0.9867) grad_norm 7.3119 (9.2911/2.4765) mem 68106MB [2022-12-19 11:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1170/1519] eta 0:05:50 lr 0.000031 time 0.9258 (1.0035) model_time 0.9255 (1.0027) loss 1.3784 (0.9869) grad_norm 10.3319 (9.2932/2.4824) mem 68106MB [2022-12-19 11:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1180/1519] eta 0:05:40 lr 0.000031 time 0.9160 (1.0034) model_time 0.9158 (1.0027) loss 1.1547 (0.9864) grad_norm 9.6918 (9.2998/2.4786) mem 68106MB [2022-12-19 11:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1190/1519] eta 0:05:30 lr 0.000031 time 0.9254 (1.0035) model_time 0.9252 (1.0027) loss 0.9332 (0.9864) grad_norm 6.4880 (9.2766/2.4725) mem 68106MB [2022-12-19 11:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1200/1519] eta 0:05:20 lr 0.000031 time 0.9294 (1.0034) model_time 0.9292 (1.0027) loss 0.9290 (0.9855) grad_norm 8.0336 (9.3102/2.5442) mem 68106MB [2022-12-19 11:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1210/1519] eta 0:05:10 lr 0.000031 time 0.9297 (1.0034) model_time 0.9295 (1.0027) loss 0.8040 (0.9851) grad_norm 17.8528 (9.2961/2.5452) mem 68106MB [2022-12-19 11:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1220/1519] eta 0:05:00 lr 0.000031 time 0.9287 (1.0034) model_time 0.9285 (1.0027) loss 0.9412 (0.9844) grad_norm 9.8475 (9.2670/2.5338) mem 68106MB [2022-12-19 11:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1230/1519] eta 0:04:49 lr 0.000031 time 0.9262 (1.0034) model_time 0.9260 (1.0027) loss 0.7477 (0.9844) grad_norm 8.4958 (9.3476/2.6518) mem 68106MB [2022-12-19 11:33:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1240/1519] eta 0:04:39 lr 0.000031 time 1.0261 (1.0034) model_time 1.0259 (1.0027) loss 0.8537 (0.9836) grad_norm 8.6276 (9.3567/2.6450) mem 68106MB [2022-12-19 11:33:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1250/1519] eta 0:04:29 lr 0.000031 time 0.9188 (1.0034) model_time 0.9187 (1.0027) loss 1.1638 (0.9846) grad_norm 8.4709 (9.3303/2.6379) mem 68106MB [2022-12-19 11:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1260/1519] eta 0:04:19 lr 0.000031 time 0.9336 (1.0034) model_time 0.9334 (1.0027) loss 0.8624 (0.9843) grad_norm 11.6196 (9.3316/2.6366) mem 68106MB [2022-12-19 11:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1270/1519] eta 0:04:09 lr 0.000031 time 0.9281 (1.0034) model_time 0.9279 (1.0026) loss 0.7639 (0.9843) grad_norm 10.4612 (9.3437/2.6251) mem 68106MB [2022-12-19 11:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1280/1519] eta 0:03:59 lr 0.000031 time 0.9276 (1.0033) model_time 0.9274 (1.0026) loss 0.9599 (0.9841) grad_norm 9.1579 (9.2873/2.6006) mem 68106MB [2022-12-19 11:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1290/1519] eta 0:03:49 lr 0.000031 time 0.9315 (1.0033) model_time 0.9313 (1.0026) loss 0.9382 (0.9847) grad_norm 13.5348 (9.3025/2.5951) mem 68106MB [2022-12-19 11:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1300/1519] eta 0:03:39 lr 0.000031 time 0.9342 (1.0033) model_time 0.9340 (1.0025) loss 0.7732 (0.9843) grad_norm 6.7828 (9.3220/2.6072) mem 68106MB [2022-12-19 11:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1310/1519] eta 0:03:29 lr 0.000031 time 0.9349 (1.0032) model_time 0.9347 (1.0025) loss 0.8137 (0.9842) grad_norm 12.2060 (9.3397/2.6116) mem 68106MB [2022-12-19 11:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1320/1519] eta 0:03:19 lr 0.000031 time 0.9128 (1.0033) model_time 0.9126 (1.0026) loss 0.8893 (0.9838) grad_norm 13.8914 (9.3466/2.6215) mem 68106MB [2022-12-19 11:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1330/1519] eta 0:03:09 lr 0.000031 time 0.9318 (1.0032) model_time 0.9315 (1.0025) loss 1.1286 (0.9841) grad_norm 11.0633 (9.3324/2.6350) mem 68106MB [2022-12-19 11:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1340/1519] eta 0:02:59 lr 0.000031 time 0.9083 (1.0033) model_time 0.9081 (1.0026) loss 0.7265 (0.9835) grad_norm 6.7558 (9.3357/2.6414) mem 68106MB [2022-12-19 11:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1350/1519] eta 0:02:49 lr 0.000031 time 0.9316 (1.0033) model_time 0.9315 (1.0026) loss 1.3365 (0.9836) grad_norm 7.7659 (9.3149/2.6439) mem 68106MB [2022-12-19 11:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1360/1519] eta 0:02:39 lr 0.000031 time 1.1783 (1.0034) model_time 1.1782 (1.0027) loss 0.8506 (0.9833) grad_norm 9.8120 (9.3033/2.6528) mem 68106MB [2022-12-19 11:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1370/1519] eta 0:02:29 lr 0.000031 time 0.9280 (1.0034) model_time 0.9278 (1.0027) loss 0.9289 (0.9832) grad_norm 9.1791 (9.2658/2.6510) mem 68106MB [2022-12-19 11:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1380/1519] eta 0:02:19 lr 0.000031 time 0.9221 (1.0033) model_time 0.9220 (1.0027) loss 1.1188 (0.9830) grad_norm 10.3602 (9.3015/2.7635) mem 68106MB [2022-12-19 11:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1390/1519] eta 0:02:09 lr 0.000031 time 0.9284 (1.0033) model_time 0.9283 (1.0026) loss 1.0013 (0.9821) grad_norm 8.7374 (9.2855/2.7707) mem 68106MB [2022-12-19 11:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1400/1519] eta 0:01:59 lr 0.000031 time 0.9209 (1.0033) model_time 0.9208 (1.0026) loss 0.8049 (0.9810) grad_norm 11.6346 (9.2917/2.7747) mem 68106MB [2022-12-19 11:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1410/1519] eta 0:01:49 lr 0.000031 time 0.9047 (1.0033) model_time 0.9046 (1.0026) loss 1.0452 (0.9813) grad_norm 7.7060 (9.3051/2.7900) mem 68106MB [2022-12-19 11:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1420/1519] eta 0:01:39 lr 0.000031 time 0.9840 (1.0034) model_time 0.9839 (1.0027) loss 0.8818 (0.9815) grad_norm 6.9127 (9.3130/2.7847) mem 68106MB [2022-12-19 11:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1430/1519] eta 0:01:29 lr 0.000031 time 0.9211 (1.0034) model_time 0.9209 (1.0027) loss 0.7638 (0.9816) grad_norm 7.7340 (9.3093/2.7784) mem 68106MB [2022-12-19 11:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1440/1519] eta 0:01:19 lr 0.000031 time 0.9302 (1.0034) model_time 0.9301 (1.0028) loss 0.9295 (0.9814) grad_norm 8.2115 (9.3325/2.8455) mem 68106MB [2022-12-19 11:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1450/1519] eta 0:01:09 lr 0.000031 time 0.9270 (1.0034) model_time 0.9269 (1.0028) loss 1.0858 (0.9819) grad_norm 5.1763 (9.2896/2.7775) mem 68106MB [2022-12-19 11:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1460/1519] eta 0:00:59 lr 0.000031 time 0.8907 (1.0035) model_time 0.8906 (1.0028) loss 0.8983 (0.9815) grad_norm 7.2694 (9.3038/2.7996) mem 68106MB [2022-12-19 11:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1470/1519] eta 0:00:49 lr 0.000031 time 0.9224 (1.0034) model_time 0.9222 (1.0028) loss 1.0707 (0.9815) grad_norm 9.9634 (9.3258/2.8034) mem 68106MB [2022-12-19 11:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1480/1519] eta 0:00:39 lr 0.000031 time 0.9236 (1.0034) model_time 0.9234 (1.0027) loss 1.2637 (0.9822) grad_norm 6.7667 (9.3129/2.8255) mem 68106MB [2022-12-19 11:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1490/1519] eta 0:00:29 lr 0.000031 time 0.9287 (1.0034) model_time 0.9285 (1.0027) loss 1.0957 (0.9821) grad_norm 11.9255 (9.3391/2.8188) mem 68106MB [2022-12-19 11:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1500/1519] eta 0:00:19 lr 0.000031 time 0.9643 (1.0034) model_time 0.9642 (1.0027) loss 0.8389 (0.9820) grad_norm 7.2565 (9.3366/2.8148) mem 68106MB [2022-12-19 11:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [16/100][1510/1519] eta 0:00:09 lr 0.000031 time 0.9252 (1.0034) model_time 0.9251 (1.0027) loss 0.8277 (0.9828) grad_norm 9.1225 (9.3165/2.8197) mem 68106MB [2022-12-19 11:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 16 training takes 0:25:24 [2022-12-19 11:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_16.pth saving...... [2022-12-19 11:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_16.pth saved !!! [2022-12-19 11:38:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.662 (0.662) Loss 0.5878 (0.5878) Acc@1 88.889 (88.889) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 11:38:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.331) Loss 0.6154 (0.5784) Acc@1 88.542 (89.015) Acc@5 97.222 (98.169) Mem 68106MB [2022-12-19 11:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.305 (0.316) Loss 0.5430 (0.5789) Acc@1 89.931 (89.170) Acc@5 98.958 (98.082) Mem 68106MB [2022-12-19 11:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.310) Loss 0.6587 (0.5820) Acc@1 88.194 (88.956) Acc@5 97.569 (97.950) Mem 68106MB [2022-12-19 11:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.307) Loss 0.5687 (0.5733) Acc@1 88.542 (89.118) Acc@5 97.569 (98.001) Mem 68106MB [2022-12-19 11:38:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.306) Loss 0.6051 (0.5688) Acc@1 85.417 (89.161) Acc@5 97.222 (98.019) Mem 68106MB [2022-12-19 11:38:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.305) Loss 0.5738 (0.5689) Acc@1 87.847 (89.065) Acc@5 97.569 (98.019) Mem 68106MB [2022-12-19 11:38:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.6260 (0.5702) Acc@1 89.236 (89.050) Acc@5 97.569 (97.985) Mem 68106MB [2022-12-19 11:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.5199 (0.5694) Acc@1 90.625 (89.086) Acc@5 97.917 (97.985) Mem 68106MB [2022-12-19 11:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:16] * Acc@1 89.072 Acc@5 97.994 [2022-12-19 11:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 89.1% [2022-12-19 11:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 11:39:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 11:39:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 89.07% [2022-12-19 11:39:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][0/1519] eta 0:35:46 lr 0.000031 time 1.4132 (1.4132) model_time 0.9884 (0.9884) loss 0.9964 (0.9964) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 11:39:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][10/1519] eta 0:26:13 lr 0.000031 time 0.9258 (1.0430) model_time 0.9257 (1.0041) loss 0.7230 (0.8888) grad_norm 6.3337 (7.8233/1.2419) mem 68106MB [2022-12-19 11:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][20/1519] eta 0:25:31 lr 0.000031 time 0.9233 (1.0218) model_time 0.9232 (1.0012) loss 0.7941 (0.9380) grad_norm 7.8062 (8.4575/1.5005) mem 68106MB [2022-12-19 11:39:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][30/1519] eta 0:25:10 lr 0.000031 time 0.9256 (1.0143) model_time 0.9255 (1.0003) loss 1.3266 (0.9529) grad_norm 7.5035 (8.6512/1.7573) mem 68106MB [2022-12-19 11:39:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][40/1519] eta 0:24:56 lr 0.000031 time 0.9273 (1.0115) model_time 0.9271 (1.0008) loss 1.3016 (0.9647) grad_norm 12.8144 (8.9606/1.7886) mem 68106MB [2022-12-19 11:39:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][50/1519] eta 0:24:41 lr 0.000031 time 0.9282 (1.0085) model_time 0.9281 (0.9999) loss 1.2112 (0.9751) grad_norm 11.4799 (9.2134/1.9475) mem 68106MB [2022-12-19 11:40:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][60/1519] eta 0:24:28 lr 0.000031 time 0.9217 (1.0068) model_time 0.9216 (0.9995) loss 1.0957 (0.9679) grad_norm 7.4749 (8.9215/1.9218) mem 68106MB [2022-12-19 11:40:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][70/1519] eta 0:24:20 lr 0.000031 time 0.9928 (1.0076) model_time 0.9927 (1.0013) loss 1.2801 (0.9805) grad_norm 8.6601 (9.0289/1.8147) mem 68106MB [2022-12-19 11:40:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][80/1519] eta 0:24:09 lr 0.000031 time 0.9422 (1.0076) model_time 0.9421 (1.0021) loss 1.4323 (0.9809) grad_norm 8.0458 (9.0330/1.7987) mem 68106MB [2022-12-19 11:40:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][90/1519] eta 0:23:59 lr 0.000031 time 0.9266 (1.0076) model_time 0.9264 (1.0026) loss 0.8882 (0.9788) grad_norm 7.4381 (9.0939/1.8757) mem 68106MB [2022-12-19 11:40:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][100/1519] eta 0:23:48 lr 0.000031 time 0.9258 (1.0070) model_time 0.9256 (1.0025) loss 0.7448 (0.9843) grad_norm 6.8486 (9.1025/1.8992) mem 68106MB [2022-12-19 11:40:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][110/1519] eta 0:23:37 lr 0.000031 time 0.9235 (1.0063) model_time 0.9234 (1.0022) loss 1.1654 (0.9950) grad_norm 11.1484 (9.3762/2.2875) mem 68106MB [2022-12-19 11:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][120/1519] eta 0:23:28 lr 0.000031 time 0.9199 (1.0069) model_time 0.9197 (1.0031) loss 1.1182 (0.9864) grad_norm 8.0112 (9.2814/2.2419) mem 68106MB [2022-12-19 11:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][130/1519] eta 0:23:19 lr 0.000031 time 0.9073 (1.0075) model_time 0.9072 (1.0039) loss 0.9730 (0.9819) grad_norm 8.7021 (9.2173/2.1824) mem 68106MB [2022-12-19 11:41:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][140/1519] eta 0:23:07 lr 0.000031 time 0.9219 (1.0061) model_time 0.9214 (1.0027) loss 0.9053 (0.9829) grad_norm inf (9.2549/2.2377) mem 68106MB [2022-12-19 11:41:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][150/1519] eta 0:22:56 lr 0.000031 time 0.9295 (1.0055) model_time 0.9293 (1.0024) loss 0.8949 (0.9759) grad_norm 7.2618 (9.3519/2.5067) mem 68106MB [2022-12-19 11:41:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][160/1519] eta 0:22:45 lr 0.000031 time 0.9221 (1.0051) model_time 0.9220 (1.0021) loss 0.9252 (0.9711) grad_norm 6.8741 (9.2471/2.4798) mem 68106MB [2022-12-19 11:41:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][170/1519] eta 0:22:35 lr 0.000031 time 0.9279 (1.0048) model_time 0.9277 (1.0020) loss 0.7337 (0.9673) grad_norm 9.8710 (9.1311/2.4952) mem 68106MB [2022-12-19 11:42:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][180/1519] eta 0:22:26 lr 0.000031 time 0.9327 (1.0054) model_time 0.9326 (1.0028) loss 1.3344 (0.9658) grad_norm 7.7306 (9.1286/2.4630) mem 68106MB [2022-12-19 11:42:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][190/1519] eta 0:22:16 lr 0.000031 time 0.9264 (1.0055) model_time 0.9262 (1.0029) loss 0.7433 (0.9620) grad_norm 9.3441 (9.0422/2.4375) mem 68106MB [2022-12-19 11:42:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][200/1519] eta 0:22:05 lr 0.000031 time 0.9278 (1.0053) model_time 0.9277 (1.0028) loss 0.9707 (0.9607) grad_norm 7.0487 (8.9494/2.4130) mem 68106MB [2022-12-19 11:42:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][210/1519] eta 0:21:56 lr 0.000031 time 0.9339 (1.0055) model_time 0.9338 (1.0031) loss 0.8542 (0.9658) grad_norm 10.0661 (8.9659/2.3897) mem 68106MB [2022-12-19 11:42:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][220/1519] eta 0:21:45 lr 0.000031 time 0.9274 (1.0053) model_time 0.9272 (1.0031) loss 0.8816 (0.9631) grad_norm 12.3598 (8.9367/2.3849) mem 68106MB [2022-12-19 11:42:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][230/1519] eta 0:21:35 lr 0.000031 time 0.9281 (1.0052) model_time 0.9280 (1.0030) loss 0.7630 (0.9639) grad_norm 5.9767 (8.8981/2.3580) mem 68106MB [2022-12-19 11:43:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][240/1519] eta 0:21:25 lr 0.000031 time 0.9263 (1.0049) model_time 0.9262 (1.0028) loss 1.0350 (0.9638) grad_norm 8.2907 (8.8537/2.3457) mem 68106MB [2022-12-19 11:43:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][250/1519] eta 0:21:15 lr 0.000031 time 0.9362 (1.0048) model_time 0.9359 (1.0027) loss 1.1462 (0.9638) grad_norm 7.9436 (8.8469/2.3050) mem 68106MB [2022-12-19 11:43:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][260/1519] eta 0:21:04 lr 0.000031 time 0.9361 (1.0047) model_time 0.9359 (1.0028) loss 1.0899 (0.9643) grad_norm 11.7813 (8.8089/2.3066) mem 68106MB [2022-12-19 11:43:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][270/1519] eta 0:20:54 lr 0.000031 time 0.9359 (1.0046) model_time 0.9357 (1.0026) loss 0.8560 (0.9629) grad_norm 8.9656 (8.7690/2.2901) mem 68106MB [2022-12-19 11:43:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][280/1519] eta 0:20:44 lr 0.000031 time 0.9039 (1.0046) model_time 0.9038 (1.0027) loss 1.0564 (0.9623) grad_norm 11.2256 (8.8001/2.2808) mem 68106MB [2022-12-19 11:43:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][290/1519] eta 0:20:34 lr 0.000031 time 0.9267 (1.0046) model_time 0.9265 (1.0028) loss 1.1247 (0.9631) grad_norm 12.8623 (8.8063/2.2876) mem 68106MB [2022-12-19 11:44:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][300/1519] eta 0:20:25 lr 0.000031 time 0.9364 (1.0052) model_time 0.9362 (1.0034) loss 0.8501 (0.9640) grad_norm 8.3577 (8.8385/2.3700) mem 68106MB [2022-12-19 11:44:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][310/1519] eta 0:20:15 lr 0.000031 time 0.9232 (1.0055) model_time 0.9231 (1.0037) loss 0.8292 (0.9641) grad_norm 12.5292 (8.8728/2.3832) mem 68106MB [2022-12-19 11:44:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][320/1519] eta 0:20:05 lr 0.000031 time 0.9211 (1.0053) model_time 0.9209 (1.0037) loss 1.0132 (0.9667) grad_norm 6.3501 (8.8510/2.3601) mem 68106MB [2022-12-19 11:44:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][330/1519] eta 0:19:55 lr 0.000031 time 0.9277 (1.0052) model_time 0.9276 (1.0035) loss 0.8380 (0.9657) grad_norm 7.6804 (8.8137/2.3396) mem 68106MB [2022-12-19 11:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][340/1519] eta 0:19:45 lr 0.000031 time 0.9290 (1.0051) model_time 0.9289 (1.0035) loss 0.7828 (0.9689) grad_norm 11.4229 (8.8441/2.3485) mem 68106MB [2022-12-19 11:44:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][350/1519] eta 0:19:34 lr 0.000031 time 0.9190 (1.0050) model_time 0.9188 (1.0034) loss 1.2001 (0.9702) grad_norm 8.5523 (8.7977/2.3420) mem 68106MB [2022-12-19 11:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][360/1519] eta 0:19:24 lr 0.000031 time 0.9315 (1.0048) model_time 0.9314 (1.0033) loss 0.8328 (0.9682) grad_norm 11.8454 (8.8078/2.3372) mem 68106MB [2022-12-19 11:45:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][370/1519] eta 0:19:14 lr 0.000031 time 0.9228 (1.0046) model_time 0.9227 (1.0031) loss 0.8595 (0.9677) grad_norm 10.8509 (8.7769/2.3297) mem 68106MB [2022-12-19 11:45:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][380/1519] eta 0:19:04 lr 0.000031 time 0.9233 (1.0045) model_time 0.9232 (1.0031) loss 0.7737 (0.9688) grad_norm 9.1390 (8.7706/2.3241) mem 68106MB [2022-12-19 11:45:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][390/1519] eta 0:18:54 lr 0.000031 time 0.9341 (1.0050) model_time 0.9340 (1.0035) loss 0.9086 (0.9680) grad_norm 8.0388 (8.7472/2.3049) mem 68106MB [2022-12-19 11:45:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][400/1519] eta 0:18:44 lr 0.000031 time 0.9024 (1.0049) model_time 0.9023 (1.0035) loss 1.2664 (0.9702) grad_norm 7.9038 (8.7325/2.2871) mem 68106MB [2022-12-19 11:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][410/1519] eta 0:18:34 lr 0.000031 time 0.9372 (1.0049) model_time 0.9371 (1.0035) loss 0.8485 (0.9698) grad_norm 7.5009 (8.7183/2.2775) mem 68106MB [2022-12-19 11:46:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][420/1519] eta 0:18:24 lr 0.000031 time 0.9291 (1.0048) model_time 0.9290 (1.0034) loss 1.4044 (0.9720) grad_norm 7.5744 (8.7335/2.2684) mem 68106MB [2022-12-19 11:46:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][430/1519] eta 0:18:14 lr 0.000031 time 0.9324 (1.0046) model_time 0.9322 (1.0033) loss 0.7902 (0.9716) grad_norm 11.2958 (8.7741/2.2921) mem 68106MB [2022-12-19 11:46:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][440/1519] eta 0:18:03 lr 0.000031 time 0.9210 (1.0046) model_time 0.9209 (1.0033) loss 0.9078 (0.9705) grad_norm 8.9500 (8.8283/2.4487) mem 68106MB [2022-12-19 11:46:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][450/1519] eta 0:17:53 lr 0.000031 time 0.9335 (1.0045) model_time 0.9333 (1.0032) loss 1.0259 (0.9702) grad_norm 6.2241 (8.8006/2.4405) mem 68106MB [2022-12-19 11:46:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][460/1519] eta 0:17:43 lr 0.000031 time 0.9267 (1.0044) model_time 0.9265 (1.0031) loss 1.2478 (0.9708) grad_norm 8.0783 (8.7784/2.4307) mem 68106MB [2022-12-19 11:46:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][470/1519] eta 0:17:33 lr 0.000031 time 0.9276 (1.0043) model_time 0.9273 (1.0030) loss 0.9846 (0.9714) grad_norm 5.6775 (8.7714/2.4235) mem 68106MB [2022-12-19 11:47:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][480/1519] eta 0:17:23 lr 0.000031 time 0.9220 (1.0042) model_time 0.9219 (1.0029) loss 1.1660 (0.9711) grad_norm 13.8119 (8.8117/2.4359) mem 68106MB [2022-12-19 11:47:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][490/1519] eta 0:17:13 lr 0.000031 time 0.9193 (1.0046) model_time 0.9192 (1.0034) loss 0.8019 (0.9696) grad_norm 6.3838 (8.7874/2.4217) mem 68106MB [2022-12-19 11:47:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][500/1519] eta 0:17:03 lr 0.000031 time 0.9224 (1.0045) model_time 0.9222 (1.0033) loss 1.0631 (0.9693) grad_norm 10.3265 (8.7995/2.4440) mem 68106MB [2022-12-19 11:47:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][510/1519] eta 0:16:53 lr 0.000031 time 0.9360 (1.0045) model_time 0.9357 (1.0033) loss 1.1219 (0.9711) grad_norm 15.7795 (8.8230/2.4625) mem 68106MB [2022-12-19 11:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][520/1519] eta 0:16:43 lr 0.000031 time 0.9295 (1.0044) model_time 0.9294 (1.0033) loss 0.9047 (0.9705) grad_norm 6.0420 (8.8031/2.4489) mem 68106MB [2022-12-19 11:47:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][530/1519] eta 0:16:33 lr 0.000031 time 0.9245 (1.0044) model_time 0.9244 (1.0032) loss 1.0387 (0.9713) grad_norm 6.1691 (8.8140/2.4688) mem 68106MB [2022-12-19 11:48:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][540/1519] eta 0:16:23 lr 0.000031 time 0.9203 (1.0043) model_time 0.9202 (1.0032) loss 0.8168 (0.9704) grad_norm 9.2397 (8.8222/2.4589) mem 68106MB [2022-12-19 11:48:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][550/1519] eta 0:16:13 lr 0.000031 time 0.9346 (1.0043) model_time 0.9344 (1.0032) loss 0.9822 (0.9705) grad_norm 6.5139 (8.8121/2.4492) mem 68106MB [2022-12-19 11:48:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][560/1519] eta 0:16:03 lr 0.000031 time 0.9248 (1.0042) model_time 0.9247 (1.0031) loss 0.9192 (0.9701) grad_norm 9.5192 (8.8019/2.4336) mem 68106MB [2022-12-19 11:48:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][570/1519] eta 0:15:52 lr 0.000031 time 0.9224 (1.0041) model_time 0.9222 (1.0030) loss 0.8359 (0.9707) grad_norm 9.5753 (8.8058/2.4284) mem 68106MB [2022-12-19 11:48:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][580/1519] eta 0:15:42 lr 0.000031 time 0.9375 (1.0040) model_time 0.9373 (1.0029) loss 1.4195 (0.9701) grad_norm 7.8295 (8.7825/2.4160) mem 68106MB [2022-12-19 11:48:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][590/1519] eta 0:15:32 lr 0.000031 time 0.9258 (1.0039) model_time 0.9257 (1.0029) loss 0.7197 (0.9710) grad_norm 10.2926 (8.7814/2.4043) mem 68106MB [2022-12-19 11:49:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][600/1519] eta 0:15:22 lr 0.000031 time 0.9261 (1.0040) model_time 0.9260 (1.0029) loss 0.7936 (0.9705) grad_norm 8.8518 (8.7617/2.3982) mem 68106MB [2022-12-19 11:49:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][610/1519] eta 0:15:12 lr 0.000031 time 0.9310 (1.0040) model_time 0.9309 (1.0030) loss 1.0109 (0.9698) grad_norm 7.1314 (8.7646/2.3948) mem 68106MB [2022-12-19 11:49:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][620/1519] eta 0:15:02 lr 0.000031 time 0.9169 (1.0044) model_time 0.9167 (1.0034) loss 0.8997 (0.9699) grad_norm 11.3055 (8.7718/2.3920) mem 68106MB [2022-12-19 11:49:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][630/1519] eta 0:14:52 lr 0.000031 time 0.9323 (1.0044) model_time 0.9321 (1.0033) loss 1.0261 (0.9692) grad_norm 7.7810 (8.7658/2.3848) mem 68106MB [2022-12-19 11:49:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][640/1519] eta 0:14:42 lr 0.000031 time 0.9317 (1.0043) model_time 0.9315 (1.0033) loss 1.2063 (0.9696) grad_norm 7.5463 (8.7557/2.3851) mem 68106MB [2022-12-19 11:49:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][650/1519] eta 0:14:32 lr 0.000031 time 0.9313 (1.0042) model_time 0.9311 (1.0032) loss 1.0146 (0.9692) grad_norm 6.0535 (8.7123/2.3632) mem 68106MB [2022-12-19 11:50:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][660/1519] eta 0:14:22 lr 0.000031 time 0.9256 (1.0042) model_time 0.9255 (1.0032) loss 0.8961 (0.9688) grad_norm 7.4562 (8.7247/2.3607) mem 68106MB [2022-12-19 11:50:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][670/1519] eta 0:14:12 lr 0.000031 time 0.9236 (1.0041) model_time 0.9233 (1.0031) loss 0.8680 (0.9676) grad_norm 7.2083 (8.7003/2.3607) mem 68106MB [2022-12-19 11:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][680/1519] eta 0:14:02 lr 0.000031 time 0.9332 (1.0040) model_time 0.9330 (1.0031) loss 0.7993 (0.9672) grad_norm 8.3718 (8.6896/2.3588) mem 68106MB [2022-12-19 11:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][690/1519] eta 0:13:52 lr 0.000031 time 0.9343 (1.0040) model_time 0.9341 (1.0030) loss 0.8644 (0.9678) grad_norm 6.5661 (8.6489/2.3506) mem 68106MB [2022-12-19 11:50:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][700/1519] eta 0:13:42 lr 0.000031 time 0.9515 (1.0040) model_time 0.9514 (1.0031) loss 0.8756 (0.9679) grad_norm 8.0256 (8.6162/2.3433) mem 68106MB [2022-12-19 11:50:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][710/1519] eta 0:13:32 lr 0.000031 time 0.9551 (1.0043) model_time 0.9549 (1.0033) loss 0.9239 (0.9679) grad_norm 8.6944 (8.5545/2.2619) mem 68106MB [2022-12-19 11:51:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][720/1519] eta 0:13:22 lr 0.000031 time 0.9302 (1.0043) model_time 0.9300 (1.0034) loss 0.8102 (0.9674) grad_norm 7.2211 (8.5347/2.2610) mem 68106MB [2022-12-19 11:51:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][730/1519] eta 0:13:12 lr 0.000031 time 0.9373 (1.0042) model_time 0.9372 (1.0033) loss 1.1011 (0.9662) grad_norm 13.4026 (8.5622/2.2988) mem 68106MB [2022-12-19 11:51:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][740/1519] eta 0:13:02 lr 0.000031 time 0.9324 (1.0042) model_time 0.9323 (1.0032) loss 1.0781 (0.9675) grad_norm 11.7720 (8.5640/2.2782) mem 68106MB [2022-12-19 11:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][750/1519] eta 0:12:52 lr 0.000031 time 0.9355 (1.0041) model_time 0.9353 (1.0032) loss 0.8074 (0.9674) grad_norm 8.1758 (8.5279/2.1964) mem 68106MB [2022-12-19 11:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][760/1519] eta 0:12:42 lr 0.000031 time 0.9211 (1.0042) model_time 0.9209 (1.0033) loss 0.7470 (0.9673) grad_norm 9.0505 (8.5341/2.1910) mem 68106MB [2022-12-19 11:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][770/1519] eta 0:12:32 lr 0.000031 time 0.9224 (1.0041) model_time 0.9223 (1.0032) loss 0.7724 (0.9672) grad_norm 6.9946 (8.5292/2.1801) mem 68106MB [2022-12-19 11:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][780/1519] eta 0:12:22 lr 0.000031 time 0.9282 (1.0041) model_time 0.9281 (1.0032) loss 0.7829 (0.9666) grad_norm 8.4231 (8.4985/2.1779) mem 68106MB [2022-12-19 11:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][790/1519] eta 0:12:11 lr 0.000031 time 0.9203 (1.0040) model_time 0.9202 (1.0031) loss 1.1799 (0.9673) grad_norm 9.1359 (8.5200/2.1935) mem 68106MB [2022-12-19 11:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][800/1519] eta 0:12:02 lr 0.000031 time 0.9328 (1.0042) model_time 0.9326 (1.0033) loss 1.1964 (0.9672) grad_norm 8.0753 (8.5359/2.1894) mem 68106MB [2022-12-19 11:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][810/1519] eta 0:11:51 lr 0.000031 time 0.9310 (1.0041) model_time 0.9308 (1.0032) loss 0.8950 (0.9668) grad_norm 9.1659 (8.5280/2.1843) mem 68106MB [2022-12-19 11:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][820/1519] eta 0:11:41 lr 0.000031 time 0.9328 (1.0041) model_time 0.9327 (1.0032) loss 1.2543 (0.9668) grad_norm 7.6461 (8.5225/2.1680) mem 68106MB [2022-12-19 11:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][830/1519] eta 0:11:31 lr 0.000031 time 0.9346 (1.0040) model_time 0.9344 (1.0032) loss 1.1170 (0.9675) grad_norm 19.3153 (8.5576/2.2505) mem 68106MB [2022-12-19 11:53:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][840/1519] eta 0:11:21 lr 0.000031 time 0.9419 (1.0040) model_time 0.9418 (1.0031) loss 0.8486 (0.9673) grad_norm 6.1393 (8.5530/2.2547) mem 68106MB [2022-12-19 11:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][850/1519] eta 0:11:11 lr 0.000031 time 0.9351 (1.0039) model_time 0.9349 (1.0031) loss 0.8866 (0.9661) grad_norm 9.7183 (8.5806/2.2681) mem 68106MB [2022-12-19 11:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][860/1519] eta 0:11:01 lr 0.000031 time 0.9272 (1.0039) model_time 0.9270 (1.0030) loss 0.9572 (0.9668) grad_norm 11.9126 (8.5824/2.2685) mem 68106MB [2022-12-19 11:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][870/1519] eta 0:10:51 lr 0.000031 time 0.9217 (1.0038) model_time 0.9215 (1.0029) loss 1.0775 (0.9673) grad_norm 8.8745 (8.6204/2.2705) mem 68106MB [2022-12-19 11:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][880/1519] eta 0:10:41 lr 0.000031 time 0.9279 (1.0038) model_time 0.9278 (1.0029) loss 0.7963 (0.9670) grad_norm 10.4411 (8.5980/2.2610) mem 68106MB [2022-12-19 11:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][890/1519] eta 0:10:31 lr 0.000031 time 0.9437 (1.0037) model_time 0.9436 (1.0029) loss 0.8181 (0.9672) grad_norm 9.4899 (8.5942/2.2450) mem 68106MB [2022-12-19 11:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][900/1519] eta 0:10:21 lr 0.000031 time 0.9376 (1.0037) model_time 0.9374 (1.0028) loss 1.1064 (0.9672) grad_norm 10.0856 (8.5816/2.1853) mem 68106MB [2022-12-19 11:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][910/1519] eta 0:10:11 lr 0.000031 time 0.9376 (1.0038) model_time 0.9375 (1.0029) loss 0.9314 (0.9675) grad_norm 8.2152 (8.5621/2.1649) mem 68106MB [2022-12-19 11:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][920/1519] eta 0:10:01 lr 0.000031 time 0.9220 (1.0039) model_time 0.9218 (1.0030) loss 0.9280 (0.9672) grad_norm 7.8360 (8.5676/2.1681) mem 68106MB [2022-12-19 11:54:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][930/1519] eta 0:09:51 lr 0.000031 time 1.1850 (1.0041) model_time 1.1848 (1.0033) loss 1.0998 (0.9676) grad_norm 7.7548 (8.5765/2.1655) mem 68106MB [2022-12-19 11:54:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][940/1519] eta 0:09:41 lr 0.000031 time 0.9233 (1.0041) model_time 0.9232 (1.0033) loss 1.1509 (0.9672) grad_norm 8.1322 (8.5416/2.1417) mem 68106MB [2022-12-19 11:54:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][950/1519] eta 0:09:31 lr 0.000031 time 0.9241 (1.0041) model_time 0.9239 (1.0032) loss 1.1658 (0.9679) grad_norm 8.3731 (8.5570/2.1371) mem 68106MB [2022-12-19 11:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][960/1519] eta 0:09:21 lr 0.000031 time 0.9292 (1.0040) model_time 0.9291 (1.0032) loss 0.8100 (0.9673) grad_norm 8.0733 (8.5792/2.1636) mem 68106MB [2022-12-19 11:55:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][970/1519] eta 0:09:11 lr 0.000031 time 0.9337 (1.0039) model_time 0.9335 (1.0031) loss 0.9679 (0.9680) grad_norm 7.8169 (8.6180/2.2077) mem 68106MB [2022-12-19 11:55:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][980/1519] eta 0:09:01 lr 0.000031 time 0.9331 (1.0041) model_time 0.9324 (1.0033) loss 0.7296 (0.9677) grad_norm 9.8795 (8.6622/2.2754) mem 68106MB [2022-12-19 11:55:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][990/1519] eta 0:08:51 lr 0.000031 time 0.9378 (1.0040) model_time 0.9376 (1.0032) loss 1.4169 (0.9678) grad_norm 5.8807 (8.6661/2.2846) mem 68106MB [2022-12-19 11:55:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1000/1519] eta 0:08:41 lr 0.000031 time 0.9331 (1.0040) model_time 0.9330 (1.0032) loss 1.0060 (0.9675) grad_norm 9.4061 (8.6581/2.2869) mem 68106MB [2022-12-19 11:55:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1010/1519] eta 0:08:31 lr 0.000031 time 0.9280 (1.0040) model_time 0.9279 (1.0032) loss 1.1074 (0.9676) grad_norm 5.8779 (8.6431/2.2847) mem 68106MB [2022-12-19 11:56:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1020/1519] eta 0:08:21 lr 0.000031 time 0.9387 (1.0045) model_time 0.9386 (1.0037) loss 0.8684 (0.9671) grad_norm 7.7241 (8.6034/2.2833) mem 68106MB [2022-12-19 11:56:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1030/1519] eta 0:08:11 lr 0.000031 time 0.9422 (1.0047) model_time 0.9420 (1.0039) loss 0.8684 (0.9667) grad_norm 13.0642 (8.5903/2.2615) mem 68106MB [2022-12-19 11:56:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1040/1519] eta 0:08:01 lr 0.000031 time 0.9327 (1.0046) model_time 0.9326 (1.0038) loss 0.8607 (0.9666) grad_norm 7.8004 (8.5322/2.1256) mem 68106MB [2022-12-19 11:56:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1050/1519] eta 0:07:51 lr 0.000031 time 0.9418 (1.0046) model_time 0.9416 (1.0038) loss 0.9920 (0.9661) grad_norm 6.9170 (8.5507/2.1309) mem 68106MB [2022-12-19 11:56:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1060/1519] eta 0:07:41 lr 0.000031 time 0.9376 (1.0046) model_time 0.9374 (1.0038) loss 1.2011 (0.9664) grad_norm 9.9148 (8.5847/2.1371) mem 68106MB [2022-12-19 11:56:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1070/1519] eta 0:07:31 lr 0.000031 time 0.9321 (1.0045) model_time 0.9319 (1.0037) loss 0.8804 (0.9657) grad_norm 11.4595 (8.5972/2.1308) mem 68106MB [2022-12-19 11:57:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1080/1519] eta 0:07:20 lr 0.000031 time 0.9273 (1.0045) model_time 0.9271 (1.0037) loss 0.7910 (0.9649) grad_norm 9.5623 (8.5766/2.1005) mem 68106MB [2022-12-19 11:57:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1090/1519] eta 0:07:10 lr 0.000031 time 0.9434 (1.0045) model_time 0.9433 (1.0037) loss 1.2292 (0.9656) grad_norm 5.9247 (8.5559/2.1122) mem 68106MB [2022-12-19 11:57:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1100/1519] eta 0:07:00 lr 0.000031 time 0.9309 (1.0045) model_time 0.9308 (1.0037) loss 1.4524 (0.9666) grad_norm 15.1464 (8.5680/2.1125) mem 68106MB [2022-12-19 11:57:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1110/1519] eta 0:06:50 lr 0.000031 time 0.9317 (1.0044) model_time 0.9316 (1.0037) loss 1.1033 (0.9665) grad_norm 6.6345 (8.5621/2.1262) mem 68106MB [2022-12-19 11:57:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1120/1519] eta 0:06:40 lr 0.000031 time 0.9303 (1.0044) model_time 0.9302 (1.0036) loss 0.8391 (0.9661) grad_norm 6.8152 (8.5595/2.1234) mem 68106MB [2022-12-19 11:57:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1130/1519] eta 0:06:30 lr 0.000031 time 0.9323 (1.0044) model_time 0.9322 (1.0036) loss 1.2337 (0.9665) grad_norm 9.8285 (8.5877/2.1636) mem 68106MB [2022-12-19 11:58:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1140/1519] eta 0:06:20 lr 0.000031 time 0.9334 (1.0043) model_time 0.9332 (1.0036) loss 1.3387 (0.9661) grad_norm 9.5550 (8.5749/2.1548) mem 68106MB [2022-12-19 11:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1150/1519] eta 0:06:10 lr 0.000031 time 0.9304 (1.0043) model_time 0.9303 (1.0035) loss 1.4793 (0.9669) grad_norm 8.0935 (8.5674/2.1475) mem 68106MB [2022-12-19 11:58:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1160/1519] eta 0:06:00 lr 0.000031 time 0.9188 (1.0042) model_time 0.9187 (1.0035) loss 0.9419 (0.9664) grad_norm 8.8184 (8.5528/2.1507) mem 68106MB [2022-12-19 11:58:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1170/1519] eta 0:05:50 lr 0.000031 time 0.9277 (1.0042) model_time 0.9276 (1.0034) loss 1.0220 (0.9659) grad_norm 7.9606 (8.5531/2.1689) mem 68106MB [2022-12-19 11:58:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1180/1519] eta 0:05:40 lr 0.000031 time 0.9206 (1.0041) model_time 0.9204 (1.0034) loss 1.1369 (0.9664) grad_norm 7.4239 (8.5490/2.1724) mem 68106MB [2022-12-19 11:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1190/1519] eta 0:05:30 lr 0.000031 time 0.9252 (1.0042) model_time 0.9250 (1.0034) loss 0.8128 (0.9670) grad_norm 9.0016 (8.5515/2.1710) mem 68106MB [2022-12-19 11:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1200/1519] eta 0:05:20 lr 0.000031 time 0.9345 (1.0042) model_time 0.9343 (1.0034) loss 1.1126 (0.9676) grad_norm 14.0138 (8.5743/2.1872) mem 68106MB [2022-12-19 11:59:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1210/1519] eta 0:05:10 lr 0.000031 time 0.9234 (1.0042) model_time 0.9233 (1.0034) loss 1.0273 (0.9677) grad_norm 10.2580 (8.5656/2.1897) mem 68106MB [2022-12-19 11:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1220/1519] eta 0:05:00 lr 0.000031 time 0.9336 (1.0041) model_time 0.9335 (1.0034) loss 1.3727 (0.9680) grad_norm 9.5600 (8.5674/2.1892) mem 68106MB [2022-12-19 11:59:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1230/1519] eta 0:04:50 lr 0.000031 time 0.9891 (1.0041) model_time 0.9889 (1.0034) loss 0.8520 (0.9677) grad_norm 10.8256 (8.5600/2.1949) mem 68106MB [2022-12-19 11:59:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1240/1519] eta 0:04:40 lr 0.000031 time 0.9321 (1.0041) model_time 0.9319 (1.0034) loss 1.1885 (0.9673) grad_norm 8.8580 (8.5602/2.1971) mem 68106MB [2022-12-19 11:59:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1250/1519] eta 0:04:30 lr 0.000031 time 0.9309 (1.0043) model_time 0.9308 (1.0035) loss 0.9280 (0.9672) grad_norm 10.6306 (8.5609/2.1977) mem 68106MB [2022-12-19 12:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1260/1519] eta 0:04:20 lr 0.000031 time 0.9322 (1.0042) model_time 0.9320 (1.0035) loss 1.0672 (0.9674) grad_norm 9.4507 (8.5384/2.2050) mem 68106MB [2022-12-19 12:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1270/1519] eta 0:04:10 lr 0.000031 time 0.9302 (1.0041) model_time 0.9301 (1.0034) loss 1.2780 (0.9674) grad_norm 7.2869 (8.5429/2.2188) mem 68106MB [2022-12-19 12:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1280/1519] eta 0:03:59 lr 0.000031 time 0.9323 (1.0041) model_time 0.9321 (1.0034) loss 0.8738 (0.9668) grad_norm 6.3993 (8.5279/2.2224) mem 68106MB [2022-12-19 12:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1290/1519] eta 0:03:49 lr 0.000031 time 0.9219 (1.0041) model_time 0.9218 (1.0034) loss 0.8441 (0.9662) grad_norm 6.7099 (8.5635/2.2212) mem 68106MB [2022-12-19 12:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1300/1519] eta 0:03:39 lr 0.000031 time 0.9352 (1.0041) model_time 0.9350 (1.0034) loss 0.8288 (0.9661) grad_norm 13.4201 (8.6055/2.2419) mem 68106MB [2022-12-19 12:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1310/1519] eta 0:03:29 lr 0.000031 time 0.9352 (1.0041) model_time 0.9350 (1.0034) loss 1.0582 (0.9670) grad_norm 8.5641 (8.6338/2.2383) mem 68106MB [2022-12-19 12:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1320/1519] eta 0:03:19 lr 0.000031 time 0.9360 (1.0041) model_time 0.9358 (1.0034) loss 1.3429 (0.9678) grad_norm 10.0522 (8.6626/2.2315) mem 68106MB [2022-12-19 12:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1330/1519] eta 0:03:09 lr 0.000031 time 0.9349 (1.0041) model_time 0.9347 (1.0034) loss 0.7512 (0.9668) grad_norm 10.6658 (8.6256/2.2025) mem 68106MB [2022-12-19 12:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1340/1519] eta 0:02:59 lr 0.000031 time 0.9267 (1.0041) model_time 0.9265 (1.0034) loss 0.9494 (0.9663) grad_norm 8.0188 (8.6052/2.2064) mem 68106MB [2022-12-19 12:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1350/1519] eta 0:02:49 lr 0.000031 time 0.9435 (1.0041) model_time 0.9433 (1.0033) loss 0.8778 (0.9662) grad_norm 7.8871 (8.5951/2.1926) mem 68106MB [2022-12-19 12:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1360/1519] eta 0:02:39 lr 0.000031 time 0.9195 (1.0040) model_time 0.9190 (1.0033) loss 0.6927 (0.9659) grad_norm 8.6752 (8.6218/2.2247) mem 68106MB [2022-12-19 12:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1370/1519] eta 0:02:29 lr 0.000031 time 0.9227 (1.0040) model_time 0.9225 (1.0033) loss 0.7534 (0.9656) grad_norm 11.9379 (8.6506/2.2277) mem 68106MB [2022-12-19 12:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1380/1519] eta 0:02:19 lr 0.000031 time 0.9173 (1.0039) model_time 0.9171 (1.0032) loss 0.8063 (0.9657) grad_norm 7.5292 (8.6785/2.2196) mem 68106MB [2022-12-19 12:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1390/1519] eta 0:02:09 lr 0.000031 time 0.9301 (1.0039) model_time 0.9299 (1.0032) loss 0.9853 (0.9659) grad_norm 8.9477 (8.6519/2.2090) mem 68106MB [2022-12-19 12:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1400/1519] eta 0:01:59 lr 0.000031 time 0.9126 (1.0039) model_time 0.9124 (1.0032) loss 0.9765 (0.9658) grad_norm 8.7219 (8.6627/2.2173) mem 68106MB [2022-12-19 12:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1410/1519] eta 0:01:49 lr 0.000031 time 1.0197 (1.0039) model_time 1.0195 (1.0032) loss 0.9552 (0.9655) grad_norm 7.4260 (8.6467/2.2148) mem 68106MB [2022-12-19 12:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1420/1519] eta 0:01:39 lr 0.000031 time 0.8827 (1.0040) model_time 0.8825 (1.0033) loss 0.7689 (0.9647) grad_norm 11.9235 (8.6886/2.2488) mem 68106MB [2022-12-19 12:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1430/1519] eta 0:01:29 lr 0.000031 time 0.9330 (1.0042) model_time 0.9329 (1.0035) loss 1.2810 (0.9647) grad_norm 8.9902 (8.6588/2.1621) mem 68106MB [2022-12-19 12:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1440/1519] eta 0:01:19 lr 0.000031 time 0.9362 (1.0042) model_time 0.9361 (1.0035) loss 1.0020 (0.9643) grad_norm 6.5938 (8.6763/2.1485) mem 68106MB [2022-12-19 12:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1450/1519] eta 0:01:09 lr 0.000031 time 0.9340 (1.0041) model_time 0.9339 (1.0034) loss 1.0696 (0.9646) grad_norm 7.6473 (8.6795/2.1748) mem 68106MB [2022-12-19 12:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1460/1519] eta 0:00:59 lr 0.000031 time 0.9293 (1.0041) model_time 0.9292 (1.0034) loss 1.2023 (0.9657) grad_norm 8.3439 (8.6870/2.1783) mem 68106MB [2022-12-19 12:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1470/1519] eta 0:00:49 lr 0.000031 time 0.9265 (1.0040) model_time 0.9263 (1.0033) loss 0.9952 (0.9656) grad_norm 11.6800 (8.6968/2.1860) mem 68106MB [2022-12-19 12:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1480/1519] eta 0:00:39 lr 0.000031 time 0.9379 (1.0040) model_time 0.9377 (1.0033) loss 1.0891 (0.9653) grad_norm 9.4311 (8.7058/2.1896) mem 68106MB [2022-12-19 12:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1490/1519] eta 0:00:29 lr 0.000031 time 0.9310 (1.0039) model_time 0.9309 (1.0033) loss 1.1412 (0.9653) grad_norm 9.4909 (8.6896/2.1910) mem 68106MB [2022-12-19 12:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1500/1519] eta 0:00:19 lr 0.000031 time 0.9315 (1.0039) model_time 0.9314 (1.0032) loss 1.1249 (0.9660) grad_norm 10.9765 (8.7057/2.2105) mem 68106MB [2022-12-19 12:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [17/100][1510/1519] eta 0:00:09 lr 0.000031 time 0.9313 (1.0040) model_time 0.9312 (1.0033) loss 0.8674 (0.9656) grad_norm 9.4027 (8.6852/2.2108) mem 68106MB [2022-12-19 12:04:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 17 training takes 0:25:25 [2022-12-19 12:04:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_17.pth saving...... [2022-12-19 12:05:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_17.pth saved !!! [2022-12-19 12:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.639 (0.639) Loss 0.5513 (0.5513) Acc@1 89.931 (89.931) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 12:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.328) Loss 0.5802 (0.5573) Acc@1 90.625 (89.552) Acc@5 96.875 (97.948) Mem 68106MB [2022-12-19 12:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.312) Loss 0.5228 (0.5607) Acc@1 91.667 (89.302) Acc@5 98.611 (97.900) Mem 68106MB [2022-12-19 12:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.307) Loss 0.6439 (0.5660) Acc@1 86.806 (89.214) Acc@5 96.875 (97.782) Mem 68106MB [2022-12-19 12:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.304) Loss 0.5633 (0.5568) Acc@1 89.236 (89.397) Acc@5 98.264 (97.908) Mem 68106MB [2022-12-19 12:05:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.303) Loss 0.5521 (0.5520) Acc@1 87.847 (89.529) Acc@5 97.569 (97.951) Mem 68106MB [2022-12-19 12:05:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.302) Loss 0.5887 (0.5527) Acc@1 87.500 (89.492) Acc@5 97.569 (97.962) Mem 68106MB [2022-12-19 12:05:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.302) Loss 0.6279 (0.5551) Acc@1 89.931 (89.437) Acc@5 98.264 (97.975) Mem 68106MB [2022-12-19 12:05:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.301) Loss 0.4906 (0.5543) Acc@1 90.278 (89.429) Acc@5 97.917 (97.994) Mem 68106MB [2022-12-19 12:05:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:17] * Acc@1 89.403 Acc@5 98.007 [2022-12-19 12:05:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 89.4% [2022-12-19 12:05:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 12:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 12:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 89.40% [2022-12-19 12:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][0/1519] eta 0:36:11 lr 0.000031 time 1.4294 (1.4294) model_time 1.0054 (1.0054) loss 0.8288 (0.8288) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 12:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][10/1519] eta 0:26:09 lr 0.000031 time 0.9180 (1.0399) model_time 0.9178 (1.0007) loss 0.9781 (1.0061) grad_norm 6.7388 (8.2196/2.1555) mem 68106MB [2022-12-19 12:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][20/1519] eta 0:25:31 lr 0.000031 time 0.9474 (1.0217) model_time 0.9473 (1.0009) loss 0.9651 (0.9906) grad_norm 6.3157 (7.8428/1.8123) mem 68106MB [2022-12-19 12:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][30/1519] eta 0:25:15 lr 0.000031 time 0.9673 (1.0176) model_time 0.9672 (1.0034) loss 0.7893 (0.9899) grad_norm 10.4023 (8.2800/1.8096) mem 68106MB [2022-12-19 12:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][40/1519] eta 0:24:58 lr 0.000031 time 0.9315 (1.0135) model_time 0.9314 (1.0027) loss 1.1091 (1.0216) grad_norm 9.0227 (8.1917/1.6169) mem 68106MB [2022-12-19 12:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][50/1519] eta 0:24:52 lr 0.000031 time 0.9273 (1.0162) model_time 0.9271 (1.0074) loss 1.0123 (1.0034) grad_norm 8.3962 (8.4013/1.9112) mem 68106MB [2022-12-19 12:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][60/1519] eta 0:24:38 lr 0.000031 time 0.9338 (1.0135) model_time 0.9337 (1.0061) loss 0.7983 (0.9991) grad_norm 5.9558 (8.3364/2.0318) mem 68106MB [2022-12-19 12:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][70/1519] eta 0:24:26 lr 0.000031 time 0.9322 (1.0118) model_time 0.9321 (1.0054) loss 0.7740 (0.9836) grad_norm 8.5818 (8.3691/2.0580) mem 68106MB [2022-12-19 12:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][80/1519] eta 0:24:13 lr 0.000031 time 0.9427 (1.0101) model_time 0.9426 (1.0044) loss 0.9295 (0.9876) grad_norm 7.8457 (8.2333/1.9646) mem 68106MB [2022-12-19 12:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][90/1519] eta 0:24:03 lr 0.000031 time 0.9313 (1.0099) model_time 0.9311 (1.0048) loss 0.8312 (0.9837) grad_norm 6.6698 (8.0887/1.9062) mem 68106MB [2022-12-19 12:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][100/1519] eta 0:23:52 lr 0.000031 time 0.9979 (1.0094) model_time 0.9978 (1.0048) loss 1.3819 (0.9888) grad_norm 9.4353 (8.1787/1.8940) mem 68106MB [2022-12-19 12:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][110/1519] eta 0:23:40 lr 0.000031 time 0.9303 (1.0083) model_time 0.9301 (1.0040) loss 1.2279 (0.9871) grad_norm 7.3371 (8.4047/2.1796) mem 68106MB [2022-12-19 12:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][120/1519] eta 0:23:31 lr 0.000031 time 0.9397 (1.0087) model_time 0.9396 (1.0048) loss 0.7701 (0.9837) grad_norm 5.9187 (8.2911/2.1624) mem 68106MB [2022-12-19 12:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][130/1519] eta 0:23:23 lr 0.000031 time 0.9410 (1.0103) model_time 0.9408 (1.0066) loss 1.2253 (0.9812) grad_norm 5.9774 (8.2344/2.1492) mem 68106MB [2022-12-19 12:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][140/1519] eta 0:23:12 lr 0.000031 time 0.9287 (1.0095) model_time 0.9284 (1.0060) loss 0.8608 (0.9813) grad_norm 7.0692 (8.2827/2.1138) mem 68106MB [2022-12-19 12:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][150/1519] eta 0:23:01 lr 0.000031 time 0.9586 (1.0090) model_time 0.9583 (1.0057) loss 0.9303 (0.9765) grad_norm 5.9944 (8.2145/2.1161) mem 68106MB [2022-12-19 12:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][160/1519] eta 0:22:50 lr 0.000031 time 0.9318 (1.0083) model_time 0.9317 (1.0052) loss 0.8315 (0.9768) grad_norm 5.6641 (8.1902/2.0703) mem 68106MB [2022-12-19 12:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][170/1519] eta 0:22:39 lr 0.000031 time 0.9397 (1.0080) model_time 0.9395 (1.0051) loss 1.0036 (0.9748) grad_norm 7.5293 (8.2024/2.0326) mem 68106MB [2022-12-19 12:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][180/1519] eta 0:22:29 lr 0.000031 time 0.8990 (1.0078) model_time 0.8989 (1.0050) loss 1.1311 (0.9766) grad_norm 7.6817 (8.2428/1.9966) mem 68106MB [2022-12-19 12:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][190/1519] eta 0:22:19 lr 0.000031 time 0.9856 (1.0076) model_time 0.9854 (1.0049) loss 1.0147 (0.9751) grad_norm 12.4948 (8.2836/2.0056) mem 68106MB [2022-12-19 12:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][200/1519] eta 0:22:08 lr 0.000031 time 0.9338 (1.0071) model_time 0.9337 (1.0045) loss 1.5560 (0.9785) grad_norm 9.0091 (8.2986/1.9879) mem 68106MB [2022-12-19 12:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][210/1519] eta 0:21:59 lr 0.000031 time 0.9872 (1.0083) model_time 0.9871 (1.0059) loss 0.9204 (0.9808) grad_norm 5.9556 (8.2487/1.9725) mem 68106MB [2022-12-19 12:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][220/1519] eta 0:21:49 lr 0.000031 time 0.9353 (1.0082) model_time 0.9352 (1.0059) loss 0.9677 (0.9812) grad_norm 7.6516 (8.2194/1.9396) mem 68106MB [2022-12-19 12:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][230/1519] eta 0:21:38 lr 0.000031 time 0.9341 (1.0078) model_time 0.9339 (1.0055) loss 1.0081 (0.9778) grad_norm 7.3549 (8.2383/1.9168) mem 68106MB [2022-12-19 12:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][240/1519] eta 0:21:28 lr 0.000031 time 0.9303 (1.0074) model_time 0.9301 (1.0052) loss 0.9957 (0.9764) grad_norm 16.0333 (8.3379/2.0250) mem 68106MB [2022-12-19 12:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][250/1519] eta 0:21:17 lr 0.000031 time 0.9274 (1.0071) model_time 0.9272 (1.0050) loss 1.0273 (0.9776) grad_norm 9.1812 (8.3259/1.9988) mem 68106MB [2022-12-19 12:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][260/1519] eta 0:21:07 lr 0.000031 time 0.9378 (1.0068) model_time 0.9376 (1.0048) loss 0.9484 (0.9771) grad_norm 6.0699 (8.3415/1.9905) mem 68106MB [2022-12-19 12:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][270/1519] eta 0:20:57 lr 0.000031 time 0.9275 (1.0066) model_time 0.9274 (1.0046) loss 0.7976 (0.9776) grad_norm 9.2752 (8.4227/2.0103) mem 68106MB [2022-12-19 12:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][280/1519] eta 0:20:47 lr 0.000031 time 0.9863 (1.0065) model_time 0.9862 (1.0046) loss 0.8383 (0.9772) grad_norm 9.0583 (8.4670/2.0053) mem 68106MB [2022-12-19 12:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][290/1519] eta 0:20:36 lr 0.000031 time 0.9412 (1.0062) model_time 0.9411 (1.0044) loss 0.8073 (0.9763) grad_norm 5.8701 (8.4169/1.9975) mem 68106MB [2022-12-19 12:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][300/1519] eta 0:20:26 lr 0.000031 time 1.0514 (1.0064) model_time 1.0512 (1.0046) loss 0.8135 (0.9769) grad_norm 5.8142 (8.3860/1.9879) mem 68106MB [2022-12-19 12:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][310/1519] eta 0:20:16 lr 0.000031 time 0.9754 (1.0063) model_time 0.9753 (1.0046) loss 0.9545 (0.9765) grad_norm 8.0225 (8.4189/2.0222) mem 68106MB [2022-12-19 12:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][320/1519] eta 0:20:06 lr 0.000031 time 0.9308 (1.0060) model_time 0.9307 (1.0043) loss 0.8993 (0.9741) grad_norm 7.3200 (8.4281/2.0108) mem 68106MB [2022-12-19 12:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][330/1519] eta 0:19:55 lr 0.000031 time 0.9308 (1.0058) model_time 0.9306 (1.0041) loss 0.7757 (0.9725) grad_norm 11.9729 (8.4731/2.0166) mem 68106MB [2022-12-19 12:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][340/1519] eta 0:19:45 lr 0.000031 time 0.9336 (1.0057) model_time 0.9333 (1.0041) loss 1.0936 (0.9734) grad_norm 13.1928 (8.5074/2.0449) mem 68106MB [2022-12-19 12:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][350/1519] eta 0:19:35 lr 0.000031 time 0.9376 (1.0054) model_time 0.9375 (1.0039) loss 1.1124 (0.9721) grad_norm 8.9052 (8.5066/2.0265) mem 68106MB [2022-12-19 12:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][360/1519] eta 0:19:25 lr 0.000031 time 0.9275 (1.0056) model_time 0.9274 (1.0040) loss 0.7509 (0.9717) grad_norm 9.4884 (8.5566/2.0551) mem 68106MB [2022-12-19 12:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][370/1519] eta 0:19:15 lr 0.000031 time 0.9882 (1.0056) model_time 0.9880 (1.0040) loss 1.2835 (0.9702) grad_norm 7.4344 (8.5668/2.0460) mem 68106MB [2022-12-19 12:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][380/1519] eta 0:19:05 lr 0.000031 time 0.9333 (1.0054) model_time 0.9332 (1.0039) loss 0.9270 (0.9698) grad_norm 10.7902 (8.6084/2.0774) mem 68106MB [2022-12-19 12:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][390/1519] eta 0:18:55 lr 0.000031 time 0.9355 (1.0054) model_time 0.9353 (1.0039) loss 1.1801 (0.9694) grad_norm 11.3698 (8.5929/2.0774) mem 68106MB [2022-12-19 12:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][400/1519] eta 0:18:45 lr 0.000031 time 0.9376 (1.0060) model_time 0.9375 (1.0045) loss 0.7505 (0.9657) grad_norm 10.1960 (8.6153/2.0837) mem 68106MB [2022-12-19 12:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][410/1519] eta 0:18:35 lr 0.000031 time 0.9303 (1.0058) model_time 0.9302 (1.0044) loss 1.2474 (0.9683) grad_norm 8.7650 (8.5967/2.0786) mem 68106MB [2022-12-19 12:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][420/1519] eta 0:18:25 lr 0.000031 time 0.9241 (1.0056) model_time 0.9240 (1.0042) loss 0.9980 (0.9676) grad_norm 10.3383 (8.6216/2.0696) mem 68106MB [2022-12-19 12:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][430/1519] eta 0:18:15 lr 0.000031 time 0.9269 (1.0055) model_time 0.9268 (1.0042) loss 0.8195 (0.9682) grad_norm 6.8031 (8.6239/2.0547) mem 68106MB [2022-12-19 12:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][440/1519] eta 0:18:04 lr 0.000031 time 0.9323 (1.0054) model_time 0.9321 (1.0041) loss 0.8369 (0.9660) grad_norm 14.3862 (8.6326/2.0758) mem 68106MB [2022-12-19 12:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][450/1519] eta 0:17:54 lr 0.000031 time 0.9285 (1.0053) model_time 0.9283 (1.0040) loss 0.7729 (0.9678) grad_norm 7.6061 (8.6200/2.0584) mem 68106MB [2022-12-19 12:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][460/1519] eta 0:17:44 lr 0.000031 time 0.9297 (1.0051) model_time 0.9295 (1.0038) loss 0.9716 (0.9665) grad_norm 10.1472 (8.6290/2.0529) mem 68106MB [2022-12-19 12:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][470/1519] eta 0:17:34 lr 0.000031 time 0.9334 (1.0050) model_time 0.9332 (1.0037) loss 1.0420 (0.9671) grad_norm 7.0141 (8.6331/2.0603) mem 68106MB [2022-12-19 12:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][480/1519] eta 0:17:24 lr 0.000031 time 0.9849 (1.0051) model_time 0.9848 (1.0038) loss 1.1081 (0.9666) grad_norm 11.5185 (8.6284/2.0624) mem 68106MB [2022-12-19 12:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][490/1519] eta 0:17:14 lr 0.000031 time 1.0129 (1.0052) model_time 1.0127 (1.0040) loss 0.9538 (0.9673) grad_norm 6.3789 (8.6224/2.0830) mem 68106MB [2022-12-19 12:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][500/1519] eta 0:17:04 lr 0.000031 time 0.9327 (1.0051) model_time 0.9326 (1.0039) loss 0.8542 (0.9677) grad_norm 7.2750 (8.5895/2.0877) mem 68106MB [2022-12-19 12:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][510/1519] eta 0:16:54 lr 0.000031 time 0.9265 (1.0050) model_time 0.9263 (1.0038) loss 0.9595 (0.9692) grad_norm 9.2505 (8.6251/2.1163) mem 68106MB [2022-12-19 12:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][520/1519] eta 0:16:43 lr 0.000031 time 0.9416 (1.0049) model_time 0.9415 (1.0037) loss 1.0996 (0.9677) grad_norm 7.0122 (8.6222/2.1325) mem 68106MB [2022-12-19 12:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][530/1519] eta 0:16:34 lr 0.000031 time 0.9252 (1.0051) model_time 0.9250 (1.0039) loss 0.9072 (0.9673) grad_norm 7.8626 (8.6589/2.1714) mem 68106MB [2022-12-19 12:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][540/1519] eta 0:16:23 lr 0.000031 time 0.9341 (1.0050) model_time 0.9339 (1.0038) loss 1.0283 (0.9675) grad_norm 6.9421 (8.6478/2.1573) mem 68106MB [2022-12-19 12:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][550/1519] eta 0:16:13 lr 0.000031 time 0.9352 (1.0050) model_time 0.9350 (1.0038) loss 0.9118 (0.9678) grad_norm 7.4859 (8.6700/2.1928) mem 68106MB [2022-12-19 12:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][560/1519] eta 0:16:03 lr 0.000031 time 0.9334 (1.0049) model_time 0.9332 (1.0037) loss 1.2478 (0.9685) grad_norm 6.1423 (8.6631/2.1802) mem 68106MB [2022-12-19 12:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][570/1519] eta 0:15:53 lr 0.000031 time 0.9354 (1.0047) model_time 0.9353 (1.0036) loss 0.9825 (0.9671) grad_norm 9.7011 (8.6619/2.1635) mem 68106MB [2022-12-19 12:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][580/1519] eta 0:15:43 lr 0.000031 time 0.9338 (1.0047) model_time 0.9337 (1.0036) loss 1.0650 (0.9660) grad_norm 9.6360 (8.6581/2.1501) mem 68106MB [2022-12-19 12:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][590/1519] eta 0:15:33 lr 0.000031 time 0.9331 (1.0046) model_time 0.9329 (1.0035) loss 1.4396 (0.9669) grad_norm 9.2503 (8.6571/2.1390) mem 68106MB [2022-12-19 12:16:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][600/1519] eta 0:15:23 lr 0.000031 time 0.9443 (1.0046) model_time 0.9442 (1.0036) loss 0.8585 (0.9669) grad_norm 10.5209 (8.6772/2.1563) mem 68106MB [2022-12-19 12:16:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][610/1519] eta 0:15:13 lr 0.000031 time 0.9342 (1.0045) model_time 0.9341 (1.0035) loss 0.8134 (0.9675) grad_norm 9.6841 (8.6879/2.1396) mem 68106MB [2022-12-19 12:16:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][620/1519] eta 0:15:03 lr 0.000031 time 0.9314 (1.0046) model_time 0.9312 (1.0035) loss 0.8582 (0.9665) grad_norm 7.3103 (8.7133/2.1398) mem 68106MB [2022-12-19 12:16:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][630/1519] eta 0:14:53 lr 0.000031 time 0.9308 (1.0046) model_time 0.9307 (1.0035) loss 0.7814 (0.9660) grad_norm 8.3006 (8.6974/2.1385) mem 68106MB [2022-12-19 12:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][640/1519] eta 0:14:42 lr 0.000031 time 0.9357 (1.0045) model_time 0.9355 (1.0034) loss 1.0033 (0.9674) grad_norm 7.8307 (8.6993/2.1458) mem 68106MB [2022-12-19 12:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][650/1519] eta 0:14:32 lr 0.000031 time 0.9315 (1.0045) model_time 0.9313 (1.0034) loss 0.7790 (0.9661) grad_norm 5.8956 (8.6773/2.1291) mem 68106MB [2022-12-19 12:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][660/1519] eta 0:14:22 lr 0.000031 time 0.9336 (1.0044) model_time 0.9334 (1.0034) loss 0.9047 (0.9659) grad_norm 13.1754 (8.7106/2.1207) mem 68106MB [2022-12-19 12:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][670/1519] eta 0:14:12 lr 0.000031 time 0.9285 (1.0043) model_time 0.9283 (1.0033) loss 0.8829 (0.9657) grad_norm 8.2270 (8.7127/2.1051) mem 68106MB [2022-12-19 12:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][680/1519] eta 0:14:02 lr 0.000031 time 0.9377 (1.0045) model_time 0.9376 (1.0035) loss 0.8471 (0.9658) grad_norm 10.1717 (8.7475/2.1089) mem 68106MB [2022-12-19 12:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][690/1519] eta 0:13:52 lr 0.000031 time 0.9297 (1.0045) model_time 0.9294 (1.0035) loss 0.7070 (0.9658) grad_norm 9.4960 (8.8051/2.1292) mem 68106MB [2022-12-19 12:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][700/1519] eta 0:13:42 lr 0.000031 time 0.9332 (1.0045) model_time 0.9331 (1.0036) loss 1.1417 (0.9668) grad_norm 9.3888 (8.7998/2.1269) mem 68106MB [2022-12-19 12:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][710/1519] eta 0:13:32 lr 0.000031 time 0.9217 (1.0048) model_time 0.9216 (1.0038) loss 0.8547 (0.9660) grad_norm 9.6040 (8.7714/2.0885) mem 68106MB [2022-12-19 12:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][720/1519] eta 0:13:22 lr 0.000031 time 0.9388 (1.0047) model_time 0.9386 (1.0038) loss 1.1727 (0.9662) grad_norm 12.7404 (8.8133/2.0868) mem 68106MB [2022-12-19 12:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][730/1519] eta 0:13:12 lr 0.000031 time 0.9365 (1.0047) model_time 0.9364 (1.0037) loss 0.9465 (0.9656) grad_norm 7.0013 (8.8299/2.0742) mem 68106MB [2022-12-19 12:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][740/1519] eta 0:13:02 lr 0.000031 time 0.9632 (1.0046) model_time 0.9631 (1.0036) loss 0.8996 (0.9653) grad_norm 7.4394 (8.8509/2.1083) mem 68106MB [2022-12-19 12:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][750/1519] eta 0:12:52 lr 0.000031 time 0.9826 (1.0046) model_time 0.9824 (1.0036) loss 0.7498 (0.9653) grad_norm 8.9235 (8.8766/2.0838) mem 68106MB [2022-12-19 12:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][760/1519] eta 0:12:42 lr 0.000031 time 0.9372 (1.0045) model_time 0.9370 (1.0036) loss 1.0233 (0.9661) grad_norm 8.5217 (8.8982/2.0912) mem 68106MB [2022-12-19 12:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][770/1519] eta 0:12:32 lr 0.000031 time 0.9291 (1.0044) model_time 0.9290 (1.0035) loss 0.8120 (0.9650) grad_norm 7.5894 (8.8799/2.0964) mem 68106MB [2022-12-19 12:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][780/1519] eta 0:12:22 lr 0.000031 time 0.9282 (1.0044) model_time 0.9281 (1.0034) loss 1.0293 (0.9659) grad_norm 7.8712 (8.8850/2.1344) mem 68106MB [2022-12-19 12:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][790/1519] eta 0:12:12 lr 0.000031 time 0.9271 (1.0043) model_time 0.9270 (1.0034) loss 0.9287 (0.9658) grad_norm 18.1506 (8.9205/2.1948) mem 68106MB [2022-12-19 12:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][800/1519] eta 0:12:02 lr 0.000031 time 0.9323 (1.0045) model_time 0.9321 (1.0035) loss 1.2072 (0.9662) grad_norm 6.2400 (8.9124/2.2142) mem 68106MB [2022-12-19 12:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][810/1519] eta 0:11:52 lr 0.000031 time 0.9276 (1.0045) model_time 0.9275 (1.0036) loss 0.8780 (0.9659) grad_norm 6.1932 (8.9222/2.2107) mem 68106MB [2022-12-19 12:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][820/1519] eta 0:11:42 lr 0.000031 time 0.9325 (1.0045) model_time 0.9322 (1.0036) loss 1.1067 (0.9665) grad_norm 8.4735 (8.9654/2.2289) mem 68106MB [2022-12-19 12:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][830/1519] eta 0:11:32 lr 0.000031 time 0.9331 (1.0044) model_time 0.9330 (1.0035) loss 1.2634 (0.9668) grad_norm 10.3580 (8.9844/2.2554) mem 68106MB [2022-12-19 12:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][840/1519] eta 0:11:21 lr 0.000031 time 0.9291 (1.0044) model_time 0.9289 (1.0035) loss 0.8606 (0.9663) grad_norm 5.5259 (8.9388/2.2241) mem 68106MB [2022-12-19 12:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][850/1519] eta 0:11:11 lr 0.000031 time 0.9276 (1.0044) model_time 0.9275 (1.0035) loss 1.1721 (0.9676) grad_norm 8.3765 (8.9496/2.2322) mem 68106MB [2022-12-19 12:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][860/1519] eta 0:11:01 lr 0.000031 time 0.9340 (1.0043) model_time 0.9338 (1.0034) loss 0.8621 (0.9665) grad_norm 8.8388 (8.9492/2.2301) mem 68106MB [2022-12-19 12:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][870/1519] eta 0:10:51 lr 0.000031 time 0.9325 (1.0042) model_time 0.9324 (1.0034) loss 0.7521 (0.9653) grad_norm 6.5497 (8.8933/2.2283) mem 68106MB [2022-12-19 12:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][880/1519] eta 0:10:41 lr 0.000031 time 0.9285 (1.0042) model_time 0.9283 (1.0033) loss 0.9282 (0.9647) grad_norm 9.7477 (8.8625/2.2291) mem 68106MB [2022-12-19 12:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][890/1519] eta 0:10:31 lr 0.000031 time 0.9318 (1.0042) model_time 0.9317 (1.0033) loss 0.8927 (0.9648) grad_norm 9.7841 (8.9255/2.2996) mem 68106MB [2022-12-19 12:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][900/1519] eta 0:10:21 lr 0.000031 time 0.9293 (1.0041) model_time 0.9291 (1.0033) loss 0.8101 (0.9651) grad_norm 8.3592 (8.9854/2.3316) mem 68106MB [2022-12-19 12:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][910/1519] eta 0:10:11 lr 0.000031 time 0.9175 (1.0041) model_time 0.9174 (1.0032) loss 0.8776 (0.9657) grad_norm 6.8798 (8.9642/2.3141) mem 68106MB [2022-12-19 12:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][920/1519] eta 0:10:01 lr 0.000031 time 0.9379 (1.0040) model_time 0.9378 (1.0032) loss 0.7625 (0.9659) grad_norm 8.8026 (8.9704/2.3071) mem 68106MB [2022-12-19 12:21:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][930/1519] eta 0:09:51 lr 0.000031 time 0.9801 (1.0040) model_time 0.9799 (1.0032) loss 0.7308 (0.9665) grad_norm 7.4598 (8.9542/2.3022) mem 68106MB [2022-12-19 12:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][940/1519] eta 0:09:41 lr 0.000031 time 0.9835 (1.0040) model_time 0.9833 (1.0032) loss 0.9543 (0.9662) grad_norm 7.8686 (8.9434/2.2853) mem 68106MB [2022-12-19 12:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][950/1519] eta 0:09:31 lr 0.000031 time 0.9341 (1.0040) model_time 0.9339 (1.0032) loss 0.8736 (0.9657) grad_norm 8.7404 (8.9483/2.2820) mem 68106MB [2022-12-19 12:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][960/1519] eta 0:09:21 lr 0.000031 time 0.9763 (1.0040) model_time 0.9761 (1.0032) loss 1.3081 (0.9662) grad_norm 9.4353 (8.9292/2.2613) mem 68106MB [2022-12-19 12:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][970/1519] eta 0:09:11 lr 0.000031 time 0.9334 (1.0039) model_time 0.9333 (1.0031) loss 1.1298 (0.9672) grad_norm 7.8945 (8.9184/2.2570) mem 68106MB [2022-12-19 12:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][980/1519] eta 0:09:01 lr 0.000031 time 0.9303 (1.0039) model_time 0.9302 (1.0031) loss 0.8353 (0.9672) grad_norm 9.2365 (8.8904/2.2313) mem 68106MB [2022-12-19 12:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][990/1519] eta 0:08:51 lr 0.000031 time 0.9360 (1.0041) model_time 0.9358 (1.0033) loss 0.9423 (0.9671) grad_norm 7.3574 (8.9297/2.2483) mem 68106MB [2022-12-19 12:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1000/1519] eta 0:08:41 lr 0.000031 time 0.9331 (1.0041) model_time 0.9330 (1.0033) loss 1.0556 (0.9664) grad_norm 7.4638 (8.9162/2.2450) mem 68106MB [2022-12-19 12:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1010/1519] eta 0:08:31 lr 0.000031 time 1.0078 (1.0041) model_time 1.0076 (1.0033) loss 1.0806 (0.9674) grad_norm 11.5409 (8.9596/2.2655) mem 68106MB [2022-12-19 12:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1020/1519] eta 0:08:21 lr 0.000031 time 0.9295 (1.0041) model_time 0.9293 (1.0033) loss 0.7743 (0.9663) grad_norm 8.6301 (8.9387/2.2652) mem 68106MB [2022-12-19 12:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1030/1519] eta 0:08:11 lr 0.000031 time 0.9309 (1.0042) model_time 0.9308 (1.0034) loss 1.0029 (0.9663) grad_norm 9.0692 (8.9477/2.2653) mem 68106MB [2022-12-19 12:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1040/1519] eta 0:08:00 lr 0.000031 time 0.9352 (1.0042) model_time 0.9351 (1.0034) loss 1.1409 (0.9667) grad_norm 10.4399 (8.9548/2.2404) mem 68106MB [2022-12-19 12:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1050/1519] eta 0:07:50 lr 0.000031 time 0.9249 (1.0041) model_time 0.9248 (1.0033) loss 1.0775 (0.9670) grad_norm 11.0121 (8.9869/2.2547) mem 68106MB [2022-12-19 12:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1060/1519] eta 0:07:40 lr 0.000031 time 0.9361 (1.0042) model_time 0.9359 (1.0034) loss 0.8556 (0.9664) grad_norm 20.6563 (9.0767/2.4812) mem 68106MB [2022-12-19 12:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1070/1519] eta 0:07:30 lr 0.000031 time 0.9410 (1.0041) model_time 0.9408 (1.0034) loss 0.9878 (0.9669) grad_norm 14.0810 (9.1022/2.4833) mem 68106MB [2022-12-19 12:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1080/1519] eta 0:07:20 lr 0.000031 time 0.9460 (1.0042) model_time 0.9458 (1.0034) loss 1.0382 (0.9677) grad_norm 18.2274 (9.1427/2.5320) mem 68106MB [2022-12-19 12:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1090/1519] eta 0:07:10 lr 0.000031 time 0.9305 (1.0042) model_time 0.9303 (1.0034) loss 0.8580 (0.9674) grad_norm 13.5658 (9.1615/2.5231) mem 68106MB [2022-12-19 12:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1100/1519] eta 0:07:00 lr 0.000031 time 0.9345 (1.0041) model_time 0.9344 (1.0034) loss 0.7973 (0.9666) grad_norm 5.8594 (9.1751/2.5098) mem 68106MB [2022-12-19 12:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1110/1519] eta 0:06:50 lr 0.000031 time 0.9280 (1.0043) model_time 0.9278 (1.0035) loss 1.3247 (0.9670) grad_norm 9.4489 (9.1354/2.4875) mem 68106MB [2022-12-19 12:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1120/1519] eta 0:06:40 lr 0.000031 time 0.9423 (1.0043) model_time 0.9421 (1.0036) loss 0.9676 (0.9676) grad_norm 9.2003 (9.1576/2.4710) mem 68106MB [2022-12-19 12:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1130/1519] eta 0:06:30 lr 0.000031 time 0.9358 (1.0043) model_time 0.9357 (1.0036) loss 0.7802 (0.9674) grad_norm 8.8324 (9.1391/2.4398) mem 68106MB [2022-12-19 12:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1140/1519] eta 0:06:20 lr 0.000031 time 0.9906 (1.0043) model_time 0.9905 (1.0036) loss 1.0251 (0.9681) grad_norm 10.3364 (9.2058/2.4961) mem 68106MB [2022-12-19 12:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1150/1519] eta 0:06:10 lr 0.000031 time 0.9503 (1.0044) model_time 0.9502 (1.0036) loss 0.7505 (0.9687) grad_norm 6.5905 (9.2155/2.4945) mem 68106MB [2022-12-19 12:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1160/1519] eta 0:06:00 lr 0.000031 time 0.9296 (1.0044) model_time 0.9294 (1.0037) loss 1.0539 (0.9691) grad_norm 8.0953 (9.2179/2.4932) mem 68106MB [2022-12-19 12:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1170/1519] eta 0:05:50 lr 0.000031 time 0.9303 (1.0045) model_time 0.9301 (1.0038) loss 1.2321 (0.9691) grad_norm 11.5829 (9.2270/2.5027) mem 68106MB [2022-12-19 12:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1180/1519] eta 0:05:40 lr 0.000031 time 0.9405 (1.0045) model_time 0.9404 (1.0038) loss 0.8115 (0.9693) grad_norm 8.3549 (9.2209/2.5023) mem 68106MB [2022-12-19 12:26:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1190/1519] eta 0:05:30 lr 0.000031 time 0.9323 (1.0044) model_time 0.9322 (1.0037) loss 0.7704 (0.9687) grad_norm 12.0505 (9.2253/2.5093) mem 68106MB [2022-12-19 12:26:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1200/1519] eta 0:05:20 lr 0.000031 time 0.9731 (1.0044) model_time 0.9730 (1.0037) loss 1.0585 (0.9688) grad_norm 9.7955 (9.2502/2.5349) mem 68106MB [2022-12-19 12:26:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1210/1519] eta 0:05:10 lr 0.000031 time 0.9346 (1.0044) model_time 0.9344 (1.0037) loss 0.8588 (0.9683) grad_norm 10.1660 (9.2725/2.5522) mem 68106MB [2022-12-19 12:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1220/1519] eta 0:05:00 lr 0.000031 time 0.9315 (1.0044) model_time 0.9313 (1.0037) loss 1.1333 (0.9687) grad_norm 8.7888 (9.2535/2.5505) mem 68106MB [2022-12-19 12:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1230/1519] eta 0:04:50 lr 0.000031 time 0.9368 (1.0044) model_time 0.9367 (1.0037) loss 0.8150 (0.9686) grad_norm 8.3198 (9.2768/2.5561) mem 68106MB [2022-12-19 12:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1240/1519] eta 0:04:40 lr 0.000031 time 0.9355 (1.0044) model_time 0.9354 (1.0037) loss 0.9663 (0.9687) grad_norm 9.0680 (9.2669/2.5618) mem 68106MB [2022-12-19 12:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1250/1519] eta 0:04:30 lr 0.000031 time 0.9393 (1.0044) model_time 0.9392 (1.0037) loss 0.9287 (0.9679) grad_norm 7.7130 (9.3661/2.7705) mem 68106MB [2022-12-19 12:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1260/1519] eta 0:04:20 lr 0.000031 time 0.9312 (1.0044) model_time 0.9311 (1.0037) loss 0.7394 (0.9678) grad_norm 9.6791 (9.3448/2.7649) mem 68106MB [2022-12-19 12:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1270/1519] eta 0:04:10 lr 0.000031 time 0.9311 (1.0044) model_time 0.9310 (1.0037) loss 0.8269 (0.9673) grad_norm 9.1679 (9.3218/2.7777) mem 68106MB [2022-12-19 12:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1280/1519] eta 0:04:00 lr 0.000031 time 0.9324 (1.0043) model_time 0.9323 (1.0036) loss 0.9113 (0.9673) grad_norm 8.1171 (9.3076/2.7797) mem 68106MB [2022-12-19 12:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1290/1519] eta 0:03:49 lr 0.000031 time 0.9395 (1.0044) model_time 0.9393 (1.0037) loss 0.7463 (0.9675) grad_norm 11.5647 (9.3002/2.7764) mem 68106MB [2022-12-19 12:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1300/1519] eta 0:03:39 lr 0.000031 time 0.9364 (1.0044) model_time 0.9363 (1.0037) loss 1.0182 (0.9671) grad_norm 11.9211 (9.3050/2.7793) mem 68106MB [2022-12-19 12:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1310/1519] eta 0:03:29 lr 0.000031 time 0.9319 (1.0044) model_time 0.9317 (1.0037) loss 1.0707 (0.9683) grad_norm 10.3574 (9.3220/2.7933) mem 68106MB [2022-12-19 12:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1320/1519] eta 0:03:19 lr 0.000031 time 0.9328 (1.0044) model_time 0.9327 (1.0037) loss 1.0760 (0.9686) grad_norm 9.0097 (9.2909/2.7910) mem 68106MB [2022-12-19 12:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1330/1519] eta 0:03:09 lr 0.000031 time 0.9357 (1.0044) model_time 0.9356 (1.0038) loss 0.6993 (0.9687) grad_norm 7.3921 (9.3136/2.7918) mem 68106MB [2022-12-19 12:28:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1340/1519] eta 0:02:59 lr 0.000031 time 0.9354 (1.0046) model_time 0.9353 (1.0039) loss 0.8452 (0.9683) grad_norm 8.9630 (9.2835/2.7670) mem 68106MB [2022-12-19 12:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1350/1519] eta 0:02:49 lr 0.000031 time 0.9331 (1.0045) model_time 0.9329 (1.0039) loss 0.9298 (0.9685) grad_norm 5.6760 (9.2724/2.7780) mem 68106MB [2022-12-19 12:28:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1360/1519] eta 0:02:39 lr 0.000031 time 0.9327 (1.0045) model_time 0.9325 (1.0039) loss 0.7077 (0.9681) grad_norm 6.3115 (9.2532/2.7772) mem 68106MB [2022-12-19 12:29:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1370/1519] eta 0:02:29 lr 0.000031 time 0.9417 (1.0045) model_time 0.9416 (1.0038) loss 0.9566 (0.9671) grad_norm 7.9712 (9.2781/2.7720) mem 68106MB [2022-12-19 12:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1380/1519] eta 0:02:19 lr 0.000031 time 0.9718 (1.0045) model_time 0.9716 (1.0038) loss 0.8662 (0.9669) grad_norm 6.5963 (9.2728/2.7485) mem 68106MB [2022-12-19 12:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1390/1519] eta 0:02:09 lr 0.000031 time 0.9315 (1.0045) model_time 0.9313 (1.0038) loss 0.8583 (0.9674) grad_norm 7.1404 (9.2219/2.7016) mem 68106MB [2022-12-19 12:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1400/1519] eta 0:01:59 lr 0.000031 time 0.9327 (1.0045) model_time 0.9326 (1.0038) loss 0.7633 (0.9676) grad_norm 6.0935 (9.2316/2.6839) mem 68106MB [2022-12-19 12:29:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1410/1519] eta 0:01:49 lr 0.000031 time 0.9368 (1.0045) model_time 0.9366 (1.0038) loss 0.7097 (0.9678) grad_norm 8.4568 (9.2412/2.6736) mem 68106MB [2022-12-19 12:29:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1420/1519] eta 0:01:39 lr 0.000031 time 0.9321 (1.0046) model_time 0.9320 (1.0039) loss 0.7282 (0.9676) grad_norm 8.0201 (9.2552/2.7722) mem 68106MB [2022-12-19 12:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1430/1519] eta 0:01:29 lr 0.000031 time 0.9294 (1.0046) model_time 0.9293 (1.0039) loss 1.3653 (0.9671) grad_norm 7.0108 (9.2947/2.8307) mem 68106MB [2022-12-19 12:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1440/1519] eta 0:01:19 lr 0.000031 time 0.9452 (1.0045) model_time 0.9451 (1.0039) loss 1.2377 (0.9675) grad_norm 7.7775 (9.3159/2.8327) mem 68106MB [2022-12-19 12:30:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1450/1519] eta 0:01:09 lr 0.000031 time 0.9395 (1.0045) model_time 0.9394 (1.0039) loss 0.9510 (0.9674) grad_norm 20.5765 (9.4127/2.9385) mem 68106MB [2022-12-19 12:30:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1460/1519] eta 0:00:59 lr 0.000031 time 0.9341 (1.0045) model_time 0.9339 (1.0039) loss 0.8704 (0.9670) grad_norm 8.0722 (9.4136/2.9540) mem 68106MB [2022-12-19 12:30:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1470/1519] eta 0:00:49 lr 0.000031 time 0.9309 (1.0045) model_time 0.9306 (1.0038) loss 0.8910 (0.9670) grad_norm 7.8855 (9.4378/2.9480) mem 68106MB [2022-12-19 12:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1480/1519] eta 0:00:39 lr 0.000031 time 0.9306 (1.0045) model_time 0.9305 (1.0039) loss 1.2831 (0.9675) grad_norm 9.3216 (9.4661/2.9444) mem 68106MB [2022-12-19 12:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1490/1519] eta 0:00:29 lr 0.000031 time 0.9397 (1.0045) model_time 0.9396 (1.0039) loss 0.9344 (0.9679) grad_norm 7.0230 (9.4431/2.8956) mem 68106MB [2022-12-19 12:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1500/1519] eta 0:00:19 lr 0.000031 time 0.9332 (1.0045) model_time 0.9331 (1.0039) loss 0.8269 (0.9685) grad_norm 5.8800 (9.3933/2.8860) mem 68106MB [2022-12-19 12:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [18/100][1510/1519] eta 0:00:09 lr 0.000031 time 0.9197 (1.0046) model_time 0.9195 (1.0040) loss 0.9893 (0.9684) grad_norm 7.4965 (9.4128/2.9205) mem 68106MB [2022-12-19 12:31:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 18 training takes 0:25:26 [2022-12-19 12:31:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_18.pth saving...... [2022-12-19 12:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_18.pth saved !!! [2022-12-19 12:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.678 (0.678) Loss 0.5729 (0.5729) Acc@1 88.542 (88.542) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 12:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.331) Loss 0.5701 (0.5450) Acc@1 90.625 (90.341) Acc@5 97.569 (98.043) Mem 68106MB [2022-12-19 12:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.314) Loss 0.5047 (0.5452) Acc@1 89.583 (89.997) Acc@5 98.958 (97.999) Mem 68106MB [2022-12-19 12:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.6596 (0.5521) Acc@1 86.806 (89.774) Acc@5 97.222 (97.928) Mem 68106MB [2022-12-19 12:32:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.306) Loss 0.5485 (0.5442) Acc@1 89.931 (89.795) Acc@5 97.917 (98.018) Mem 68106MB [2022-12-19 12:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.305) Loss 0.5686 (0.5407) Acc@1 86.458 (89.849) Acc@5 98.264 (98.094) Mem 68106MB [2022-12-19 12:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.304) Loss 0.5700 (0.5415) Acc@1 88.542 (89.817) Acc@5 97.917 (98.087) Mem 68106MB [2022-12-19 12:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.303) Loss 0.6132 (0.5446) Acc@1 89.583 (89.745) Acc@5 98.264 (98.068) Mem 68106MB [2022-12-19 12:32:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.302) Loss 0.4927 (0.5435) Acc@1 90.972 (89.781) Acc@5 97.917 (98.088) Mem 68106MB [2022-12-19 12:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:18] * Acc@1 89.755 Acc@5 98.109 [2022-12-19 12:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 89.8% [2022-12-19 12:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 12:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 12:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 89.76% [2022-12-19 12:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][0/1519] eta 0:36:53 lr 0.000031 time 1.4571 (1.4571) model_time 0.9768 (0.9768) loss 0.7060 (0.7060) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 12:32:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][10/1519] eta 0:26:19 lr 0.000031 time 0.9280 (1.0465) model_time 0.9279 (1.0024) loss 0.7332 (0.9470) grad_norm 8.4185 (9.8158/3.5818) mem 68106MB [2022-12-19 12:33:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][20/1519] eta 0:25:36 lr 0.000031 time 0.9233 (1.0250) model_time 0.9232 (1.0017) loss 1.1292 (0.9318) grad_norm 8.9197 (10.5000/3.2050) mem 68106MB [2022-12-19 12:33:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][30/1519] eta 0:25:18 lr 0.000031 time 0.9326 (1.0198) model_time 0.9324 (1.0039) loss 1.0256 (0.9804) grad_norm 8.6171 (9.7753/2.9219) mem 68106MB [2022-12-19 12:33:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][40/1519] eta 0:25:00 lr 0.000031 time 0.9308 (1.0149) model_time 0.9306 (1.0027) loss 1.1415 (0.9734) grad_norm 9.3645 (9.6052/2.8455) mem 68106MB [2022-12-19 12:33:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][50/1519] eta 0:24:47 lr 0.000031 time 0.9362 (1.0128) model_time 0.9360 (1.0029) loss 0.6855 (0.9563) grad_norm 8.0617 (9.4555/2.6064) mem 68106MB [2022-12-19 12:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][60/1519] eta 0:24:34 lr 0.000031 time 0.9249 (1.0105) model_time 0.9247 (1.0022) loss 1.7994 (0.9733) grad_norm 6.8990 (9.2034/2.4629) mem 68106MB [2022-12-19 12:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][70/1519] eta 0:24:21 lr 0.000031 time 0.9306 (1.0086) model_time 0.9304 (1.0015) loss 0.9226 (0.9683) grad_norm 10.0765 (9.2863/2.4633) mem 68106MB [2022-12-19 12:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][80/1519] eta 0:24:11 lr 0.000031 time 1.0065 (1.0090) model_time 1.0064 (1.0027) loss 0.7569 (0.9763) grad_norm 7.2745 (9.3750/2.6812) mem 68106MB [2022-12-19 12:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][90/1519] eta 0:24:00 lr 0.000031 time 0.9443 (1.0083) model_time 0.9441 (1.0026) loss 0.7900 (0.9695) grad_norm 13.5085 (9.2171/2.7305) mem 68106MB [2022-12-19 12:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][100/1519] eta 0:23:49 lr 0.000031 time 0.9442 (1.0075) model_time 0.9440 (1.0023) loss 0.7926 (0.9587) grad_norm 7.2081 (9.0674/2.6549) mem 68106MB [2022-12-19 12:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][110/1519] eta 0:23:38 lr 0.000031 time 0.9242 (1.0067) model_time 0.9241 (1.0020) loss 0.9638 (0.9560) grad_norm 9.0907 (8.9308/2.5904) mem 68106MB [2022-12-19 12:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][120/1519] eta 0:23:27 lr 0.000031 time 0.9307 (1.0060) model_time 0.9305 (1.0017) loss 1.0180 (0.9493) grad_norm 6.3024 (8.9809/2.9059) mem 68106MB [2022-12-19 12:34:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][130/1519] eta 0:23:18 lr 0.000031 time 0.9349 (1.0065) model_time 0.9347 (1.0024) loss 1.2012 (0.9475) grad_norm 9.5468 (9.0245/2.8281) mem 68106MB [2022-12-19 12:35:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][140/1519] eta 0:23:07 lr 0.000031 time 0.9412 (1.0061) model_time 0.9408 (1.0023) loss 0.8192 (0.9544) grad_norm 13.9938 (9.1507/2.8992) mem 68106MB [2022-12-19 12:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][150/1519] eta 0:22:57 lr 0.000031 time 0.9637 (1.0065) model_time 0.9635 (1.0029) loss 1.1599 (0.9512) grad_norm 10.8112 (9.0770/2.8401) mem 68106MB [2022-12-19 12:35:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][160/1519] eta 0:22:48 lr 0.000031 time 0.9286 (1.0068) model_time 0.9281 (1.0034) loss 1.1697 (0.9484) grad_norm 6.7734 (9.3259/3.0501) mem 68106MB [2022-12-19 12:35:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][170/1519] eta 0:22:37 lr 0.000031 time 0.9356 (1.0065) model_time 0.9354 (1.0033) loss 1.0067 (0.9440) grad_norm 9.5601 (9.2451/2.9853) mem 68106MB [2022-12-19 12:35:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][180/1519] eta 0:22:27 lr 0.000031 time 0.9324 (1.0064) model_time 0.9323 (1.0033) loss 1.3418 (0.9438) grad_norm 8.9963 (9.1910/2.9308) mem 68106MB [2022-12-19 12:35:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][190/1519] eta 0:22:16 lr 0.000031 time 0.9261 (1.0059) model_time 0.9259 (1.0030) loss 0.9536 (0.9463) grad_norm 7.1653 (9.1414/2.8838) mem 68106MB [2022-12-19 12:36:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][200/1519] eta 0:22:06 lr 0.000031 time 0.9255 (1.0059) model_time 0.9253 (1.0031) loss 0.9024 (0.9482) grad_norm 8.7353 (9.1085/2.8657) mem 68106MB [2022-12-19 12:36:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][210/1519] eta 0:21:56 lr 0.000031 time 0.9195 (1.0057) model_time 0.9194 (1.0030) loss 0.8984 (0.9474) grad_norm 8.0891 (9.0363/2.8186) mem 68106MB [2022-12-19 12:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][220/1519] eta 0:21:46 lr 0.000031 time 0.9272 (1.0059) model_time 0.9270 (1.0033) loss 1.0705 (0.9523) grad_norm 6.9658 (9.0204/2.7807) mem 68106MB [2022-12-19 12:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][230/1519] eta 0:21:36 lr 0.000031 time 0.9279 (1.0056) model_time 0.9277 (1.0031) loss 0.8287 (0.9543) grad_norm 8.3257 (9.1005/2.8622) mem 68106MB [2022-12-19 12:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][240/1519] eta 0:21:26 lr 0.000031 time 0.9381 (1.0055) model_time 0.9380 (1.0032) loss 1.0734 (0.9557) grad_norm 10.3274 (9.1097/2.8259) mem 68106MB [2022-12-19 12:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][250/1519] eta 0:21:15 lr 0.000031 time 0.9268 (1.0053) model_time 0.9266 (1.0030) loss 0.9754 (0.9537) grad_norm 5.6122 (9.0870/2.7911) mem 68106MB [2022-12-19 12:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][260/1519] eta 0:21:05 lr 0.000031 time 1.0065 (1.0054) model_time 1.0063 (1.0032) loss 0.7679 (0.9536) grad_norm 8.5449 (9.0453/2.7573) mem 68106MB [2022-12-19 12:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][270/1519] eta 0:20:55 lr 0.000031 time 0.9349 (1.0051) model_time 0.9347 (1.0029) loss 0.9287 (0.9528) grad_norm 7.9084 (8.9814/2.7277) mem 68106MB [2022-12-19 12:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][280/1519] eta 0:20:45 lr 0.000031 time 0.9309 (1.0049) model_time 0.9307 (1.0028) loss 1.0247 (0.9580) grad_norm 7.8296 (8.9177/2.7047) mem 68106MB [2022-12-19 12:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][290/1519] eta 0:20:34 lr 0.000031 time 0.9303 (1.0049) model_time 0.9301 (1.0028) loss 1.0353 (0.9567) grad_norm 8.3102 (8.8612/2.6765) mem 68106MB [2022-12-19 12:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][300/1519] eta 0:20:25 lr 0.000031 time 0.9259 (1.0050) model_time 0.9255 (1.0030) loss 1.3564 (0.9566) grad_norm 10.9306 (8.8935/2.6761) mem 68106MB [2022-12-19 12:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][310/1519] eta 0:20:15 lr 0.000031 time 0.9248 (1.0050) model_time 0.9247 (1.0031) loss 1.0890 (0.9554) grad_norm 6.7505 (8.8547/2.6468) mem 68106MB [2022-12-19 12:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][320/1519] eta 0:20:04 lr 0.000031 time 0.9239 (1.0048) model_time 0.9237 (1.0029) loss 1.2081 (0.9551) grad_norm 6.8290 (8.8517/2.6294) mem 68106MB [2022-12-19 12:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][330/1519] eta 0:19:54 lr 0.000031 time 0.9318 (1.0046) model_time 0.9315 (1.0028) loss 1.0277 (0.9558) grad_norm 11.0349 (8.8455/2.6017) mem 68106MB [2022-12-19 12:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][340/1519] eta 0:19:44 lr 0.000031 time 0.9375 (1.0047) model_time 0.9373 (1.0029) loss 0.9216 (0.9573) grad_norm 6.9857 (8.8044/2.5897) mem 68106MB [2022-12-19 12:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][350/1519] eta 0:19:34 lr 0.000031 time 0.9298 (1.0046) model_time 0.9296 (1.0028) loss 1.1255 (0.9559) grad_norm 8.2946 (8.8094/2.5564) mem 68106MB [2022-12-19 12:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][360/1519] eta 0:19:24 lr 0.000031 time 0.9259 (1.0046) model_time 0.9257 (1.0029) loss 1.2384 (0.9589) grad_norm 8.2859 (8.8369/2.5340) mem 68106MB [2022-12-19 12:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][370/1519] eta 0:19:14 lr 0.000031 time 0.9398 (1.0045) model_time 0.9396 (1.0028) loss 0.9955 (0.9588) grad_norm 7.3942 (8.8399/2.5437) mem 68106MB [2022-12-19 12:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][380/1519] eta 0:19:04 lr 0.000031 time 0.9334 (1.0047) model_time 0.9332 (1.0030) loss 0.9719 (0.9578) grad_norm 7.4879 (8.8276/2.5234) mem 68106MB [2022-12-19 12:39:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][390/1519] eta 0:18:54 lr 0.000031 time 0.9223 (1.0046) model_time 0.9221 (1.0030) loss 1.2355 (0.9606) grad_norm 9.0613 (8.8180/2.5021) mem 68106MB [2022-12-19 12:39:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][400/1519] eta 0:18:44 lr 0.000031 time 0.9128 (1.0047) model_time 0.9126 (1.0031) loss 0.7261 (0.9591) grad_norm 8.1131 (8.8001/2.4817) mem 68106MB [2022-12-19 12:39:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][410/1519] eta 0:18:34 lr 0.000031 time 0.9329 (1.0046) model_time 0.9327 (1.0030) loss 1.2755 (0.9594) grad_norm 7.7081 (8.7899/2.4812) mem 68106MB [2022-12-19 12:39:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][420/1519] eta 0:18:24 lr 0.000031 time 0.9315 (1.0046) model_time 0.9313 (1.0031) loss 0.7118 (0.9580) grad_norm 11.5443 (8.7781/2.4679) mem 68106MB [2022-12-19 12:40:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][430/1519] eta 0:18:13 lr 0.000031 time 0.9240 (1.0045) model_time 0.9239 (1.0030) loss 0.7879 (0.9570) grad_norm 7.7070 (8.7845/2.4844) mem 68106MB [2022-12-19 12:40:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][440/1519] eta 0:18:04 lr 0.000031 time 1.0963 (1.0048) model_time 1.0962 (1.0033) loss 0.7281 (0.9569) grad_norm 7.8331 (8.7741/2.4669) mem 68106MB [2022-12-19 12:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][450/1519] eta 0:17:54 lr 0.000031 time 0.9217 (1.0048) model_time 0.9216 (1.0033) loss 0.8037 (0.9570) grad_norm 10.2991 (8.7559/2.4535) mem 68106MB [2022-12-19 12:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][460/1519] eta 0:17:44 lr 0.000031 time 0.9240 (1.0048) model_time 0.9239 (1.0033) loss 0.7469 (0.9564) grad_norm 10.4991 (8.7765/2.4367) mem 68106MB [2022-12-19 12:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][470/1519] eta 0:17:34 lr 0.000031 time 0.9146 (1.0053) model_time 0.9145 (1.0039) loss 0.9754 (0.9564) grad_norm 10.1766 (8.7743/2.4204) mem 68106MB [2022-12-19 12:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][480/1519] eta 0:17:25 lr 0.000031 time 0.9532 (1.0058) model_time 0.9530 (1.0044) loss 0.9967 (0.9572) grad_norm 9.3527 (8.8191/2.4460) mem 68106MB [2022-12-19 12:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][490/1519] eta 0:17:14 lr 0.000031 time 0.9296 (1.0056) model_time 0.9295 (1.0043) loss 0.8301 (0.9571) grad_norm 6.5746 (8.8171/2.4461) mem 68106MB [2022-12-19 12:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][500/1519] eta 0:17:04 lr 0.000031 time 0.9056 (1.0055) model_time 0.9054 (1.0042) loss 1.0786 (0.9573) grad_norm 8.0255 (8.8221/2.4544) mem 68106MB [2022-12-19 12:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][510/1519] eta 0:16:54 lr 0.000031 time 0.9363 (1.0054) model_time 0.9361 (1.0041) loss 0.7299 (0.9567) grad_norm 5.7784 (8.8007/2.4445) mem 68106MB [2022-12-19 12:41:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][520/1519] eta 0:16:44 lr 0.000031 time 0.9068 (1.0054) model_time 0.9062 (1.0041) loss 0.8851 (0.9550) grad_norm 10.9870 (8.8011/2.4279) mem 68106MB [2022-12-19 12:41:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][530/1519] eta 0:16:34 lr 0.000031 time 0.9363 (1.0056) model_time 0.9362 (1.0043) loss 1.1003 (0.9559) grad_norm 8.6478 (8.8085/2.4270) mem 68106MB [2022-12-19 12:41:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][540/1519] eta 0:16:24 lr 0.000031 time 0.9539 (1.0056) model_time 0.9538 (1.0043) loss 0.8371 (0.9563) grad_norm 7.7779 (8.7926/2.4170) mem 68106MB [2022-12-19 12:42:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][550/1519] eta 0:16:14 lr 0.000031 time 0.9337 (1.0054) model_time 0.9335 (1.0042) loss 0.8132 (0.9555) grad_norm 7.2501 (8.7976/2.4030) mem 68106MB [2022-12-19 12:42:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][560/1519] eta 0:16:04 lr 0.000031 time 0.9282 (1.0053) model_time 0.9281 (1.0041) loss 0.8620 (0.9565) grad_norm 9.4402 (8.7973/2.3826) mem 68106MB [2022-12-19 12:42:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][570/1519] eta 0:15:53 lr 0.000031 time 0.9281 (1.0052) model_time 0.9280 (1.0040) loss 0.9850 (0.9569) grad_norm 14.4167 (8.8224/2.3920) mem 68106MB [2022-12-19 12:42:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][580/1519] eta 0:15:43 lr 0.000031 time 0.9331 (1.0052) model_time 0.9330 (1.0040) loss 1.2414 (0.9577) grad_norm 6.5008 (8.8103/2.3950) mem 68106MB [2022-12-19 12:42:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][590/1519] eta 0:15:33 lr 0.000031 time 0.9308 (1.0051) model_time 0.9306 (1.0039) loss 0.8382 (0.9577) grad_norm 12.8527 (8.8281/2.4161) mem 68106MB [2022-12-19 12:42:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][600/1519] eta 0:15:23 lr 0.000031 time 0.9289 (1.0050) model_time 0.9287 (1.0038) loss 1.1687 (0.9587) grad_norm 9.2484 (8.8388/2.4057) mem 68106MB [2022-12-19 12:43:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][610/1519] eta 0:15:13 lr 0.000031 time 0.9338 (1.0049) model_time 0.9337 (1.0037) loss 0.7333 (0.9600) grad_norm 10.8578 (8.8505/2.3857) mem 68106MB [2022-12-19 12:43:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][620/1519] eta 0:15:03 lr 0.000031 time 0.9296 (1.0050) model_time 0.9294 (1.0039) loss 0.7780 (0.9589) grad_norm 9.8157 (8.8419/2.3863) mem 68106MB [2022-12-19 12:43:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][630/1519] eta 0:14:53 lr 0.000031 time 0.9263 (1.0050) model_time 0.9262 (1.0039) loss 0.9071 (0.9580) grad_norm 11.1294 (8.8441/2.3884) mem 68106MB [2022-12-19 12:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][640/1519] eta 0:14:43 lr 0.000031 time 0.9307 (1.0049) model_time 0.9305 (1.0037) loss 0.7898 (0.9590) grad_norm 7.4937 (8.8260/2.3697) mem 68106MB [2022-12-19 12:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][650/1519] eta 0:14:33 lr 0.000031 time 1.1800 (1.0053) model_time 1.1799 (1.0041) loss 1.1110 (0.9579) grad_norm 7.4138 (8.8201/2.3685) mem 68106MB [2022-12-19 12:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][660/1519] eta 0:14:23 lr 0.000031 time 0.9347 (1.0052) model_time 0.9345 (1.0041) loss 0.8672 (0.9570) grad_norm 8.1251 (8.8771/2.4378) mem 68106MB [2022-12-19 12:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][670/1519] eta 0:14:13 lr 0.000031 time 0.9328 (1.0052) model_time 0.9326 (1.0041) loss 0.9642 (0.9570) grad_norm 14.4213 (8.9755/2.9116) mem 68106MB [2022-12-19 12:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][680/1519] eta 0:14:03 lr 0.000031 time 0.9311 (1.0051) model_time 0.9309 (1.0040) loss 1.1017 (0.9569) grad_norm 7.6178 (8.9350/2.8755) mem 68106MB [2022-12-19 12:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][690/1519] eta 0:13:53 lr 0.000031 time 0.9242 (1.0052) model_time 0.9240 (1.0041) loss 0.8031 (0.9575) grad_norm 6.9217 (8.9205/2.8613) mem 68106MB [2022-12-19 12:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][700/1519] eta 0:13:43 lr 0.000031 time 0.9296 (1.0051) model_time 0.9295 (1.0040) loss 0.9486 (0.9569) grad_norm 7.2082 (8.9439/2.8695) mem 68106MB [2022-12-19 12:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][710/1519] eta 0:13:33 lr 0.000031 time 0.9256 (1.0053) model_time 0.9254 (1.0042) loss 0.8445 (0.9571) grad_norm 12.9585 (9.0062/2.9126) mem 68106MB [2022-12-19 12:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][720/1519] eta 0:13:23 lr 0.000031 time 0.9210 (1.0052) model_time 0.9209 (1.0041) loss 1.4868 (0.9572) grad_norm 11.3841 (9.0079/2.8410) mem 68106MB [2022-12-19 12:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][730/1519] eta 0:13:13 lr 0.000031 time 0.9358 (1.0051) model_time 0.9356 (1.0040) loss 0.9690 (0.9573) grad_norm 13.9565 (9.0358/2.8625) mem 68106MB [2022-12-19 12:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][740/1519] eta 0:13:02 lr 0.000031 time 0.9205 (1.0050) model_time 0.9204 (1.0040) loss 0.8469 (0.9575) grad_norm 9.0311 (8.9873/2.8301) mem 68106MB [2022-12-19 12:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][750/1519] eta 0:12:52 lr 0.000031 time 0.9314 (1.0050) model_time 0.9312 (1.0039) loss 0.7505 (0.9566) grad_norm 8.6281 (9.0122/2.8273) mem 68106MB [2022-12-19 12:45:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][760/1519] eta 0:12:42 lr 0.000031 time 0.9310 (1.0050) model_time 0.9308 (1.0040) loss 0.7299 (0.9569) grad_norm 7.1903 (8.9636/2.7524) mem 68106MB [2022-12-19 12:45:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][770/1519] eta 0:12:32 lr 0.000031 time 0.9874 (1.0052) model_time 0.9872 (1.0042) loss 0.8513 (0.9565) grad_norm 8.7943 (8.9647/2.7551) mem 68106MB [2022-12-19 12:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][780/1519] eta 0:12:22 lr 0.000031 time 0.9250 (1.0052) model_time 0.9248 (1.0042) loss 0.8640 (0.9566) grad_norm 6.5430 (8.9574/2.7550) mem 68106MB [2022-12-19 12:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][790/1519] eta 0:12:12 lr 0.000031 time 0.9136 (1.0053) model_time 0.9135 (1.0043) loss 1.0656 (0.9568) grad_norm 7.8705 (8.9657/2.7497) mem 68106MB [2022-12-19 12:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][800/1519] eta 0:12:02 lr 0.000031 time 0.9289 (1.0053) model_time 0.9288 (1.0043) loss 0.8978 (0.9565) grad_norm 7.7018 (8.9603/2.7424) mem 68106MB [2022-12-19 12:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][810/1519] eta 0:11:52 lr 0.000031 time 0.9301 (1.0053) model_time 0.9300 (1.0043) loss 0.7314 (0.9571) grad_norm 7.7615 (8.9891/2.7513) mem 68106MB [2022-12-19 12:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][820/1519] eta 0:11:42 lr 0.000031 time 0.9343 (1.0052) model_time 0.9336 (1.0042) loss 1.5302 (0.9580) grad_norm 9.1095 (8.9881/2.7488) mem 68106MB [2022-12-19 12:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][830/1519] eta 0:11:32 lr 0.000031 time 0.9492 (1.0051) model_time 0.9491 (1.0042) loss 1.0597 (0.9571) grad_norm 7.7445 (8.9636/2.7023) mem 68106MB [2022-12-19 12:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][840/1519] eta 0:11:22 lr 0.000031 time 0.9186 (1.0053) model_time 0.9185 (1.0043) loss 1.1932 (0.9578) grad_norm 7.5627 (8.9919/2.7285) mem 68106MB [2022-12-19 12:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][850/1519] eta 0:11:12 lr 0.000031 time 0.9294 (1.0052) model_time 0.9293 (1.0042) loss 1.0309 (0.9566) grad_norm 19.0237 (9.0310/2.7925) mem 68106MB [2022-12-19 12:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][860/1519] eta 0:11:02 lr 0.000031 time 0.9233 (1.0051) model_time 0.9231 (1.0042) loss 0.9429 (0.9558) grad_norm 11.3292 (9.0808/2.8034) mem 68106MB [2022-12-19 12:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][870/1519] eta 0:10:52 lr 0.000031 time 0.9202 (1.0052) model_time 0.9201 (1.0042) loss 1.0263 (0.9554) grad_norm 5.6576 (9.0761/2.8116) mem 68106MB [2022-12-19 12:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][880/1519] eta 0:10:42 lr 0.000031 time 0.9272 (1.0051) model_time 0.9270 (1.0042) loss 0.8849 (0.9557) grad_norm 7.7531 (9.1034/2.8054) mem 68106MB [2022-12-19 12:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][890/1519] eta 0:10:32 lr 0.000031 time 0.9316 (1.0051) model_time 0.9314 (1.0042) loss 0.7484 (0.9547) grad_norm 8.2442 (9.1741/2.8582) mem 68106MB [2022-12-19 12:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][900/1519] eta 0:10:22 lr 0.000031 time 0.9221 (1.0050) model_time 0.9220 (1.0041) loss 1.1107 (0.9555) grad_norm 7.2992 (9.1631/2.8474) mem 68106MB [2022-12-19 12:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][910/1519] eta 0:10:12 lr 0.000031 time 0.9229 (1.0050) model_time 0.9228 (1.0040) loss 1.0602 (0.9548) grad_norm 6.7499 (9.1903/2.8497) mem 68106MB [2022-12-19 12:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][920/1519] eta 0:10:01 lr 0.000031 time 0.9333 (1.0049) model_time 0.9331 (1.0040) loss 1.1124 (0.9550) grad_norm 8.1690 (9.2337/2.9035) mem 68106MB [2022-12-19 12:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][930/1519] eta 0:09:51 lr 0.000031 time 0.9238 (1.0049) model_time 0.9237 (1.0040) loss 1.4303 (0.9547) grad_norm 8.9245 (9.2568/2.9008) mem 68106MB [2022-12-19 12:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][940/1519] eta 0:09:41 lr 0.000031 time 0.9250 (1.0050) model_time 0.9248 (1.0041) loss 1.1079 (0.9548) grad_norm 8.5518 (9.3167/2.9195) mem 68106MB [2022-12-19 12:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][950/1519] eta 0:09:31 lr 0.000031 time 0.9229 (1.0049) model_time 0.9227 (1.0040) loss 1.3165 (0.9557) grad_norm 10.3014 (9.3491/2.9470) mem 68106MB [2022-12-19 12:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][960/1519] eta 0:09:21 lr 0.000031 time 0.9246 (1.0049) model_time 0.9244 (1.0040) loss 0.9710 (0.9555) grad_norm 9.0275 (9.3511/2.9462) mem 68106MB [2022-12-19 12:49:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][970/1519] eta 0:09:11 lr 0.000031 time 0.9310 (1.0049) model_time 0.9308 (1.0040) loss 0.9142 (0.9555) grad_norm 8.2151 (9.3724/2.9308) mem 68106MB [2022-12-19 12:49:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][980/1519] eta 0:09:01 lr 0.000031 time 0.9478 (1.0049) model_time 0.9476 (1.0040) loss 0.8188 (0.9556) grad_norm 5.9760 (9.3831/3.0041) mem 68106MB [2022-12-19 12:49:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][990/1519] eta 0:08:51 lr 0.000031 time 0.9345 (1.0049) model_time 0.9343 (1.0040) loss 0.9305 (0.9555) grad_norm 7.6378 (9.4149/3.0342) mem 68106MB [2022-12-19 12:49:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1000/1519] eta 0:08:41 lr 0.000031 time 0.9310 (1.0049) model_time 0.9309 (1.0040) loss 1.2511 (0.9558) grad_norm 6.8433 (9.3955/3.0427) mem 68106MB [2022-12-19 12:49:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1010/1519] eta 0:08:31 lr 0.000031 time 0.9328 (1.0049) model_time 0.9326 (1.0040) loss 0.9660 (0.9557) grad_norm 8.8594 (9.3929/3.0322) mem 68106MB [2022-12-19 12:49:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1020/1519] eta 0:08:21 lr 0.000031 time 0.9237 (1.0050) model_time 0.9235 (1.0041) loss 0.8388 (0.9556) grad_norm 9.2440 (9.3887/3.0269) mem 68106MB [2022-12-19 12:50:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1030/1519] eta 0:08:11 lr 0.000031 time 0.9351 (1.0050) model_time 0.9349 (1.0041) loss 1.0710 (0.9556) grad_norm 6.5121 (9.3710/3.0185) mem 68106MB [2022-12-19 12:50:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1040/1519] eta 0:08:01 lr 0.000031 time 0.9309 (1.0050) model_time 0.9307 (1.0042) loss 1.1546 (0.9555) grad_norm 6.2127 (9.3527/3.0246) mem 68106MB [2022-12-19 12:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1050/1519] eta 0:07:51 lr 0.000031 time 0.9257 (1.0050) model_time 0.9255 (1.0041) loss 1.0079 (0.9559) grad_norm 10.4092 (9.3628/3.0199) mem 68106MB [2022-12-19 12:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1060/1519] eta 0:07:41 lr 0.000031 time 0.9269 (1.0051) model_time 0.9268 (1.0042) loss 1.0019 (0.9559) grad_norm 10.4632 (9.3713/3.0346) mem 68106MB [2022-12-19 12:50:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1070/1519] eta 0:07:31 lr 0.000031 time 0.9299 (1.0050) model_time 0.9297 (1.0042) loss 1.2671 (0.9559) grad_norm 7.3294 (9.3906/3.0438) mem 68106MB [2022-12-19 12:50:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1080/1519] eta 0:07:21 lr 0.000031 time 0.9323 (1.0049) model_time 0.9319 (1.0041) loss 0.7801 (0.9557) grad_norm 8.1363 (9.3636/3.0278) mem 68106MB [2022-12-19 12:51:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1090/1519] eta 0:07:11 lr 0.000031 time 0.9283 (1.0050) model_time 0.9281 (1.0041) loss 0.8655 (0.9558) grad_norm 10.0898 (9.3751/3.0163) mem 68106MB [2022-12-19 12:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1100/1519] eta 0:07:01 lr 0.000031 time 0.9287 (1.0051) model_time 0.9285 (1.0043) loss 0.7694 (0.9558) grad_norm 8.1020 (9.3776/3.0143) mem 68106MB [2022-12-19 12:51:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1110/1519] eta 0:06:51 lr 0.000031 time 0.9349 (1.0053) model_time 0.9347 (1.0045) loss 0.8252 (0.9563) grad_norm 6.6300 (9.3785/3.0105) mem 68106MB [2022-12-19 12:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1120/1519] eta 0:06:41 lr 0.000031 time 0.9348 (1.0052) model_time 0.9347 (1.0044) loss 0.7988 (0.9561) grad_norm 6.5374 (9.3661/3.0154) mem 68106MB [2022-12-19 12:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1130/1519] eta 0:06:31 lr 0.000031 time 0.9307 (1.0052) model_time 0.9306 (1.0043) loss 1.1724 (0.9561) grad_norm 9.9648 (9.3656/3.0017) mem 68106MB [2022-12-19 12:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1140/1519] eta 0:06:20 lr 0.000031 time 0.9292 (1.0051) model_time 0.9290 (1.0043) loss 0.8347 (0.9565) grad_norm 6.6719 (9.3526/3.0030) mem 68106MB [2022-12-19 12:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1150/1519] eta 0:06:10 lr 0.000031 time 0.9277 (1.0053) model_time 0.9276 (1.0045) loss 0.8135 (0.9562) grad_norm 12.1073 (9.3482/3.0137) mem 68106MB [2022-12-19 12:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1160/1519] eta 0:06:00 lr 0.000031 time 0.9354 (1.0052) model_time 0.9352 (1.0044) loss 1.0206 (0.9569) grad_norm 6.8845 (9.3184/3.0303) mem 68106MB [2022-12-19 12:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1170/1519] eta 0:05:50 lr 0.000031 time 0.9236 (1.0052) model_time 0.9235 (1.0044) loss 1.2816 (0.9574) grad_norm 9.1290 (9.3141/3.0269) mem 68106MB [2022-12-19 12:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1180/1519] eta 0:05:40 lr 0.000031 time 0.9303 (1.0052) model_time 0.9301 (1.0044) loss 0.7686 (0.9565) grad_norm 7.8355 (9.3228/3.0135) mem 68106MB [2022-12-19 12:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1190/1519] eta 0:05:30 lr 0.000031 time 0.9290 (1.0052) model_time 0.9288 (1.0044) loss 1.0884 (0.9574) grad_norm 7.5925 (9.3021/2.9885) mem 68106MB [2022-12-19 12:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1200/1519] eta 0:05:20 lr 0.000031 time 0.9176 (1.0052) model_time 0.9174 (1.0044) loss 1.2313 (0.9580) grad_norm 6.5401 (9.3055/3.0039) mem 68106MB [2022-12-19 12:53:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1210/1519] eta 0:05:10 lr 0.000031 time 0.9300 (1.0051) model_time 0.9299 (1.0043) loss 1.0133 (0.9585) grad_norm 7.1624 (9.2739/3.0003) mem 68106MB [2022-12-19 12:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1220/1519] eta 0:05:00 lr 0.000031 time 0.9316 (1.0051) model_time 0.9313 (1.0043) loss 1.0140 (0.9585) grad_norm 10.0010 (9.2484/2.9706) mem 68106MB [2022-12-19 12:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1230/1519] eta 0:04:50 lr 0.000031 time 0.9386 (1.0051) model_time 0.9384 (1.0043) loss 1.1822 (0.9581) grad_norm 8.0442 (9.2472/2.9690) mem 68106MB [2022-12-19 12:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1240/1519] eta 0:04:40 lr 0.000031 time 0.9700 (1.0050) model_time 0.9698 (1.0042) loss 0.7494 (0.9584) grad_norm 7.8881 (9.2713/2.9812) mem 68106MB [2022-12-19 12:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1250/1519] eta 0:04:30 lr 0.000031 time 0.9270 (1.0051) model_time 0.9268 (1.0043) loss 1.2190 (0.9588) grad_norm 8.5759 (9.2801/2.9853) mem 68106MB [2022-12-19 12:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1260/1519] eta 0:04:20 lr 0.000031 time 0.9192 (1.0050) model_time 0.9191 (1.0042) loss 0.8236 (0.9582) grad_norm 6.2027 (9.2171/2.9419) mem 68106MB [2022-12-19 12:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1270/1519] eta 0:04:10 lr 0.000031 time 0.9260 (1.0050) model_time 0.9258 (1.0043) loss 0.7285 (0.9588) grad_norm 10.6953 (9.1046/2.4717) mem 68106MB [2022-12-19 12:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1280/1519] eta 0:04:00 lr 0.000031 time 0.9321 (1.0051) model_time 0.9320 (1.0044) loss 0.9185 (0.9584) grad_norm 7.8066 (9.1019/2.4706) mem 68106MB [2022-12-19 12:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1290/1519] eta 0:03:50 lr 0.000031 time 0.9225 (1.0051) model_time 0.9223 (1.0043) loss 0.8671 (0.9594) grad_norm 8.5137 (9.1562/2.4647) mem 68106MB [2022-12-19 12:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1300/1519] eta 0:03:40 lr 0.000031 time 0.9307 (1.0050) model_time 0.9305 (1.0043) loss 0.8805 (0.9594) grad_norm 9.3663 (9.1380/2.4569) mem 68106MB [2022-12-19 12:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1310/1519] eta 0:03:30 lr 0.000031 time 0.9664 (1.0050) model_time 0.9663 (1.0043) loss 1.0305 (0.9594) grad_norm 7.5574 (9.1136/2.4219) mem 68106MB [2022-12-19 12:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1320/1519] eta 0:03:19 lr 0.000031 time 0.9193 (1.0050) model_time 0.9192 (1.0042) loss 0.9813 (0.9591) grad_norm 6.7807 (9.1066/2.4286) mem 68106MB [2022-12-19 12:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1330/1519] eta 0:03:09 lr 0.000031 time 0.9243 (1.0052) model_time 0.9241 (1.0044) loss 0.7226 (0.9587) grad_norm 5.9179 (9.0512/2.4055) mem 68106MB [2022-12-19 12:55:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1340/1519] eta 0:02:59 lr 0.000031 time 0.9418 (1.0051) model_time 0.9416 (1.0044) loss 0.9646 (0.9582) grad_norm 7.8187 (9.0902/2.4244) mem 68106MB [2022-12-19 12:55:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1350/1519] eta 0:02:49 lr 0.000031 time 0.9283 (1.0051) model_time 0.9282 (1.0043) loss 1.2960 (0.9587) grad_norm 10.0568 (9.0772/2.4291) mem 68106MB [2022-12-19 12:55:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1360/1519] eta 0:02:39 lr 0.000031 time 0.9276 (1.0050) model_time 0.9275 (1.0043) loss 0.7464 (0.9588) grad_norm 8.8862 (9.0631/2.4179) mem 68106MB [2022-12-19 12:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1370/1519] eta 0:02:29 lr 0.000031 time 0.9208 (1.0049) model_time 0.9206 (1.0042) loss 1.1346 (0.9589) grad_norm 6.5775 (9.0648/2.4199) mem 68106MB [2022-12-19 12:55:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1380/1519] eta 0:02:19 lr 0.000031 time 0.9220 (1.0050) model_time 0.9219 (1.0041) loss 0.8175 (0.9592) grad_norm 9.2797 (9.0863/2.4170) mem 68106MB [2022-12-19 12:56:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1390/1519] eta 0:02:09 lr 0.000031 time 0.9196 (1.0050) model_time 0.9194 (1.0041) loss 1.0654 (0.9594) grad_norm 5.5808 (9.0852/2.4276) mem 68106MB [2022-12-19 12:56:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1400/1519] eta 0:01:59 lr 0.000031 time 0.9790 (1.0049) model_time 0.9789 (1.0041) loss 0.8416 (0.9593) grad_norm 8.2564 (9.0854/2.4258) mem 68106MB [2022-12-19 12:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1410/1519] eta 0:01:49 lr 0.000031 time 0.9253 (1.0049) model_time 0.9252 (1.0040) loss 0.8296 (0.9598) grad_norm 6.8382 (9.0834/2.4382) mem 68106MB [2022-12-19 12:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1420/1519] eta 0:01:39 lr 0.000031 time 1.0511 (1.0050) model_time 1.0509 (1.0041) loss 0.8601 (0.9595) grad_norm 10.8669 (9.1180/2.4932) mem 68106MB [2022-12-19 12:56:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1430/1519] eta 0:01:29 lr 0.000031 time 0.9233 (1.0049) model_time 0.9232 (1.0041) loss 0.7897 (0.9593) grad_norm 10.5490 (9.1565/2.5256) mem 68106MB [2022-12-19 12:56:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1440/1519] eta 0:01:19 lr 0.000031 time 0.9306 (1.0049) model_time 0.9303 (1.0040) loss 0.8088 (0.9593) grad_norm 9.0571 (9.1125/2.4922) mem 68106MB [2022-12-19 12:57:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1450/1519] eta 0:01:09 lr 0.000031 time 0.9480 (1.0049) model_time 0.9478 (1.0041) loss 1.0539 (0.9593) grad_norm 13.8429 (9.0928/2.4343) mem 68106MB [2022-12-19 12:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1460/1519] eta 0:00:59 lr 0.000031 time 0.9297 (1.0049) model_time 0.9295 (1.0041) loss 0.7098 (0.9590) grad_norm 7.3528 (9.0410/2.4178) mem 68106MB [2022-12-19 12:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1470/1519] eta 0:00:49 lr 0.000031 time 0.9392 (1.0049) model_time 0.9390 (1.0040) loss 1.1871 (0.9593) grad_norm 13.5477 (9.0713/2.4169) mem 68106MB [2022-12-19 12:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1480/1519] eta 0:00:39 lr 0.000031 time 0.9322 (1.0049) model_time 0.9321 (1.0040) loss 1.1177 (0.9591) grad_norm 8.5531 (9.0401/2.4281) mem 68106MB [2022-12-19 12:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1490/1519] eta 0:00:29 lr 0.000031 time 0.9904 (1.0049) model_time 0.9902 (1.0040) loss 0.8551 (0.9589) grad_norm 11.1223 (9.0006/2.3573) mem 68106MB [2022-12-19 12:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1500/1519] eta 0:00:19 lr 0.000031 time 0.9302 (1.0048) model_time 0.9301 (1.0040) loss 0.8496 (0.9589) grad_norm 13.2792 (9.0100/2.3698) mem 68106MB [2022-12-19 12:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [19/100][1510/1519] eta 0:00:09 lr 0.000031 time 0.9253 (1.0048) model_time 0.9252 (1.0040) loss 0.9720 (0.9583) grad_norm 9.6783 (9.0061/2.3576) mem 68106MB [2022-12-19 12:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 19 training takes 0:25:26 [2022-12-19 12:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_19.pth saving...... [2022-12-19 12:58:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_19.pth saved !!! [2022-12-19 12:58:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.626 (0.626) Loss 0.5777 (0.5777) Acc@1 89.583 (89.583) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 12:58:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.326) Loss 0.5510 (0.5339) Acc@1 90.278 (90.120) Acc@5 97.569 (97.885) Mem 68106MB [2022-12-19 12:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.312) Loss 0.4971 (0.5364) Acc@1 90.625 (89.864) Acc@5 98.958 (97.834) Mem 68106MB [2022-12-19 12:58:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.307) Loss 0.6374 (0.5446) Acc@1 88.542 (89.729) Acc@5 97.569 (97.827) Mem 68106MB [2022-12-19 12:58:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.305) Loss 0.5425 (0.5362) Acc@1 89.583 (89.761) Acc@5 98.611 (97.951) Mem 68106MB [2022-12-19 12:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.304) Loss 0.5702 (0.5343) Acc@1 86.806 (89.740) Acc@5 98.264 (97.985) Mem 68106MB [2022-12-19 12:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.304 (0.303) Loss 0.5334 (0.5349) Acc@1 87.847 (89.714) Acc@5 97.569 (98.013) Mem 68106MB [2022-12-19 12:58:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.300 (0.303) Loss 0.5799 (0.5368) Acc@1 89.931 (89.666) Acc@5 98.264 (97.990) Mem 68106MB [2022-12-19 12:59:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.4678 (0.5357) Acc@1 92.361 (89.776) Acc@5 97.917 (98.002) Mem 68106MB [2022-12-19 12:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:19] * Acc@1 89.768 Acc@5 98.019 [2022-12-19 12:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 89.8% [2022-12-19 12:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 12:59:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 12:59:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 89.77% [2022-12-19 12:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][0/1519] eta 0:32:09 lr 0.000031 time 1.2702 (1.2702) model_time 0.9043 (0.9043) loss 1.3367 (1.3367) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 12:59:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][10/1519] eta 0:25:53 lr 0.000031 time 0.9314 (1.0293) model_time 0.9312 (0.9957) loss 0.9750 (1.0175) grad_norm 10.6625 (8.0629/1.4535) mem 68106MB [2022-12-19 12:59:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][20/1519] eta 0:25:23 lr 0.000031 time 0.9323 (1.0162) model_time 0.9322 (0.9985) loss 1.0437 (1.0249) grad_norm 7.0487 (7.7119/1.1462) mem 68106MB [2022-12-19 13:00:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][30/1519] eta 0:25:04 lr 0.000031 time 0.9225 (1.0101) model_time 0.9224 (0.9980) loss 1.0055 (1.0178) grad_norm 15.4405 (8.4618/2.5098) mem 68106MB [2022-12-19 13:00:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][40/1519] eta 0:24:55 lr 0.000031 time 0.9098 (1.0111) model_time 0.9096 (1.0018) loss 0.7889 (0.9795) grad_norm 11.6027 (8.9635/2.7333) mem 68106MB [2022-12-19 13:00:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][50/1519] eta 0:24:42 lr 0.000031 time 0.9264 (1.0091) model_time 0.9262 (1.0016) loss 0.7308 (0.9719) grad_norm 8.6087 (9.0305/2.5562) mem 68106MB [2022-12-19 13:00:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][60/1519] eta 0:24:33 lr 0.000031 time 0.9214 (1.0098) model_time 0.9212 (1.0035) loss 0.9439 (0.9666) grad_norm 6.7645 (9.1784/2.8441) mem 68106MB [2022-12-19 13:00:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][70/1519] eta 0:24:23 lr 0.000031 time 0.9282 (1.0097) model_time 0.9281 (1.0042) loss 1.1492 (0.9586) grad_norm 12.1942 (8.9323/2.8625) mem 68106MB [2022-12-19 13:00:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][80/1519] eta 0:24:10 lr 0.000031 time 0.9244 (1.0081) model_time 0.9243 (1.0033) loss 0.8072 (0.9471) grad_norm 7.7313 (8.9027/2.8177) mem 68106MB [2022-12-19 13:01:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][90/1519] eta 0:23:58 lr 0.000031 time 0.9212 (1.0070) model_time 0.9209 (1.0026) loss 1.0534 (0.9506) grad_norm 12.1084 (9.0699/2.8124) mem 68106MB [2022-12-19 13:01:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][100/1519] eta 0:23:47 lr 0.000031 time 0.9371 (1.0062) model_time 0.9369 (1.0022) loss 1.0163 (0.9493) grad_norm 8.6926 (9.1352/2.7520) mem 68106MB [2022-12-19 13:01:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][110/1519] eta 0:23:37 lr 0.000031 time 0.9605 (1.0058) model_time 0.9604 (1.0021) loss 0.7832 (0.9483) grad_norm 11.0316 (9.2134/2.7615) mem 68106MB [2022-12-19 13:01:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][120/1519] eta 0:23:26 lr 0.000031 time 0.9267 (1.0057) model_time 0.9266 (1.0023) loss 0.8177 (0.9457) grad_norm 7.1580 (9.2813/2.7341) mem 68106MB [2022-12-19 13:01:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][130/1519] eta 0:23:16 lr 0.000031 time 0.9781 (1.0053) model_time 0.9779 (1.0022) loss 1.2561 (0.9473) grad_norm 6.5191 (9.1287/2.7208) mem 68106MB [2022-12-19 13:01:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][140/1519] eta 0:23:06 lr 0.000031 time 1.0653 (1.0058) model_time 1.0651 (1.0028) loss 0.9010 (0.9537) grad_norm 12.8087 (9.2638/2.7940) mem 68106MB [2022-12-19 13:02:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][150/1519] eta 0:22:56 lr 0.000031 time 0.9294 (1.0055) model_time 0.9293 (1.0027) loss 0.9947 (0.9561) grad_norm 8.2925 (9.1897/2.7210) mem 68106MB [2022-12-19 13:02:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][160/1519] eta 0:22:45 lr 0.000031 time 0.9207 (1.0051) model_time 0.9206 (1.0025) loss 0.9907 (0.9550) grad_norm 7.1167 (9.2212/2.7144) mem 68106MB [2022-12-19 13:02:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][170/1519] eta 0:22:35 lr 0.000031 time 0.9217 (1.0051) model_time 0.9215 (1.0026) loss 0.7730 (0.9487) grad_norm 7.4910 (9.3068/2.7802) mem 68106MB [2022-12-19 13:02:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][180/1519] eta 0:22:25 lr 0.000031 time 0.9295 (1.0052) model_time 0.9293 (1.0028) loss 0.7385 (0.9441) grad_norm 7.6298 (9.2617/2.7146) mem 68106MB [2022-12-19 13:02:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][190/1519] eta 0:22:15 lr 0.000031 time 0.9144 (1.0051) model_time 0.9142 (1.0028) loss 0.7418 (0.9406) grad_norm 7.9992 (9.1924/2.6707) mem 68106MB [2022-12-19 13:02:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][200/1519] eta 0:22:05 lr 0.000031 time 0.9298 (1.0050) model_time 0.9297 (1.0028) loss 0.7950 (0.9403) grad_norm 5.9635 (9.1432/2.6775) mem 68106MB [2022-12-19 13:03:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][210/1519] eta 0:21:55 lr 0.000031 time 0.9338 (1.0048) model_time 0.9336 (1.0027) loss 0.8798 (0.9382) grad_norm 9.0650 (9.0920/2.6342) mem 68106MB [2022-12-19 13:03:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][220/1519] eta 0:21:45 lr 0.000031 time 0.9274 (1.0051) model_time 0.9271 (1.0031) loss 0.9857 (0.9387) grad_norm 6.3162 (9.0309/2.6045) mem 68106MB [2022-12-19 13:03:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][230/1519] eta 0:21:35 lr 0.000031 time 0.9240 (1.0049) model_time 0.9238 (1.0030) loss 0.7749 (0.9394) grad_norm 7.7266 (8.9537/2.5920) mem 68106MB [2022-12-19 13:03:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][240/1519] eta 0:21:25 lr 0.000031 time 0.9214 (1.0049) model_time 0.9213 (1.0030) loss 0.8104 (0.9390) grad_norm 9.1530 (9.0239/2.7554) mem 68106MB [2022-12-19 13:03:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][250/1519] eta 0:21:17 lr 0.000031 time 0.9293 (1.0066) model_time 0.9292 (1.0048) loss 1.1329 (0.9466) grad_norm 6.1741 (9.0293/2.7457) mem 68106MB [2022-12-19 13:03:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][260/1519] eta 0:21:07 lr 0.000031 time 0.9255 (1.0069) model_time 0.9253 (1.0052) loss 1.2842 (0.9482) grad_norm 20.0882 (9.1194/2.8785) mem 68106MB [2022-12-19 13:04:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][270/1519] eta 0:20:57 lr 0.000031 time 0.9252 (1.0068) model_time 0.9251 (1.0050) loss 0.7283 (0.9442) grad_norm 9.9870 (9.1169/2.8340) mem 68106MB [2022-12-19 13:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][280/1519] eta 0:20:47 lr 0.000031 time 0.9297 (1.0070) model_time 0.9296 (1.0054) loss 0.7384 (0.9476) grad_norm 8.7692 (9.1172/2.7866) mem 68106MB [2022-12-19 13:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][290/1519] eta 0:20:37 lr 0.000031 time 0.9325 (1.0070) model_time 0.9323 (1.0054) loss 1.1062 (0.9502) grad_norm 8.0900 (9.1579/2.7669) mem 68106MB [2022-12-19 13:04:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][300/1519] eta 0:20:27 lr 0.000030 time 0.9341 (1.0069) model_time 0.9340 (1.0053) loss 0.6774 (0.9459) grad_norm 7.2436 (9.1373/2.7594) mem 68106MB [2022-12-19 13:04:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][310/1519] eta 0:20:17 lr 0.000030 time 0.9318 (1.0067) model_time 0.9316 (1.0051) loss 0.9581 (0.9442) grad_norm 7.4954 (9.1428/2.7367) mem 68106MB [2022-12-19 13:04:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][320/1519] eta 0:20:06 lr 0.000030 time 0.9201 (1.0065) model_time 0.9199 (1.0050) loss 1.0007 (0.9464) grad_norm 7.0486 (9.1276/2.7112) mem 68106MB [2022-12-19 13:05:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][330/1519] eta 0:19:56 lr 0.000030 time 0.9199 (1.0064) model_time 0.9197 (1.0049) loss 1.0844 (0.9511) grad_norm 10.5339 (9.1284/2.6892) mem 68106MB [2022-12-19 13:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][340/1519] eta 0:19:46 lr 0.000030 time 0.9320 (1.0062) model_time 0.9318 (1.0047) loss 1.1701 (0.9517) grad_norm 8.4900 (9.1042/2.6648) mem 68106MB [2022-12-19 13:05:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][350/1519] eta 0:19:36 lr 0.000030 time 0.9455 (1.0061) model_time 0.9454 (1.0046) loss 0.7199 (0.9490) grad_norm 8.2380 (9.0796/2.6367) mem 68106MB [2022-12-19 13:05:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][360/1519] eta 0:19:26 lr 0.000030 time 0.9283 (1.0061) model_time 0.9281 (1.0047) loss 0.7925 (0.9489) grad_norm 8.2246 (9.0776/2.6107) mem 68106MB [2022-12-19 13:05:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][370/1519] eta 0:19:15 lr 0.000030 time 0.9196 (1.0060) model_time 0.9194 (1.0046) loss 0.8676 (0.9478) grad_norm 7.8824 (9.0497/2.5998) mem 68106MB [2022-12-19 13:05:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][380/1519] eta 0:19:05 lr 0.000030 time 0.9186 (1.0058) model_time 0.9184 (1.0045) loss 1.0180 (0.9461) grad_norm 12.3041 (9.1040/2.5895) mem 68106MB [2022-12-19 13:06:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][390/1519] eta 0:18:55 lr 0.000030 time 0.9335 (1.0061) model_time 0.9333 (1.0047) loss 0.8774 (0.9444) grad_norm 6.9630 (9.0867/2.5729) mem 68106MB [2022-12-19 13:06:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][400/1519] eta 0:18:45 lr 0.000030 time 0.9221 (1.0060) model_time 0.9220 (1.0047) loss 0.9901 (0.9443) grad_norm 8.3984 (9.1096/2.5579) mem 68106MB [2022-12-19 13:06:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][410/1519] eta 0:18:35 lr 0.000030 time 0.9241 (1.0059) model_time 0.9240 (1.0046) loss 1.0276 (0.9438) grad_norm 10.2068 (9.1231/2.5491) mem 68106MB [2022-12-19 13:06:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][420/1519] eta 0:18:25 lr 0.000030 time 0.9332 (1.0058) model_time 0.9331 (1.0045) loss 0.9514 (0.9430) grad_norm 5.4018 (9.0737/2.5476) mem 68106MB [2022-12-19 13:06:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][430/1519] eta 0:18:15 lr 0.000030 time 0.9658 (1.0059) model_time 0.9656 (1.0047) loss 0.7402 (0.9442) grad_norm 12.5032 (9.0884/2.5535) mem 68106MB [2022-12-19 13:06:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][440/1519] eta 0:18:05 lr 0.000030 time 0.9214 (1.0058) model_time 0.9213 (1.0046) loss 0.8169 (0.9439) grad_norm 8.0953 (9.0684/2.5339) mem 68106MB [2022-12-19 13:07:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][450/1519] eta 0:17:55 lr 0.000030 time 0.9254 (1.0059) model_time 0.9253 (1.0047) loss 1.2088 (0.9441) grad_norm 8.3955 (9.0938/2.5603) mem 68106MB [2022-12-19 13:07:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][460/1519] eta 0:17:45 lr 0.000030 time 0.9242 (1.0062) model_time 0.9241 (1.0050) loss 1.0267 (0.9434) grad_norm 9.2502 (9.1010/2.5506) mem 68106MB [2022-12-19 13:07:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][470/1519] eta 0:17:35 lr 0.000030 time 0.9320 (1.0062) model_time 0.9318 (1.0050) loss 0.6930 (0.9440) grad_norm 8.8988 (9.0829/2.5391) mem 68106MB [2022-12-19 13:07:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][480/1519] eta 0:17:25 lr 0.000030 time 0.9197 (1.0060) model_time 0.9196 (1.0049) loss 0.8327 (0.9456) grad_norm 9.1380 (9.0587/2.5257) mem 68106MB [2022-12-19 13:07:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][490/1519] eta 0:17:15 lr 0.000030 time 0.9209 (1.0058) model_time 0.9207 (1.0047) loss 0.7529 (0.9455) grad_norm 6.4282 (9.0329/2.5172) mem 68106MB [2022-12-19 13:07:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][500/1519] eta 0:17:04 lr 0.000030 time 0.9239 (1.0059) model_time 0.9238 (1.0048) loss 0.7255 (0.9454) grad_norm 9.3322 (9.0593/2.5243) mem 68106MB [2022-12-19 13:08:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][510/1519] eta 0:16:54 lr 0.000030 time 0.9237 (1.0057) model_time 0.9236 (1.0046) loss 0.7801 (0.9454) grad_norm 9.9078 (9.0466/2.5063) mem 68106MB [2022-12-19 13:08:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][520/1519] eta 0:16:44 lr 0.000030 time 0.9266 (1.0056) model_time 0.9265 (1.0045) loss 0.9168 (0.9454) grad_norm 8.3946 (9.0516/2.4877) mem 68106MB [2022-12-19 13:08:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][530/1519] eta 0:16:34 lr 0.000030 time 0.9313 (1.0056) model_time 0.9312 (1.0045) loss 1.1347 (0.9457) grad_norm 10.1108 (9.0481/2.4684) mem 68106MB [2022-12-19 13:08:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][540/1519] eta 0:16:24 lr 0.000030 time 0.9202 (1.0054) model_time 0.9201 (1.0044) loss 0.7609 (0.9454) grad_norm 16.1537 (9.0789/2.4900) mem 68106MB [2022-12-19 13:08:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][550/1519] eta 0:16:14 lr 0.000030 time 0.9115 (1.0054) model_time 0.9114 (1.0044) loss 0.7591 (0.9442) grad_norm 11.0251 (9.0416/2.4980) mem 68106MB [2022-12-19 13:08:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][560/1519] eta 0:16:04 lr 0.000030 time 0.9237 (1.0053) model_time 0.9235 (1.0043) loss 0.7278 (0.9447) grad_norm 12.8147 (9.0309/2.4973) mem 68106MB [2022-12-19 13:09:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][570/1519] eta 0:15:54 lr 0.000030 time 0.9243 (1.0053) model_time 0.9242 (1.0043) loss 1.1403 (0.9444) grad_norm 11.5689 (9.0251/2.4843) mem 68106MB [2022-12-19 13:09:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][580/1519] eta 0:15:43 lr 0.000030 time 0.9229 (1.0053) model_time 0.9228 (1.0043) loss 0.7249 (0.9434) grad_norm 7.9269 (9.0194/2.4698) mem 68106MB [2022-12-19 13:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][590/1519] eta 0:15:33 lr 0.000030 time 0.9331 (1.0052) model_time 0.9328 (1.0042) loss 0.7580 (0.9444) grad_norm 9.0390 (9.0666/2.5146) mem 68106MB [2022-12-19 13:09:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][600/1519] eta 0:15:23 lr 0.000030 time 0.9219 (1.0050) model_time 0.9218 (1.0041) loss 0.8159 (0.9440) grad_norm 9.9714 (9.0840/2.5013) mem 68106MB [2022-12-19 13:09:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][610/1519] eta 0:15:13 lr 0.000030 time 0.9200 (1.0050) model_time 0.9198 (1.0040) loss 0.8590 (0.9441) grad_norm 6.8575 (9.0799/2.5007) mem 68106MB [2022-12-19 13:09:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][620/1519] eta 0:15:03 lr 0.000030 time 0.9153 (1.0050) model_time 0.9151 (1.0040) loss 1.0920 (0.9453) grad_norm 6.8146 (9.0947/2.5000) mem 68106MB [2022-12-19 13:10:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][630/1519] eta 0:14:53 lr 0.000030 time 0.9210 (1.0050) model_time 0.9208 (1.0041) loss 0.9369 (0.9458) grad_norm 7.0588 (9.0735/2.4597) mem 68106MB [2022-12-19 13:10:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][640/1519] eta 0:14:43 lr 0.000030 time 0.9344 (1.0050) model_time 0.9342 (1.0041) loss 1.3916 (0.9476) grad_norm 6.8887 (9.0317/2.4331) mem 68106MB [2022-12-19 13:10:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][650/1519] eta 0:14:33 lr 0.000030 time 0.9204 (1.0049) model_time 0.9203 (1.0040) loss 0.8254 (0.9476) grad_norm 9.1460 (9.0440/2.4529) mem 68106MB [2022-12-19 13:10:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][660/1519] eta 0:14:23 lr 0.000030 time 0.9291 (1.0048) model_time 0.9290 (1.0039) loss 0.7261 (0.9464) grad_norm 10.3290 (9.0138/2.4069) mem 68106MB [2022-12-19 13:10:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][670/1519] eta 0:14:13 lr 0.000030 time 0.9175 (1.0051) model_time 0.9174 (1.0042) loss 0.9766 (0.9471) grad_norm 7.7975 (9.0235/2.3820) mem 68106MB [2022-12-19 13:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][680/1519] eta 0:14:03 lr 0.000030 time 0.9309 (1.0050) model_time 0.9307 (1.0041) loss 0.9582 (0.9464) grad_norm 11.4625 (9.0204/2.3832) mem 68106MB [2022-12-19 13:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][690/1519] eta 0:13:53 lr 0.000030 time 0.9265 (1.0049) model_time 0.9263 (1.0040) loss 1.0242 (0.9461) grad_norm 9.6236 (9.0067/2.3805) mem 68106MB [2022-12-19 13:11:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][700/1519] eta 0:13:43 lr 0.000030 time 0.9193 (1.0049) model_time 0.9191 (1.0040) loss 1.1150 (0.9456) grad_norm 8.7457 (8.9784/2.3693) mem 68106MB [2022-12-19 13:11:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][710/1519] eta 0:13:32 lr 0.000030 time 0.9297 (1.0049) model_time 0.9295 (1.0040) loss 0.8291 (0.9461) grad_norm 7.1074 (8.9346/2.3506) mem 68106MB [2022-12-19 13:11:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][720/1519] eta 0:13:22 lr 0.000030 time 0.9265 (1.0049) model_time 0.9264 (1.0040) loss 0.8778 (0.9466) grad_norm 7.7428 (8.8923/2.3357) mem 68106MB [2022-12-19 13:11:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][730/1519] eta 0:13:12 lr 0.000030 time 0.9343 (1.0048) model_time 0.9342 (1.0039) loss 1.0461 (0.9469) grad_norm 7.7924 (8.8946/2.3254) mem 68106MB [2022-12-19 13:11:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][740/1519] eta 0:13:02 lr 0.000030 time 0.9271 (1.0048) model_time 0.9269 (1.0039) loss 0.7474 (0.9463) grad_norm 6.3943 (8.8462/2.2815) mem 68106MB [2022-12-19 13:12:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][750/1519] eta 0:12:52 lr 0.000030 time 0.9291 (1.0047) model_time 0.9288 (1.0038) loss 0.7128 (0.9450) grad_norm 6.6105 (8.8424/2.2830) mem 68106MB [2022-12-19 13:12:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][760/1519] eta 0:12:42 lr 0.000030 time 0.9297 (1.0046) model_time 0.9295 (1.0037) loss 0.7787 (0.9446) grad_norm 7.0415 (8.8667/2.3412) mem 68106MB [2022-12-19 13:12:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][770/1519] eta 0:12:32 lr 0.000030 time 0.9245 (1.0047) model_time 0.9244 (1.0038) loss 0.7351 (0.9446) grad_norm 5.9578 (8.8250/2.3017) mem 68106MB [2022-12-19 13:12:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][780/1519] eta 0:12:22 lr 0.000030 time 0.9327 (1.0046) model_time 0.9325 (1.0038) loss 1.1637 (0.9448) grad_norm 8.2970 (8.8295/2.3191) mem 68106MB [2022-12-19 13:12:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][790/1519] eta 0:12:12 lr 0.000030 time 0.9640 (1.0046) model_time 0.9638 (1.0037) loss 0.9496 (0.9452) grad_norm 11.8105 (8.8472/2.3282) mem 68106MB [2022-12-19 13:12:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][800/1519] eta 0:12:02 lr 0.000030 time 0.9245 (1.0045) model_time 0.9244 (1.0037) loss 0.9792 (0.9460) grad_norm 11.8158 (8.8639/2.3130) mem 68106MB [2022-12-19 13:13:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][810/1519] eta 0:11:52 lr 0.000030 time 0.9242 (1.0044) model_time 0.9241 (1.0036) loss 0.9373 (0.9476) grad_norm 8.6625 (8.8709/2.3126) mem 68106MB [2022-12-19 13:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][820/1519] eta 0:11:42 lr 0.000030 time 0.9258 (1.0044) model_time 0.9256 (1.0036) loss 0.8451 (0.9476) grad_norm 7.6344 (8.8820/2.3110) mem 68106MB [2022-12-19 13:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][830/1519] eta 0:11:32 lr 0.000030 time 0.9308 (1.0044) model_time 0.9307 (1.0035) loss 0.8718 (0.9479) grad_norm 8.5975 (8.8960/2.2995) mem 68106MB [2022-12-19 13:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][840/1519] eta 0:11:22 lr 0.000030 time 0.9249 (1.0046) model_time 0.9247 (1.0038) loss 0.9853 (0.9492) grad_norm 7.3314 (8.8536/2.2054) mem 68106MB [2022-12-19 13:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][850/1519] eta 0:11:12 lr 0.000030 time 0.9298 (1.0047) model_time 0.9296 (1.0039) loss 0.8102 (0.9490) grad_norm 6.4229 (8.8145/2.1979) mem 68106MB [2022-12-19 13:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][860/1519] eta 0:11:02 lr 0.000030 time 0.9216 (1.0047) model_time 0.9214 (1.0039) loss 0.7893 (0.9499) grad_norm 7.8913 (8.7818/2.0950) mem 68106MB [2022-12-19 13:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][870/1519] eta 0:10:52 lr 0.000030 time 0.9434 (1.0047) model_time 0.9433 (1.0039) loss 1.2239 (0.9506) grad_norm 13.7199 (8.7885/2.1302) mem 68106MB [2022-12-19 13:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][880/1519] eta 0:10:42 lr 0.000030 time 0.9548 (1.0049) model_time 0.9546 (1.0041) loss 1.1640 (0.9510) grad_norm 6.8137 (8.7689/2.1360) mem 68106MB [2022-12-19 13:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][890/1519] eta 0:10:32 lr 0.000030 time 0.9306 (1.0049) model_time 0.9305 (1.0041) loss 0.7268 (0.9499) grad_norm 8.4715 (8.7291/2.1173) mem 68106MB [2022-12-19 13:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][900/1519] eta 0:10:22 lr 0.000030 time 0.9352 (1.0049) model_time 0.9350 (1.0041) loss 1.0575 (0.9497) grad_norm 7.8155 (8.7317/2.1007) mem 68106MB [2022-12-19 13:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][910/1519] eta 0:10:11 lr 0.000030 time 0.9300 (1.0048) model_time 0.9296 (1.0040) loss 1.4261 (0.9505) grad_norm 7.2728 (8.7230/2.0913) mem 68106MB [2022-12-19 13:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][920/1519] eta 0:10:01 lr 0.000030 time 0.9344 (1.0048) model_time 0.9342 (1.0040) loss 1.0964 (0.9510) grad_norm 8.5720 (8.7475/2.1144) mem 68106MB [2022-12-19 13:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][930/1519] eta 0:09:51 lr 0.000030 time 0.9244 (1.0048) model_time 0.9242 (1.0040) loss 1.0546 (0.9511) grad_norm 6.7276 (8.7280/2.1096) mem 68106MB [2022-12-19 13:15:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][940/1519] eta 0:09:41 lr 0.000030 time 0.9332 (1.0048) model_time 0.9330 (1.0040) loss 1.4002 (0.9517) grad_norm 7.3647 (8.7650/2.1523) mem 68106MB [2022-12-19 13:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][950/1519] eta 0:09:31 lr 0.000030 time 0.9279 (1.0048) model_time 0.9277 (1.0040) loss 0.9726 (0.9534) grad_norm 11.5102 (8.8222/2.2431) mem 68106MB [2022-12-19 13:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][960/1519] eta 0:09:21 lr 0.000030 time 0.9231 (1.0047) model_time 0.9230 (1.0040) loss 0.8924 (0.9532) grad_norm 7.0077 (8.8626/2.2864) mem 68106MB [2022-12-19 13:15:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][970/1519] eta 0:09:11 lr 0.000030 time 0.9745 (1.0047) model_time 0.9744 (1.0040) loss 0.8238 (0.9527) grad_norm 8.3670 (8.9020/2.2890) mem 68106MB [2022-12-19 13:15:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][980/1519] eta 0:09:01 lr 0.000030 time 0.9312 (1.0047) model_time 0.9309 (1.0040) loss 0.7654 (0.9529) grad_norm 9.3993 (8.8725/2.2761) mem 68106MB [2022-12-19 13:16:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][990/1519] eta 0:08:51 lr 0.000030 time 0.9299 (1.0047) model_time 0.9297 (1.0040) loss 0.7236 (0.9533) grad_norm 6.7401 (8.8515/2.2790) mem 68106MB [2022-12-19 13:16:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1000/1519] eta 0:08:41 lr 0.000030 time 0.9276 (1.0047) model_time 0.9274 (1.0039) loss 1.1134 (0.9538) grad_norm 7.9693 (8.8043/2.2755) mem 68106MB [2022-12-19 13:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1010/1519] eta 0:08:31 lr 0.000030 time 0.9296 (1.0047) model_time 0.9295 (1.0039) loss 0.7024 (0.9531) grad_norm 7.3855 (8.7793/2.2698) mem 68106MB [2022-12-19 13:16:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1020/1519] eta 0:08:21 lr 0.000030 time 0.9440 (1.0047) model_time 0.9437 (1.0039) loss 0.9147 (0.9532) grad_norm 9.4675 (8.7972/2.2572) mem 68106MB [2022-12-19 13:16:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1030/1519] eta 0:08:11 lr 0.000030 time 0.9301 (1.0046) model_time 0.9300 (1.0039) loss 0.7603 (0.9532) grad_norm 7.7043 (8.7878/2.2350) mem 68106MB [2022-12-19 13:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1040/1519] eta 0:08:01 lr 0.000030 time 0.9275 (1.0046) model_time 0.9273 (1.0038) loss 1.2456 (0.9521) grad_norm 10.3235 (8.7883/2.2406) mem 68106MB [2022-12-19 13:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1050/1519] eta 0:07:51 lr 0.000030 time 0.9234 (1.0045) model_time 0.9233 (1.0038) loss 1.1711 (0.9524) grad_norm 11.3544 (8.7691/2.2059) mem 68106MB [2022-12-19 13:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1060/1519] eta 0:07:41 lr 0.000030 time 0.9281 (1.0045) model_time 0.9279 (1.0038) loss 1.1197 (0.9529) grad_norm 7.3319 (8.7766/2.2316) mem 68106MB [2022-12-19 13:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1070/1519] eta 0:07:31 lr 0.000030 time 0.9313 (1.0045) model_time 0.9312 (1.0038) loss 0.8927 (0.9526) grad_norm 7.1473 (8.7969/2.2270) mem 68106MB [2022-12-19 13:17:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1080/1519] eta 0:07:20 lr 0.000030 time 0.9032 (1.0045) model_time 0.9031 (1.0038) loss 0.7672 (0.9523) grad_norm 7.8855 (8.7790/2.2328) mem 68106MB [2022-12-19 13:17:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1090/1519] eta 0:07:10 lr 0.000030 time 0.9531 (1.0044) model_time 0.9530 (1.0037) loss 0.8839 (0.9525) grad_norm 8.0638 (8.7939/2.2205) mem 68106MB [2022-12-19 13:17:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1100/1519] eta 0:07:00 lr 0.000030 time 0.9244 (1.0044) model_time 0.9243 (1.0037) loss 0.8408 (0.9526) grad_norm 13.4330 (8.7709/2.2128) mem 68106MB [2022-12-19 13:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1110/1519] eta 0:06:50 lr 0.000030 time 0.9437 (1.0044) model_time 0.9434 (1.0036) loss 1.2669 (0.9532) grad_norm 15.5210 (8.8497/2.3001) mem 68106MB [2022-12-19 13:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1120/1519] eta 0:06:40 lr 0.000030 time 0.9303 (1.0043) model_time 0.9300 (1.0036) loss 1.2112 (0.9537) grad_norm 7.4739 (8.8383/2.3007) mem 68106MB [2022-12-19 13:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1130/1519] eta 0:06:30 lr 0.000030 time 0.9271 (1.0043) model_time 0.9269 (1.0036) loss 0.8550 (0.9535) grad_norm 7.5351 (8.8267/2.3002) mem 68106MB [2022-12-19 13:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1140/1519] eta 0:06:20 lr 0.000030 time 0.9363 (1.0042) model_time 0.9361 (1.0035) loss 0.8286 (0.9539) grad_norm 7.2081 (8.8134/2.2956) mem 68106MB [2022-12-19 13:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1150/1519] eta 0:06:10 lr 0.000030 time 0.9258 (1.0041) model_time 0.9255 (1.0034) loss 0.8460 (0.9532) grad_norm 6.8060 (8.8093/2.2385) mem 68106MB [2022-12-19 13:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1160/1519] eta 0:06:00 lr 0.000030 time 0.9167 (1.0043) model_time 0.9166 (1.0036) loss 0.7625 (0.9529) grad_norm 9.8303 (8.8145/2.2301) mem 68106MB [2022-12-19 13:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1170/1519] eta 0:05:50 lr 0.000030 time 0.9244 (1.0042) model_time 0.9243 (1.0035) loss 0.7883 (0.9524) grad_norm 10.4193 (8.8299/2.2504) mem 68106MB [2022-12-19 13:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1180/1519] eta 0:05:40 lr 0.000030 time 0.9206 (1.0042) model_time 0.9204 (1.0035) loss 0.8647 (0.9516) grad_norm 6.1828 (8.8200/2.2650) mem 68106MB [2022-12-19 13:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1190/1519] eta 0:05:30 lr 0.000030 time 0.9230 (1.0043) model_time 0.9228 (1.0036) loss 0.8298 (0.9518) grad_norm 6.2611 (8.7530/2.1963) mem 68106MB [2022-12-19 13:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1200/1519] eta 0:05:20 lr 0.000030 time 0.9336 (1.0042) model_time 0.9335 (1.0035) loss 0.8562 (0.9512) grad_norm 9.7566 (8.7154/2.1952) mem 68106MB [2022-12-19 13:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1210/1519] eta 0:05:10 lr 0.000030 time 0.9271 (1.0042) model_time 0.9268 (1.0035) loss 1.3469 (0.9519) grad_norm 5.9679 (8.7142/2.2044) mem 68106MB [2022-12-19 13:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1220/1519] eta 0:05:00 lr 0.000030 time 0.9332 (1.0041) model_time 0.9330 (1.0034) loss 0.8240 (0.9519) grad_norm 6.0908 (8.7029/2.2044) mem 68106MB [2022-12-19 13:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1230/1519] eta 0:04:50 lr 0.000030 time 0.9344 (1.0041) model_time 0.9343 (1.0034) loss 1.0079 (0.9518) grad_norm 9.1927 (8.7099/2.2032) mem 68106MB [2022-12-19 13:20:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1240/1519] eta 0:04:40 lr 0.000030 time 0.9270 (1.0041) model_time 0.9269 (1.0034) loss 1.0708 (0.9524) grad_norm 9.7318 (8.7230/2.2033) mem 68106MB [2022-12-19 13:20:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1250/1519] eta 0:04:30 lr 0.000030 time 0.9273 (1.0041) model_time 0.9271 (1.0034) loss 0.8517 (0.9521) grad_norm 6.8760 (8.7189/2.1796) mem 68106MB [2022-12-19 13:20:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1260/1519] eta 0:04:20 lr 0.000030 time 0.9634 (1.0042) model_time 0.9631 (1.0035) loss 1.6331 (0.9517) grad_norm 10.0552 (8.7250/2.1849) mem 68106MB [2022-12-19 13:20:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1270/1519] eta 0:04:10 lr 0.000030 time 0.9324 (1.0041) model_time 0.9323 (1.0035) loss 0.8476 (0.9527) grad_norm 10.5688 (8.7425/2.1960) mem 68106MB [2022-12-19 13:20:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1280/1519] eta 0:03:59 lr 0.000030 time 0.9206 (1.0041) model_time 0.9204 (1.0034) loss 0.7448 (0.9530) grad_norm 15.6514 (8.8011/2.2426) mem 68106MB [2022-12-19 13:21:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1290/1519] eta 0:03:49 lr 0.000030 time 0.9763 (1.0041) model_time 0.9762 (1.0035) loss 1.0963 (0.9538) grad_norm 7.9082 (8.7520/2.2257) mem 68106MB [2022-12-19 13:21:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1300/1519] eta 0:03:39 lr 0.000030 time 0.9249 (1.0041) model_time 0.9248 (1.0034) loss 0.8176 (0.9538) grad_norm 7.1698 (8.7430/2.2385) mem 68106MB [2022-12-19 13:21:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1310/1519] eta 0:03:29 lr 0.000030 time 0.9217 (1.0040) model_time 0.9215 (1.0034) loss 0.9524 (0.9534) grad_norm 9.0191 (8.7568/2.2374) mem 68106MB [2022-12-19 13:21:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1320/1519] eta 0:03:19 lr 0.000030 time 0.9281 (1.0040) model_time 0.9279 (1.0033) loss 0.6956 (0.9533) grad_norm 8.5457 (8.7693/2.2322) mem 68106MB [2022-12-19 13:21:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1330/1519] eta 0:03:09 lr 0.000030 time 0.9223 (1.0040) model_time 0.9222 (1.0033) loss 0.8944 (0.9532) grad_norm 7.8710 (8.8043/2.2322) mem 68106MB [2022-12-19 13:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1340/1519] eta 0:02:59 lr 0.000030 time 0.9247 (1.0040) model_time 0.9245 (1.0034) loss 0.7562 (0.9534) grad_norm 8.0826 (8.8220/2.2492) mem 68106MB [2022-12-19 13:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1350/1519] eta 0:02:49 lr 0.000030 time 0.9273 (1.0040) model_time 0.9272 (1.0033) loss 0.9236 (0.9540) grad_norm 11.0742 (8.8477/2.2493) mem 68106MB [2022-12-19 13:22:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1360/1519] eta 0:02:39 lr 0.000030 time 0.9985 (1.0040) model_time 0.9984 (1.0034) loss 0.9702 (0.9535) grad_norm 8.8212 (8.7958/2.1682) mem 68106MB [2022-12-19 13:22:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1370/1519] eta 0:02:29 lr 0.000030 time 0.9210 (1.0042) model_time 0.9209 (1.0035) loss 0.8282 (0.9531) grad_norm 7.8250 (8.7859/2.1644) mem 68106MB [2022-12-19 13:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1380/1519] eta 0:02:19 lr 0.000030 time 0.9126 (1.0042) model_time 0.9124 (1.0035) loss 1.1188 (0.9530) grad_norm 12.7704 (8.8246/2.1605) mem 68106MB [2022-12-19 13:22:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1390/1519] eta 0:02:09 lr 0.000030 time 0.9216 (1.0041) model_time 0.9214 (1.0035) loss 0.8176 (0.9527) grad_norm 11.5634 (8.8358/2.1585) mem 68106MB [2022-12-19 13:22:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1400/1519] eta 0:01:59 lr 0.000030 time 0.9305 (1.0041) model_time 0.9303 (1.0035) loss 1.0161 (0.9526) grad_norm 8.4782 (8.8133/2.1532) mem 68106MB [2022-12-19 13:23:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1410/1519] eta 0:01:49 lr 0.000030 time 0.9180 (1.0041) model_time 0.9178 (1.0034) loss 0.8054 (0.9522) grad_norm 6.3406 (8.7871/2.1474) mem 68106MB [2022-12-19 13:23:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1420/1519] eta 0:01:39 lr 0.000030 time 0.9352 (1.0041) model_time 0.9350 (1.0035) loss 0.8607 (0.9524) grad_norm 7.5307 (8.7758/2.1450) mem 68106MB [2022-12-19 13:23:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1430/1519] eta 0:01:29 lr 0.000030 time 0.9258 (1.0041) model_time 0.9255 (1.0034) loss 1.2668 (0.9522) grad_norm 12.8031 (8.8002/2.1590) mem 68106MB [2022-12-19 13:23:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1440/1519] eta 0:01:19 lr 0.000030 time 0.9220 (1.0041) model_time 0.9219 (1.0034) loss 1.3567 (0.9519) grad_norm 10.4512 (8.8090/2.1568) mem 68106MB [2022-12-19 13:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1450/1519] eta 0:01:09 lr 0.000030 time 0.9226 (1.0041) model_time 0.9225 (1.0034) loss 1.0062 (0.9522) grad_norm 10.8132 (8.8457/2.1546) mem 68106MB [2022-12-19 13:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1460/1519] eta 0:00:59 lr 0.000030 time 0.9305 (1.0040) model_time 0.9303 (1.0034) loss 0.9035 (0.9520) grad_norm 7.2508 (8.8511/2.1689) mem 68106MB [2022-12-19 13:24:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1470/1519] eta 0:00:49 lr 0.000030 time 1.0102 (1.0041) model_time 1.0100 (1.0034) loss 1.0277 (0.9514) grad_norm 9.0684 (8.8689/2.1653) mem 68106MB [2022-12-19 13:24:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1480/1519] eta 0:00:39 lr 0.000030 time 0.9391 (1.0041) model_time 0.9388 (1.0034) loss 1.1048 (0.9512) grad_norm 9.2944 (8.8783/2.1539) mem 68106MB [2022-12-19 13:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1490/1519] eta 0:00:29 lr 0.000030 time 0.9237 (1.0040) model_time 0.9236 (1.0034) loss 0.9312 (0.9520) grad_norm 10.3609 (8.8975/2.1492) mem 68106MB [2022-12-19 13:24:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1500/1519] eta 0:00:19 lr 0.000030 time 0.9172 (1.0040) model_time 0.9171 (1.0034) loss 0.9308 (0.9512) grad_norm 8.8246 (8.8918/2.1458) mem 68106MB [2022-12-19 13:24:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [20/100][1510/1519] eta 0:00:09 lr 0.000030 time 0.9272 (1.0040) model_time 0.9271 (1.0034) loss 1.0989 (0.9514) grad_norm 7.3952 (8.8744/2.1449) mem 68106MB [2022-12-19 13:24:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 20 training takes 0:25:25 [2022-12-19 13:24:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_20.pth saving...... [2022-12-19 13:25:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_20.pth saved !!! [2022-12-19 13:25:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.5495 (0.5495) Acc@1 90.625 (90.625) Acc@5 98.264 (98.264) Mem 68106MB [2022-12-19 13:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.334) Loss 0.5401 (0.5215) Acc@1 90.278 (90.436) Acc@5 97.222 (98.201) Mem 68106MB [2022-12-19 13:25:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.317) Loss 0.4913 (0.5251) Acc@1 91.667 (90.129) Acc@5 98.958 (98.065) Mem 68106MB [2022-12-19 13:25:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 0.6330 (0.5317) Acc@1 88.194 (89.998) Acc@5 96.875 (97.995) Mem 68106MB [2022-12-19 13:25:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.308) Loss 0.5132 (0.5247) Acc@1 89.236 (90.032) Acc@5 98.611 (98.086) Mem 68106MB [2022-12-19 13:25:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.5458 (0.5211) Acc@1 85.417 (89.978) Acc@5 98.958 (98.148) Mem 68106MB [2022-12-19 13:25:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.295 (0.305) Loss 0.5500 (0.5224) Acc@1 87.500 (89.959) Acc@5 97.917 (98.167) Mem 68106MB [2022-12-19 13:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.300 (0.304) Loss 0.5881 (0.5234) Acc@1 90.278 (89.916) Acc@5 97.569 (98.161) Mem 68106MB [2022-12-19 13:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.299 (0.303) Loss 0.4562 (0.5225) Acc@1 92.361 (89.999) Acc@5 98.264 (98.165) Mem 68106MB [2022-12-19 13:25:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:20] * Acc@1 89.997 Acc@5 98.175 [2022-12-19 13:25:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 90.0% [2022-12-19 13:25:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 13:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 13:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.00% [2022-12-19 13:26:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][0/1519] eta 0:33:04 lr 0.000030 time 1.3064 (1.3064) model_time 0.9089 (0.9089) loss 0.9201 (0.9201) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 13:26:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][10/1519] eta 0:25:56 lr 0.000030 time 0.9229 (1.0316) model_time 0.9226 (0.9951) loss 0.9965 (1.0050) grad_norm 7.0851 (8.9692/3.4774) mem 68106MB [2022-12-19 13:26:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][20/1519] eta 0:25:22 lr 0.000030 time 0.9295 (1.0157) model_time 0.9294 (0.9965) loss 0.8343 (0.9781) grad_norm 8.5719 (8.7979/3.2630) mem 68106MB [2022-12-19 13:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][30/1519] eta 0:25:12 lr 0.000030 time 0.9301 (1.0157) model_time 0.9299 (1.0025) loss 1.0318 (0.9684) grad_norm 9.6272 (8.7648/2.7389) mem 68106MB [2022-12-19 13:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][40/1519] eta 0:24:58 lr 0.000030 time 0.9244 (1.0129) model_time 0.9243 (1.0028) loss 0.7153 (0.9837) grad_norm 8.3585 (9.0378/2.7295) mem 68106MB [2022-12-19 13:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][50/1519] eta 0:24:44 lr 0.000030 time 0.9278 (1.0104) model_time 0.9277 (1.0022) loss 0.8520 (0.9743) grad_norm 7.1507 (8.7559/2.5810) mem 68106MB [2022-12-19 13:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][60/1519] eta 0:24:32 lr 0.000030 time 0.9335 (1.0094) model_time 0.9334 (1.0025) loss 0.8237 (0.9570) grad_norm 9.2131 (8.8658/2.6444) mem 68106MB [2022-12-19 13:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][70/1519] eta 0:24:21 lr 0.000030 time 0.9384 (1.0085) model_time 0.9383 (1.0025) loss 1.3998 (0.9846) grad_norm 6.6518 (8.7473/2.4974) mem 68106MB [2022-12-19 13:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][80/1519] eta 0:24:10 lr 0.000030 time 0.9264 (1.0077) model_time 0.9262 (1.0024) loss 0.8665 (0.9864) grad_norm 10.5081 (8.9421/2.5445) mem 68106MB [2022-12-19 13:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][90/1519] eta 0:23:58 lr 0.000030 time 0.9283 (1.0067) model_time 0.9282 (1.0020) loss 0.8828 (0.9855) grad_norm 9.3971 (9.1105/2.5860) mem 68106MB [2022-12-19 13:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][100/1519] eta 0:23:47 lr 0.000030 time 0.9319 (1.0060) model_time 0.9317 (1.0017) loss 0.9232 (0.9861) grad_norm 6.8294 (9.0119/2.4906) mem 68106MB [2022-12-19 13:28:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][110/1519] eta 0:23:37 lr 0.000030 time 0.9266 (1.0058) model_time 0.9264 (1.0018) loss 1.0188 (0.9920) grad_norm 7.1588 (8.9327/2.4437) mem 68106MB [2022-12-19 13:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][120/1519] eta 0:23:26 lr 0.000030 time 0.9228 (1.0052) model_time 0.9226 (1.0015) loss 1.3540 (0.9883) grad_norm 6.1566 (8.9845/2.5530) mem 68106MB [2022-12-19 13:28:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][130/1519] eta 0:23:16 lr 0.000030 time 0.9195 (1.0052) model_time 0.9193 (1.0018) loss 0.8168 (0.9823) grad_norm 7.4564 (8.9381/2.5269) mem 68106MB [2022-12-19 13:28:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][140/1519] eta 0:23:06 lr 0.000030 time 1.0437 (1.0054) model_time 1.0434 (1.0022) loss 0.7357 (0.9789) grad_norm 7.9078 (8.9648/2.5694) mem 68106MB [2022-12-19 13:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][150/1519] eta 0:22:56 lr 0.000030 time 0.9199 (1.0058) model_time 0.9198 (1.0028) loss 0.8896 (0.9782) grad_norm 9.0480 (9.0399/2.5446) mem 68106MB [2022-12-19 13:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][160/1519] eta 0:22:47 lr 0.000030 time 1.0271 (1.0063) model_time 1.0270 (1.0035) loss 0.8791 (0.9815) grad_norm 9.1891 (8.9376/2.5125) mem 68106MB [2022-12-19 13:29:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][170/1519] eta 0:22:36 lr 0.000030 time 0.9303 (1.0059) model_time 0.9300 (1.0030) loss 0.9469 (0.9813) grad_norm 6.0643 (8.9008/2.4767) mem 68106MB [2022-12-19 13:29:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][180/1519] eta 0:22:26 lr 0.000030 time 0.9216 (1.0057) model_time 0.9215 (1.0029) loss 0.6928 (0.9778) grad_norm 8.8791 (8.8351/2.4303) mem 68106MB [2022-12-19 13:29:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][190/1519] eta 0:22:16 lr 0.000030 time 0.9214 (1.0053) model_time 0.9213 (1.0027) loss 0.8216 (0.9721) grad_norm 6.8636 (8.7784/2.4066) mem 68106MB [2022-12-19 13:29:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][200/1519] eta 0:22:05 lr 0.000030 time 0.9357 (1.0051) model_time 0.9355 (1.0026) loss 0.8869 (0.9690) grad_norm 5.8961 (8.7455/2.3694) mem 68106MB [2022-12-19 13:29:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][210/1519] eta 0:21:55 lr 0.000030 time 0.9783 (1.0051) model_time 0.9782 (1.0026) loss 1.2708 (0.9655) grad_norm 9.0994 (8.7326/2.3461) mem 68106MB [2022-12-19 13:29:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][220/1519] eta 0:21:45 lr 0.000030 time 0.9245 (1.0048) model_time 0.9244 (1.0025) loss 0.6796 (0.9651) grad_norm 6.4589 (8.7103/2.3514) mem 68106MB [2022-12-19 13:30:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][230/1519] eta 0:21:34 lr 0.000030 time 0.9273 (1.0046) model_time 0.9272 (1.0023) loss 0.7193 (0.9633) grad_norm 9.1840 (8.6656/2.3247) mem 68106MB [2022-12-19 13:30:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][240/1519] eta 0:21:24 lr 0.000030 time 0.9243 (1.0044) model_time 0.9242 (1.0022) loss 0.8984 (0.9588) grad_norm 10.5234 (8.6558/2.2998) mem 68106MB [2022-12-19 13:30:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][250/1519] eta 0:21:14 lr 0.000030 time 0.9168 (1.0041) model_time 0.9166 (1.0020) loss 0.8350 (0.9600) grad_norm 13.5039 (8.7332/2.3209) mem 68106MB [2022-12-19 13:30:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][260/1519] eta 0:21:03 lr 0.000030 time 0.9302 (1.0039) model_time 0.9300 (1.0019) loss 0.7023 (0.9614) grad_norm 9.2619 (8.7507/2.3055) mem 68106MB [2022-12-19 13:30:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][270/1519] eta 0:20:53 lr 0.000030 time 0.9299 (1.0039) model_time 0.9297 (1.0019) loss 0.9929 (0.9598) grad_norm 7.0747 (8.7739/2.2795) mem 68106MB [2022-12-19 13:30:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][280/1519] eta 0:20:43 lr 0.000030 time 0.9214 (1.0036) model_time 0.9212 (1.0017) loss 0.9853 (0.9622) grad_norm 11.6456 (8.7796/2.2603) mem 68106MB [2022-12-19 13:31:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][290/1519] eta 0:20:33 lr 0.000030 time 0.9273 (1.0033) model_time 0.9271 (1.0014) loss 0.6984 (0.9577) grad_norm 8.9618 (8.7602/2.2404) mem 68106MB [2022-12-19 13:31:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][300/1519] eta 0:20:23 lr 0.000030 time 0.9726 (1.0037) model_time 0.9723 (1.0019) loss 0.7839 (0.9567) grad_norm 6.9822 (8.7777/2.2458) mem 68106MB [2022-12-19 13:31:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][310/1519] eta 0:20:13 lr 0.000030 time 0.9257 (1.0037) model_time 0.9255 (1.0020) loss 0.7788 (0.9548) grad_norm 11.5852 (8.8012/2.2357) mem 68106MB [2022-12-19 13:31:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][320/1519] eta 0:20:03 lr 0.000030 time 0.9314 (1.0036) model_time 0.9312 (1.0018) loss 1.0151 (0.9561) grad_norm 8.1397 (8.7855/2.2088) mem 68106MB [2022-12-19 13:31:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][330/1519] eta 0:19:53 lr 0.000030 time 0.9294 (1.0036) model_time 0.9293 (1.0020) loss 0.8697 (0.9563) grad_norm 7.8797 (8.7895/2.1837) mem 68106MB [2022-12-19 13:31:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][340/1519] eta 0:19:43 lr 0.000030 time 0.9326 (1.0037) model_time 0.9323 (1.0021) loss 1.1712 (0.9553) grad_norm 7.5294 (8.7507/2.1690) mem 68106MB [2022-12-19 13:32:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][350/1519] eta 0:19:33 lr 0.000030 time 0.9269 (1.0038) model_time 0.9267 (1.0022) loss 0.9030 (0.9545) grad_norm 6.6695 (8.7340/2.1490) mem 68106MB [2022-12-19 13:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][360/1519] eta 0:19:23 lr 0.000030 time 0.9310 (1.0038) model_time 0.9308 (1.0022) loss 0.9700 (0.9563) grad_norm 7.0105 (8.7630/2.1504) mem 68106MB [2022-12-19 13:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][370/1519] eta 0:19:13 lr 0.000030 time 0.9302 (1.0038) model_time 0.9301 (1.0022) loss 1.2762 (0.9570) grad_norm 12.2088 (8.7914/2.1574) mem 68106MB [2022-12-19 13:32:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][380/1519] eta 0:19:03 lr 0.000030 time 0.9289 (1.0036) model_time 0.9287 (1.0021) loss 0.9852 (0.9582) grad_norm 6.2054 (8.8191/2.2613) mem 68106MB [2022-12-19 13:32:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][390/1519] eta 0:18:52 lr 0.000030 time 0.9298 (1.0034) model_time 0.9297 (1.0019) loss 0.7636 (0.9574) grad_norm 13.2355 (8.7856/2.2880) mem 68106MB [2022-12-19 13:32:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][400/1519] eta 0:18:42 lr 0.000030 time 0.9164 (1.0033) model_time 0.9162 (1.0018) loss 0.8291 (0.9566) grad_norm 6.8668 (8.7568/2.2853) mem 68106MB [2022-12-19 13:33:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][410/1519] eta 0:18:32 lr 0.000030 time 0.9350 (1.0031) model_time 0.9348 (1.0017) loss 0.7798 (0.9584) grad_norm 5.5426 (8.7330/2.2803) mem 68106MB [2022-12-19 13:33:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][420/1519] eta 0:18:22 lr 0.000030 time 0.9276 (1.0030) model_time 0.9274 (1.0016) loss 0.9154 (0.9585) grad_norm 9.0581 (8.7131/2.2655) mem 68106MB [2022-12-19 13:33:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][430/1519] eta 0:18:12 lr 0.000030 time 0.9327 (1.0029) model_time 0.9322 (1.0016) loss 1.0301 (0.9573) grad_norm 16.5618 (8.7246/2.3179) mem 68106MB [2022-12-19 13:33:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][440/1519] eta 0:18:02 lr 0.000030 time 0.9328 (1.0028) model_time 0.9326 (1.0014) loss 0.8134 (0.9560) grad_norm 8.8882 (8.7305/2.2973) mem 68106MB [2022-12-19 13:33:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][450/1519] eta 0:17:52 lr 0.000030 time 0.9363 (1.0029) model_time 0.9361 (1.0015) loss 0.8271 (0.9573) grad_norm 10.5452 (8.7394/2.2809) mem 68106MB [2022-12-19 13:33:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][460/1519] eta 0:17:42 lr 0.000030 time 0.9752 (1.0032) model_time 0.9748 (1.0019) loss 0.9780 (0.9557) grad_norm 9.7698 (8.7404/2.2685) mem 68106MB [2022-12-19 13:34:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][470/1519] eta 0:17:32 lr 0.000030 time 0.9182 (1.0035) model_time 0.9180 (1.0021) loss 0.7179 (0.9563) grad_norm 6.9939 (8.7606/2.2668) mem 68106MB [2022-12-19 13:34:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][480/1519] eta 0:17:23 lr 0.000030 time 0.9317 (1.0040) model_time 0.9316 (1.0028) loss 0.8558 (0.9560) grad_norm 9.1531 (8.7485/2.2475) mem 68106MB [2022-12-19 13:34:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][490/1519] eta 0:17:13 lr 0.000030 time 0.9301 (1.0040) model_time 0.9298 (1.0027) loss 1.0412 (0.9554) grad_norm 13.7884 (8.7639/2.2498) mem 68106MB [2022-12-19 13:34:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][500/1519] eta 0:17:02 lr 0.000030 time 0.9284 (1.0039) model_time 0.9283 (1.0026) loss 0.8950 (0.9543) grad_norm 8.2024 (8.7768/2.2512) mem 68106MB [2022-12-19 13:34:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][510/1519] eta 0:16:52 lr 0.000030 time 0.9317 (1.0039) model_time 0.9316 (1.0026) loss 0.8791 (0.9538) grad_norm 6.9034 (8.7871/2.2430) mem 68106MB [2022-12-19 13:34:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][520/1519] eta 0:16:42 lr 0.000030 time 0.9333 (1.0038) model_time 0.9332 (1.0026) loss 0.8646 (0.9535) grad_norm 11.1903 (8.8039/2.2417) mem 68106MB [2022-12-19 13:35:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][530/1519] eta 0:16:32 lr 0.000030 time 0.9310 (1.0038) model_time 0.9309 (1.0026) loss 0.7628 (0.9542) grad_norm 6.6971 (8.8340/2.2725) mem 68106MB [2022-12-19 13:35:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][540/1519] eta 0:16:22 lr 0.000030 time 0.9245 (1.0037) model_time 0.9244 (1.0025) loss 0.7952 (0.9525) grad_norm 9.2271 (8.8238/2.2547) mem 68106MB [2022-12-19 13:35:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][550/1519] eta 0:16:12 lr 0.000030 time 0.9322 (1.0039) model_time 0.9320 (1.0027) loss 0.9157 (0.9514) grad_norm 10.9775 (8.8241/2.2400) mem 68106MB [2022-12-19 13:35:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][560/1519] eta 0:16:02 lr 0.000030 time 0.9316 (1.0038) model_time 0.9314 (1.0026) loss 1.1175 (0.9512) grad_norm 7.3174 (8.8141/2.2303) mem 68106MB [2022-12-19 13:35:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][570/1519] eta 0:15:52 lr 0.000030 time 0.9133 (1.0039) model_time 0.9131 (1.0027) loss 0.8276 (0.9508) grad_norm 8.7976 (8.8427/2.2391) mem 68106MB [2022-12-19 13:35:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][580/1519] eta 0:15:42 lr 0.000030 time 0.9317 (1.0039) model_time 0.9316 (1.0027) loss 0.8137 (0.9503) grad_norm 11.0726 (8.8328/2.2313) mem 68106MB [2022-12-19 13:36:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][590/1519] eta 0:15:32 lr 0.000030 time 0.9323 (1.0038) model_time 0.9321 (1.0027) loss 0.7271 (0.9498) grad_norm 7.3475 (8.8546/2.2338) mem 68106MB [2022-12-19 13:36:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][600/1519] eta 0:15:22 lr 0.000030 time 0.9300 (1.0037) model_time 0.9298 (1.0026) loss 0.8729 (0.9486) grad_norm 9.8568 (8.8343/2.2300) mem 68106MB [2022-12-19 13:36:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][610/1519] eta 0:15:12 lr 0.000030 time 0.9464 (1.0037) model_time 0.9463 (1.0026) loss 0.8954 (0.9502) grad_norm 8.6741 (8.9087/2.2881) mem 68106MB [2022-12-19 13:36:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][620/1519] eta 0:15:02 lr 0.000030 time 0.9311 (1.0038) model_time 0.9309 (1.0027) loss 0.7623 (0.9489) grad_norm 8.5045 (8.9147/2.2566) mem 68106MB [2022-12-19 13:36:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][630/1519] eta 0:14:52 lr 0.000030 time 0.9279 (1.0037) model_time 0.9278 (1.0026) loss 0.8336 (0.9489) grad_norm 6.2508 (8.9266/2.2691) mem 68106MB [2022-12-19 13:36:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][640/1519] eta 0:14:42 lr 0.000030 time 0.9829 (1.0037) model_time 0.9827 (1.0027) loss 1.1749 (0.9503) grad_norm 12.3739 (8.9093/2.2573) mem 68106MB [2022-12-19 13:37:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][650/1519] eta 0:14:32 lr 0.000030 time 0.9320 (1.0038) model_time 0.9318 (1.0028) loss 1.0956 (0.9498) grad_norm 9.1117 (8.9193/2.2614) mem 68106MB [2022-12-19 13:37:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][660/1519] eta 0:14:22 lr 0.000030 time 0.9374 (1.0042) model_time 0.9373 (1.0032) loss 0.7086 (0.9493) grad_norm 8.3618 (8.9002/2.2344) mem 68106MB [2022-12-19 13:37:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][670/1519] eta 0:14:12 lr 0.000030 time 0.9462 (1.0045) model_time 0.9461 (1.0034) loss 1.2876 (0.9503) grad_norm 7.3180 (8.9051/2.2409) mem 68106MB [2022-12-19 13:37:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][680/1519] eta 0:14:02 lr 0.000030 time 0.9294 (1.0044) model_time 0.9293 (1.0034) loss 1.0137 (0.9506) grad_norm 7.7304 (8.8632/2.2216) mem 68106MB [2022-12-19 13:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][690/1519] eta 0:13:52 lr 0.000030 time 0.9310 (1.0043) model_time 0.9308 (1.0033) loss 0.8049 (0.9501) grad_norm 8.3822 (8.8550/2.2135) mem 68106MB [2022-12-19 13:37:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][700/1519] eta 0:13:42 lr 0.000030 time 0.9332 (1.0042) model_time 0.9331 (1.0032) loss 1.0061 (0.9500) grad_norm 9.3028 (8.8799/2.2120) mem 68106MB [2022-12-19 13:38:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][710/1519] eta 0:13:32 lr 0.000030 time 0.9290 (1.0042) model_time 0.9289 (1.0032) loss 1.1912 (0.9511) grad_norm 12.1154 (8.9161/2.2200) mem 68106MB [2022-12-19 13:38:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][720/1519] eta 0:13:22 lr 0.000030 time 0.9340 (1.0041) model_time 0.9339 (1.0031) loss 0.9123 (0.9508) grad_norm 10.4232 (8.9195/2.1992) mem 68106MB [2022-12-19 13:38:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][730/1519] eta 0:13:12 lr 0.000030 time 0.9721 (1.0041) model_time 0.9720 (1.0032) loss 0.7330 (0.9494) grad_norm 8.8519 (8.9209/2.1861) mem 68106MB [2022-12-19 13:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][740/1519] eta 0:13:02 lr 0.000030 time 0.9370 (1.0041) model_time 0.9369 (1.0031) loss 0.9027 (0.9498) grad_norm 8.4411 (8.9116/2.1513) mem 68106MB [2022-12-19 13:38:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][750/1519] eta 0:12:52 lr 0.000030 time 0.9316 (1.0040) model_time 0.9314 (1.0031) loss 0.7612 (0.9493) grad_norm 11.4216 (8.9182/2.1704) mem 68106MB [2022-12-19 13:38:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][760/1519] eta 0:12:42 lr 0.000030 time 0.9327 (1.0041) model_time 0.9325 (1.0032) loss 0.8415 (0.9496) grad_norm 8.9053 (8.9640/2.1844) mem 68106MB [2022-12-19 13:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][770/1519] eta 0:12:32 lr 0.000030 time 0.9336 (1.0041) model_time 0.9335 (1.0032) loss 0.8280 (0.9484) grad_norm 7.5531 (8.9559/2.1824) mem 68106MB [2022-12-19 13:39:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][780/1519] eta 0:12:22 lr 0.000030 time 0.9358 (1.0042) model_time 0.9357 (1.0033) loss 0.8768 (0.9477) grad_norm 8.8259 (8.9919/2.1965) mem 68106MB [2022-12-19 13:39:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][790/1519] eta 0:12:12 lr 0.000030 time 0.9303 (1.0043) model_time 0.9302 (1.0034) loss 1.2058 (0.9481) grad_norm 11.1102 (9.0191/2.1875) mem 68106MB [2022-12-19 13:39:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][800/1519] eta 0:12:02 lr 0.000030 time 0.9302 (1.0043) model_time 0.9301 (1.0034) loss 0.9086 (0.9476) grad_norm 6.5228 (8.9954/2.2012) mem 68106MB [2022-12-19 13:39:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][810/1519] eta 0:11:52 lr 0.000030 time 0.9358 (1.0043) model_time 0.9357 (1.0033) loss 0.8543 (0.9474) grad_norm 7.9843 (9.0398/2.2690) mem 68106MB [2022-12-19 13:39:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][820/1519] eta 0:11:41 lr 0.000030 time 0.9322 (1.0042) model_time 0.9321 (1.0033) loss 0.9041 (0.9466) grad_norm 9.2166 (9.0385/2.2561) mem 68106MB [2022-12-19 13:40:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][830/1519] eta 0:11:31 lr 0.000030 time 0.9283 (1.0041) model_time 0.9281 (1.0032) loss 0.8598 (0.9468) grad_norm 9.2157 (9.0441/2.2528) mem 68106MB [2022-12-19 13:40:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][840/1519] eta 0:11:22 lr 0.000030 time 0.9295 (1.0044) model_time 0.9293 (1.0035) loss 1.0659 (0.9464) grad_norm 8.1637 (9.0522/2.2533) mem 68106MB [2022-12-19 13:40:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][850/1519] eta 0:11:11 lr 0.000030 time 0.9470 (1.0044) model_time 0.9469 (1.0035) loss 0.8280 (0.9462) grad_norm 8.6978 (9.0224/2.2315) mem 68106MB [2022-12-19 13:40:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][860/1519] eta 0:11:01 lr 0.000030 time 0.9272 (1.0045) model_time 0.9270 (1.0036) loss 1.5414 (0.9471) grad_norm 8.3743 (8.9968/2.2277) mem 68106MB [2022-12-19 13:40:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][870/1519] eta 0:10:51 lr 0.000030 time 0.9377 (1.0044) model_time 0.9376 (1.0036) loss 0.7252 (0.9481) grad_norm 8.9169 (8.9752/2.2282) mem 68106MB [2022-12-19 13:40:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][880/1519] eta 0:10:41 lr 0.000030 time 0.9362 (1.0044) model_time 0.9361 (1.0035) loss 1.1104 (0.9474) grad_norm 10.3184 (8.9608/2.2298) mem 68106MB [2022-12-19 13:41:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][890/1519] eta 0:10:31 lr 0.000030 time 0.9422 (1.0044) model_time 0.9421 (1.0035) loss 1.1538 (0.9473) grad_norm 9.0570 (8.9680/2.2269) mem 68106MB [2022-12-19 13:41:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][900/1519] eta 0:10:21 lr 0.000030 time 0.9304 (1.0043) model_time 0.9303 (1.0035) loss 1.1014 (0.9469) grad_norm 8.5143 (8.9411/2.2198) mem 68106MB [2022-12-19 13:41:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][910/1519] eta 0:10:11 lr 0.000030 time 0.9782 (1.0043) model_time 0.9781 (1.0035) loss 0.9267 (0.9465) grad_norm 6.3072 (8.9337/2.2252) mem 68106MB [2022-12-19 13:41:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][920/1519] eta 0:10:01 lr 0.000030 time 0.9422 (1.0043) model_time 0.9421 (1.0034) loss 0.8405 (0.9460) grad_norm 9.3109 (8.9299/2.2284) mem 68106MB [2022-12-19 13:41:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][930/1519] eta 0:09:51 lr 0.000030 time 1.0229 (1.0043) model_time 1.0228 (1.0035) loss 1.2164 (0.9459) grad_norm 10.0476 (8.9243/2.2293) mem 68106MB [2022-12-19 13:41:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][940/1519] eta 0:09:41 lr 0.000030 time 0.9392 (1.0043) model_time 0.9391 (1.0034) loss 0.7415 (0.9463) grad_norm 8.4825 (8.9494/2.2316) mem 68106MB [2022-12-19 13:42:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][950/1519] eta 0:09:31 lr 0.000030 time 0.9425 (1.0043) model_time 0.9424 (1.0035) loss 0.7398 (0.9459) grad_norm 6.9601 (8.9739/2.2528) mem 68106MB [2022-12-19 13:42:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][960/1519] eta 0:09:21 lr 0.000030 time 0.9910 (1.0044) model_time 0.9909 (1.0036) loss 0.9393 (0.9457) grad_norm 8.5901 (8.9635/2.2426) mem 68106MB [2022-12-19 13:42:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][970/1519] eta 0:09:11 lr 0.000030 time 0.9335 (1.0047) model_time 0.9334 (1.0038) loss 0.7677 (0.9455) grad_norm 8.4779 (8.9637/2.2370) mem 68106MB [2022-12-19 13:42:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][980/1519] eta 0:09:01 lr 0.000030 time 0.9256 (1.0046) model_time 0.9254 (1.0038) loss 0.8752 (0.9455) grad_norm 8.5635 (8.9697/2.1663) mem 68106MB [2022-12-19 13:42:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][990/1519] eta 0:08:51 lr 0.000030 time 0.9337 (1.0046) model_time 0.9335 (1.0038) loss 1.6232 (0.9451) grad_norm 7.0256 (8.9688/2.1366) mem 68106MB [2022-12-19 13:42:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1000/1519] eta 0:08:41 lr 0.000030 time 0.9346 (1.0045) model_time 0.9344 (1.0037) loss 0.8797 (0.9449) grad_norm 7.0228 (9.0001/2.1243) mem 68106MB [2022-12-19 13:43:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1010/1519] eta 0:08:31 lr 0.000030 time 0.9319 (1.0045) model_time 0.9318 (1.0037) loss 0.8030 (0.9454) grad_norm 14.1985 (9.0372/2.1327) mem 68106MB [2022-12-19 13:43:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1020/1519] eta 0:08:21 lr 0.000030 time 0.9289 (1.0045) model_time 0.9288 (1.0037) loss 0.9412 (0.9461) grad_norm 7.5523 (9.0375/2.1286) mem 68106MB [2022-12-19 13:43:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1030/1519] eta 0:08:11 lr 0.000030 time 0.9536 (1.0045) model_time 0.9534 (1.0037) loss 1.2877 (0.9465) grad_norm 8.4296 (9.0257/2.0706) mem 68106MB [2022-12-19 13:43:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1040/1519] eta 0:08:01 lr 0.000030 time 0.9346 (1.0044) model_time 0.9344 (1.0036) loss 1.1966 (0.9469) grad_norm 6.8875 (9.0220/2.0815) mem 68106MB [2022-12-19 13:43:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1050/1519] eta 0:07:51 lr 0.000030 time 0.9344 (1.0044) model_time 0.9343 (1.0036) loss 0.7808 (0.9467) grad_norm 7.0774 (9.0090/2.0861) mem 68106MB [2022-12-19 13:43:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1060/1519] eta 0:07:40 lr 0.000030 time 0.9317 (1.0043) model_time 0.9315 (1.0035) loss 1.2796 (0.9469) grad_norm 12.8798 (9.0276/2.0935) mem 68106MB [2022-12-19 13:44:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1070/1519] eta 0:07:30 lr 0.000030 time 0.9750 (1.0044) model_time 0.9748 (1.0036) loss 0.9191 (0.9471) grad_norm 6.9827 (9.0063/2.0838) mem 68106MB [2022-12-19 13:44:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1080/1519] eta 0:07:20 lr 0.000030 time 0.9310 (1.0045) model_time 0.9308 (1.0037) loss 1.0271 (0.9466) grad_norm 7.4994 (9.0016/2.0928) mem 68106MB [2022-12-19 13:44:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1090/1519] eta 0:07:10 lr 0.000030 time 0.9371 (1.0046) model_time 0.9369 (1.0038) loss 1.1557 (0.9468) grad_norm 9.1444 (9.0053/2.0896) mem 68106MB [2022-12-19 13:44:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1100/1519] eta 0:07:00 lr 0.000030 time 0.9364 (1.0046) model_time 0.9363 (1.0038) loss 0.9503 (0.9469) grad_norm 9.3444 (8.9866/2.0781) mem 68106MB [2022-12-19 13:44:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1110/1519] eta 0:06:50 lr 0.000030 time 0.9363 (1.0046) model_time 0.9362 (1.0038) loss 0.9301 (0.9474) grad_norm 8.4910 (9.0037/2.1056) mem 68106MB [2022-12-19 13:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1120/1519] eta 0:06:40 lr 0.000030 time 0.9357 (1.0045) model_time 0.9355 (1.0038) loss 0.8261 (0.9481) grad_norm 7.8860 (8.9823/2.0985) mem 68106MB [2022-12-19 13:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1130/1519] eta 0:06:30 lr 0.000030 time 0.9360 (1.0045) model_time 0.9358 (1.0038) loss 0.7836 (0.9471) grad_norm 9.2179 (8.9853/2.0723) mem 68106MB [2022-12-19 13:45:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1140/1519] eta 0:06:20 lr 0.000030 time 0.9901 (1.0045) model_time 0.9899 (1.0038) loss 0.9067 (0.9466) grad_norm 9.6218 (8.9967/2.0691) mem 68106MB [2022-12-19 13:45:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1150/1519] eta 0:06:10 lr 0.000030 time 0.9317 (1.0045) model_time 0.9315 (1.0038) loss 1.1216 (0.9469) grad_norm 12.2311 (9.0354/2.0947) mem 68106MB [2022-12-19 13:45:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1160/1519] eta 0:06:00 lr 0.000030 time 0.9338 (1.0045) model_time 0.9337 (1.0038) loss 0.9041 (0.9469) grad_norm 6.6359 (9.0565/2.1009) mem 68106MB [2022-12-19 13:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1170/1519] eta 0:05:50 lr 0.000030 time 0.9692 (1.0045) model_time 0.9691 (1.0038) loss 1.0847 (0.9473) grad_norm 12.6421 (9.0685/2.1194) mem 68106MB [2022-12-19 13:45:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1180/1519] eta 0:05:40 lr 0.000030 time 0.9339 (1.0045) model_time 0.9337 (1.0038) loss 0.7140 (0.9465) grad_norm 8.1086 (9.0707/2.1107) mem 68106MB [2022-12-19 13:46:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1190/1519] eta 0:05:30 lr 0.000030 time 0.9383 (1.0045) model_time 0.9381 (1.0038) loss 0.8551 (0.9464) grad_norm 7.7015 (9.0338/2.1024) mem 68106MB [2022-12-19 13:46:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1200/1519] eta 0:05:20 lr 0.000030 time 0.9380 (1.0045) model_time 0.9379 (1.0038) loss 1.1768 (0.9468) grad_norm 6.6386 (9.0525/2.1184) mem 68106MB [2022-12-19 13:46:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1210/1519] eta 0:05:10 lr 0.000030 time 0.9352 (1.0045) model_time 0.9351 (1.0037) loss 1.2354 (0.9462) grad_norm 11.9295 (8.9765/2.0252) mem 68106MB [2022-12-19 13:46:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1220/1519] eta 0:05:00 lr 0.000030 time 0.9332 (1.0045) model_time 0.9330 (1.0037) loss 0.9370 (0.9461) grad_norm 7.0674 (8.9536/2.0326) mem 68106MB [2022-12-19 13:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1230/1519] eta 0:04:50 lr 0.000030 time 0.9401 (1.0045) model_time 0.9400 (1.0038) loss 1.2811 (0.9465) grad_norm 10.8357 (8.9615/2.0297) mem 68106MB [2022-12-19 13:46:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1240/1519] eta 0:04:40 lr 0.000030 time 0.9405 (1.0045) model_time 0.9403 (1.0037) loss 0.9889 (0.9466) grad_norm 11.1580 (8.9682/2.0228) mem 68106MB [2022-12-19 13:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1250/1519] eta 0:04:30 lr 0.000030 time 0.9526 (1.0045) model_time 0.9524 (1.0038) loss 1.0314 (0.9471) grad_norm 10.4489 (8.9952/2.0219) mem 68106MB [2022-12-19 13:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1260/1519] eta 0:04:20 lr 0.000030 time 0.9265 (1.0045) model_time 0.9263 (1.0038) loss 1.1183 (0.9465) grad_norm 6.7435 (9.0025/2.0502) mem 68106MB [2022-12-19 13:47:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1270/1519] eta 0:04:10 lr 0.000030 time 0.9330 (1.0045) model_time 0.9329 (1.0038) loss 1.4428 (0.9467) grad_norm 8.7679 (9.0141/2.0493) mem 68106MB [2022-12-19 13:47:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1280/1519] eta 0:04:00 lr 0.000030 time 0.9378 (1.0046) model_time 0.9376 (1.0039) loss 0.8849 (0.9470) grad_norm 11.8353 (9.0746/2.0919) mem 68106MB [2022-12-19 13:47:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1290/1519] eta 0:03:50 lr 0.000030 time 1.0368 (1.0047) model_time 1.0367 (1.0040) loss 1.0074 (0.9469) grad_norm 8.0603 (9.1064/2.1945) mem 68106MB [2022-12-19 13:47:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1300/1519] eta 0:03:40 lr 0.000030 time 0.9320 (1.0046) model_time 0.9319 (1.0039) loss 1.3136 (0.9471) grad_norm 25.5574 (9.1500/2.3931) mem 68106MB [2022-12-19 13:48:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1310/1519] eta 0:03:29 lr 0.000030 time 0.9313 (1.0046) model_time 0.9312 (1.0039) loss 0.9676 (0.9465) grad_norm 6.2585 (9.1117/2.3863) mem 68106MB [2022-12-19 13:48:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1320/1519] eta 0:03:19 lr 0.000030 time 0.9296 (1.0045) model_time 0.9295 (1.0038) loss 0.8096 (0.9469) grad_norm 8.0218 (9.0966/2.3645) mem 68106MB [2022-12-19 13:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1330/1519] eta 0:03:09 lr 0.000030 time 0.9323 (1.0045) model_time 0.9322 (1.0038) loss 1.0430 (0.9481) grad_norm 8.3116 (9.1241/2.3843) mem 68106MB [2022-12-19 13:48:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1340/1519] eta 0:02:59 lr 0.000030 time 0.9347 (1.0045) model_time 0.9346 (1.0038) loss 0.9380 (0.9479) grad_norm 10.5074 (9.1387/2.3860) mem 68106MB [2022-12-19 13:48:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1350/1519] eta 0:02:49 lr 0.000030 time 0.9304 (1.0045) model_time 0.9302 (1.0038) loss 0.8415 (0.9475) grad_norm 9.3747 (9.0921/2.3648) mem 68106MB [2022-12-19 13:48:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1360/1519] eta 0:02:39 lr 0.000030 time 0.9400 (1.0045) model_time 0.9398 (1.0038) loss 0.7597 (0.9476) grad_norm 7.2931 (9.0738/2.3541) mem 68106MB [2022-12-19 13:49:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1370/1519] eta 0:02:29 lr 0.000030 time 0.9366 (1.0044) model_time 0.9364 (1.0038) loss 0.9474 (0.9484) grad_norm 7.6369 (9.0760/2.3482) mem 68106MB [2022-12-19 13:49:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1380/1519] eta 0:02:19 lr 0.000030 time 0.9404 (1.0044) model_time 0.9403 (1.0038) loss 0.9281 (0.9481) grad_norm 12.1345 (9.0722/2.3374) mem 68106MB [2022-12-19 13:49:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1390/1519] eta 0:02:09 lr 0.000030 time 0.9365 (1.0045) model_time 0.9364 (1.0038) loss 0.7409 (0.9481) grad_norm 6.5240 (9.0477/2.3443) mem 68106MB [2022-12-19 13:49:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1400/1519] eta 0:01:59 lr 0.000030 time 0.9306 (1.0044) model_time 0.9301 (1.0038) loss 0.8464 (0.9477) grad_norm 5.9852 (9.0604/2.3346) mem 68106MB [2022-12-19 13:49:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1410/1519] eta 0:01:49 lr 0.000030 time 1.0449 (1.0045) model_time 1.0448 (1.0038) loss 0.8645 (0.9483) grad_norm 7.9554 (9.0085/2.2602) mem 68106MB [2022-12-19 13:49:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1420/1519] eta 0:01:39 lr 0.000030 time 0.9331 (1.0045) model_time 0.9330 (1.0038) loss 0.9638 (0.9484) grad_norm 9.5980 (9.0164/2.2570) mem 68106MB [2022-12-19 13:50:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1430/1519] eta 0:01:29 lr 0.000030 time 0.9210 (1.0045) model_time 0.9209 (1.0038) loss 0.8443 (0.9482) grad_norm 6.3267 (9.0116/2.2626) mem 68106MB [2022-12-19 13:50:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1440/1519] eta 0:01:19 lr 0.000030 time 0.9154 (1.0045) model_time 0.9152 (1.0038) loss 0.9900 (0.9479) grad_norm 6.3779 (9.0014/2.2631) mem 68106MB [2022-12-19 13:50:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1450/1519] eta 0:01:09 lr 0.000030 time 0.9393 (1.0045) model_time 0.9392 (1.0038) loss 0.8297 (0.9476) grad_norm 9.8813 (8.9836/2.2729) mem 68106MB [2022-12-19 13:50:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1460/1519] eta 0:00:59 lr 0.000030 time 0.9255 (1.0046) model_time 0.9253 (1.0039) loss 1.2525 (0.9478) grad_norm 7.3565 (8.9972/2.2719) mem 68106MB [2022-12-19 13:50:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1470/1519] eta 0:00:49 lr 0.000030 time 1.0046 (1.0047) model_time 1.0045 (1.0041) loss 0.9430 (0.9477) grad_norm 6.7208 (8.9950/2.2788) mem 68106MB [2022-12-19 13:50:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1480/1519] eta 0:00:39 lr 0.000030 time 0.9299 (1.0047) model_time 0.9298 (1.0040) loss 0.7937 (0.9473) grad_norm 7.4668 (9.0013/2.2701) mem 68106MB [2022-12-19 13:51:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1490/1519] eta 0:00:29 lr 0.000030 time 0.9294 (1.0047) model_time 0.9292 (1.0040) loss 0.8115 (0.9477) grad_norm 9.8701 (9.0001/2.2679) mem 68106MB [2022-12-19 13:51:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1500/1519] eta 0:00:19 lr 0.000030 time 0.9340 (1.0046) model_time 0.9339 (1.0040) loss 1.1182 (0.9484) grad_norm 8.5655 (9.0378/2.2662) mem 68106MB [2022-12-19 13:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [21/100][1510/1519] eta 0:00:09 lr 0.000030 time 0.9309 (1.0046) model_time 0.9304 (1.0040) loss 0.9066 (0.9486) grad_norm 6.9192 (9.0206/2.2613) mem 68106MB [2022-12-19 13:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 21 training takes 0:25:25 [2022-12-19 13:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_21.pth saving...... [2022-12-19 13:52:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_21.pth saved !!! [2022-12-19 13:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.710 (0.710) Loss 0.5446 (0.5446) Acc@1 89.236 (89.236) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 13:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.333) Loss 0.5554 (0.5234) Acc@1 89.583 (90.593) Acc@5 97.569 (98.264) Mem 68106MB [2022-12-19 13:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.300 (0.317) Loss 0.4796 (0.5235) Acc@1 91.667 (90.526) Acc@5 98.958 (98.099) Mem 68106MB [2022-12-19 13:52:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.311) Loss 0.6403 (0.5296) Acc@1 87.847 (90.222) Acc@5 97.569 (98.073) Mem 68106MB [2022-12-19 13:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.308) Loss 0.5005 (0.5209) Acc@1 89.583 (90.337) Acc@5 98.611 (98.179) Mem 68106MB [2022-12-19 13:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.310 (0.306) Loss 0.5156 (0.5179) Acc@1 87.500 (90.298) Acc@5 99.306 (98.230) Mem 68106MB [2022-12-19 13:52:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.306) Loss 0.5539 (0.5187) Acc@1 87.500 (90.278) Acc@5 98.264 (98.258) Mem 68106MB [2022-12-19 13:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.301 (0.305) Loss 0.5713 (0.5206) Acc@1 90.278 (90.234) Acc@5 98.611 (98.259) Mem 68106MB [2022-12-19 13:52:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.304) Loss 0.4720 (0.5184) Acc@1 91.319 (90.299) Acc@5 97.222 (98.285) Mem 68106MB [2022-12-19 13:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:21] * Acc@1 90.296 Acc@5 98.281 [2022-12-19 13:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 90.3% [2022-12-19 13:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 13:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 13:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.30% [2022-12-19 13:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][0/1519] eta 0:35:35 lr 0.000030 time 1.4061 (1.4061) model_time 0.9576 (0.9576) loss 0.8326 (0.8326) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 13:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][10/1519] eta 0:26:15 lr 0.000030 time 0.9786 (1.0443) model_time 0.9785 (1.0031) loss 1.0301 (0.9648) grad_norm 8.2806 (8.3013/1.0219) mem 68106MB [2022-12-19 13:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][20/1519] eta 0:25:36 lr 0.000030 time 0.9311 (1.0253) model_time 0.9309 (1.0035) loss 0.9022 (0.9642) grad_norm 11.3067 (8.8142/1.2221) mem 68106MB [2022-12-19 13:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][30/1519] eta 0:25:17 lr 0.000030 time 0.9398 (1.0190) model_time 0.9397 (1.0041) loss 0.8120 (0.9457) grad_norm 8.6559 (8.7428/1.1283) mem 68106MB [2022-12-19 13:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][40/1519] eta 0:25:07 lr 0.000030 time 1.0982 (1.0190) model_time 1.0981 (1.0076) loss 0.8632 (0.9319) grad_norm 8.3716 (9.1110/2.0673) mem 68106MB [2022-12-19 13:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][50/1519] eta 0:24:58 lr 0.000030 time 0.9281 (1.0203) model_time 0.9276 (1.0110) loss 1.3172 (0.9553) grad_norm 8.1430 (9.0516/1.9089) mem 68106MB [2022-12-19 13:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][60/1519] eta 0:24:45 lr 0.000030 time 1.0150 (1.0185) model_time 1.0148 (1.0107) loss 0.8045 (0.9567) grad_norm 8.2805 (9.0025/1.8307) mem 68106MB [2022-12-19 13:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][70/1519] eta 0:24:32 lr 0.000030 time 0.9376 (1.0163) model_time 0.9374 (1.0095) loss 1.0087 (0.9572) grad_norm 11.0433 (8.8795/1.8255) mem 68106MB [2022-12-19 13:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][80/1519] eta 0:24:22 lr 0.000030 time 0.9230 (1.0162) model_time 0.9229 (1.0102) loss 0.6795 (0.9473) grad_norm 6.6382 (8.7110/1.8174) mem 68106MB [2022-12-19 13:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][90/1519] eta 0:24:09 lr 0.000030 time 0.9266 (1.0142) model_time 0.9265 (1.0088) loss 0.9567 (0.9441) grad_norm 11.8704 (8.7385/1.8074) mem 68106MB [2022-12-19 13:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][100/1519] eta 0:23:57 lr 0.000030 time 0.9301 (1.0127) model_time 0.9299 (1.0079) loss 0.7511 (0.9436) grad_norm 8.8705 (8.6100/1.7976) mem 68106MB [2022-12-19 13:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][110/1519] eta 0:23:46 lr 0.000030 time 0.9411 (1.0122) model_time 0.9410 (1.0077) loss 1.0738 (0.9386) grad_norm 11.6641 (8.6122/1.7952) mem 68106MB [2022-12-19 13:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][120/1519] eta 0:23:35 lr 0.000030 time 0.9934 (1.0116) model_time 0.9933 (1.0075) loss 0.7609 (0.9377) grad_norm 6.9763 (8.6717/1.7897) mem 68106MB [2022-12-19 13:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][130/1519] eta 0:23:26 lr 0.000030 time 0.9241 (1.0123) model_time 0.9239 (1.0084) loss 1.0171 (0.9403) grad_norm 9.3248 (8.6054/1.7527) mem 68106MB [2022-12-19 13:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][140/1519] eta 0:23:14 lr 0.000030 time 0.9289 (1.0112) model_time 0.9287 (1.0076) loss 0.7720 (0.9414) grad_norm 6.6428 (8.5272/1.7374) mem 68106MB [2022-12-19 13:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][150/1519] eta 0:23:03 lr 0.000030 time 0.9258 (1.0108) model_time 0.9256 (1.0074) loss 0.8789 (0.9440) grad_norm 8.3224 (8.4699/1.7156) mem 68106MB [2022-12-19 13:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][160/1519] eta 0:22:52 lr 0.000030 time 0.9279 (1.0101) model_time 0.9277 (1.0069) loss 1.0185 (0.9459) grad_norm 7.2696 (8.4368/1.6780) mem 68106MB [2022-12-19 13:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][170/1519] eta 0:22:42 lr 0.000030 time 0.9235 (1.0097) model_time 0.9234 (1.0066) loss 0.8947 (0.9437) grad_norm 8.1793 (8.4161/1.6608) mem 68106MB [2022-12-19 13:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][180/1519] eta 0:22:31 lr 0.000030 time 0.9313 (1.0092) model_time 0.9311 (1.0063) loss 0.9819 (0.9421) grad_norm 9.8070 (8.5365/1.6938) mem 68106MB [2022-12-19 13:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][190/1519] eta 0:22:20 lr 0.000030 time 0.9254 (1.0086) model_time 0.9252 (1.0058) loss 0.8770 (0.9408) grad_norm 14.5814 (8.5591/1.7889) mem 68106MB [2022-12-19 13:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][200/1519] eta 0:22:09 lr 0.000030 time 0.9298 (1.0083) model_time 0.9296 (1.0057) loss 0.6958 (0.9376) grad_norm 12.6629 (8.6289/1.8213) mem 68106MB [2022-12-19 13:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][210/1519] eta 0:21:59 lr 0.000030 time 0.9121 (1.0078) model_time 0.9120 (1.0053) loss 0.7954 (0.9357) grad_norm 8.7818 (8.6380/1.8141) mem 68106MB [2022-12-19 13:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][220/1519] eta 0:21:49 lr 0.000030 time 0.9320 (1.0079) model_time 0.9318 (1.0055) loss 0.8472 (0.9354) grad_norm 8.4497 (8.6975/1.9031) mem 68106MB [2022-12-19 13:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][230/1519] eta 0:21:38 lr 0.000030 time 0.9324 (1.0076) model_time 0.9323 (1.0052) loss 1.0267 (0.9345) grad_norm 6.9034 (8.7597/1.9543) mem 68106MB [2022-12-19 13:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][240/1519] eta 0:21:28 lr 0.000030 time 0.9258 (1.0075) model_time 0.9256 (1.0052) loss 0.9383 (0.9358) grad_norm 8.3470 (8.7298/1.9252) mem 68106MB [2022-12-19 13:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][250/1519] eta 0:21:18 lr 0.000030 time 0.9268 (1.0075) model_time 0.9266 (1.0053) loss 0.8589 (0.9345) grad_norm 12.3653 (8.8235/1.9635) mem 68106MB [2022-12-19 13:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][260/1519] eta 0:21:10 lr 0.000030 time 0.9318 (1.0092) model_time 0.9316 (1.0071) loss 1.3866 (0.9371) grad_norm 7.3852 (8.8621/2.0106) mem 68106MB [2022-12-19 13:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][270/1519] eta 0:21:00 lr 0.000030 time 0.9250 (1.0090) model_time 0.9249 (1.0069) loss 1.1711 (0.9391) grad_norm 13.0278 (8.8820/2.0196) mem 68106MB [2022-12-19 13:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][280/1519] eta 0:20:49 lr 0.000030 time 0.9203 (1.0086) model_time 0.9201 (1.0066) loss 1.1159 (0.9388) grad_norm 13.4989 (8.9324/2.0582) mem 68106MB [2022-12-19 13:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][290/1519] eta 0:20:39 lr 0.000030 time 0.9242 (1.0084) model_time 0.9241 (1.0065) loss 0.7890 (0.9408) grad_norm 8.4164 (8.9542/2.1001) mem 68106MB [2022-12-19 13:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][300/1519] eta 0:20:29 lr 0.000030 time 0.9228 (1.0087) model_time 0.9227 (1.0068) loss 0.9528 (0.9394) grad_norm 8.8431 (8.9339/2.0808) mem 68106MB [2022-12-19 13:58:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][310/1519] eta 0:20:19 lr 0.000030 time 0.9314 (1.0084) model_time 0.9312 (1.0066) loss 1.0044 (0.9412) grad_norm 11.3715 (8.9734/2.0741) mem 68106MB [2022-12-19 13:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][320/1519] eta 0:20:08 lr 0.000030 time 0.9339 (1.0080) model_time 0.9337 (1.0063) loss 0.7468 (0.9381) grad_norm 7.0536 (8.9368/2.0694) mem 68106MB [2022-12-19 13:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][330/1519] eta 0:19:58 lr 0.000030 time 0.9231 (1.0078) model_time 0.9230 (1.0061) loss 0.8648 (0.9370) grad_norm 7.7435 (8.9323/2.0527) mem 68106MB [2022-12-19 13:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][340/1519] eta 0:19:47 lr 0.000030 time 0.9244 (1.0076) model_time 0.9242 (1.0059) loss 0.8832 (0.9379) grad_norm 7.8469 (8.9849/2.1126) mem 68106MB [2022-12-19 13:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][350/1519] eta 0:19:37 lr 0.000030 time 0.9217 (1.0075) model_time 0.9215 (1.0058) loss 0.7585 (0.9363) grad_norm 7.5541 (8.9420/2.1004) mem 68106MB [2022-12-19 13:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][360/1519] eta 0:19:28 lr 0.000030 time 0.9285 (1.0079) model_time 0.9284 (1.0063) loss 1.2538 (0.9353) grad_norm 10.6238 (8.9372/2.0915) mem 68106MB [2022-12-19 13:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][370/1519] eta 0:19:17 lr 0.000030 time 0.9254 (1.0077) model_time 0.9253 (1.0062) loss 0.9246 (0.9346) grad_norm 7.0478 (8.8974/2.0792) mem 68106MB [2022-12-19 13:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][380/1519] eta 0:19:08 lr 0.000030 time 0.9239 (1.0082) model_time 0.9238 (1.0066) loss 1.0115 (0.9377) grad_norm 9.4770 (8.8814/2.0978) mem 68106MB [2022-12-19 13:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][390/1519] eta 0:18:58 lr 0.000030 time 0.9220 (1.0083) model_time 0.9219 (1.0068) loss 0.7008 (0.9364) grad_norm 8.1298 (8.8700/2.0862) mem 68106MB [2022-12-19 13:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][400/1519] eta 0:18:48 lr 0.000030 time 0.9322 (1.0080) model_time 0.9320 (1.0066) loss 1.0914 (0.9349) grad_norm 8.9212 (8.8502/2.0686) mem 68106MB [2022-12-19 13:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][410/1519] eta 0:18:37 lr 0.000030 time 0.9191 (1.0078) model_time 0.9190 (1.0064) loss 0.9034 (0.9347) grad_norm 8.4191 (8.8407/2.0504) mem 68106MB [2022-12-19 13:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][420/1519] eta 0:18:27 lr 0.000030 time 0.9211 (1.0078) model_time 0.9210 (1.0063) loss 0.9089 (0.9349) grad_norm 7.5990 (8.8110/2.0386) mem 68106MB [2022-12-19 14:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][430/1519] eta 0:18:17 lr 0.000030 time 0.9326 (1.0075) model_time 0.9325 (1.0061) loss 0.8209 (0.9351) grad_norm 9.3081 (8.7923/2.0458) mem 68106MB [2022-12-19 14:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][440/1519] eta 0:18:07 lr 0.000030 time 0.9155 (1.0076) model_time 0.9154 (1.0062) loss 0.7607 (0.9357) grad_norm 15.2523 (8.8295/2.0737) mem 68106MB [2022-12-19 14:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][450/1519] eta 0:17:56 lr 0.000030 time 0.9226 (1.0074) model_time 0.9225 (1.0060) loss 1.0147 (0.9352) grad_norm 8.9569 (8.8559/2.0742) mem 68106MB [2022-12-19 14:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][460/1519] eta 0:17:46 lr 0.000030 time 0.9316 (1.0071) model_time 0.9315 (1.0058) loss 1.3475 (0.9347) grad_norm 7.1185 (8.8187/2.0672) mem 68106MB [2022-12-19 14:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][470/1519] eta 0:17:36 lr 0.000030 time 0.9361 (1.0071) model_time 0.9359 (1.0058) loss 1.2975 (0.9353) grad_norm 8.9122 (8.8088/2.0527) mem 68106MB [2022-12-19 14:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][480/1519] eta 0:17:26 lr 0.000030 time 0.9726 (1.0071) model_time 0.9725 (1.0058) loss 1.0297 (0.9361) grad_norm 10.6875 (8.8446/2.0583) mem 68106MB [2022-12-19 14:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][490/1519] eta 0:17:16 lr 0.000030 time 0.9253 (1.0069) model_time 0.9252 (1.0056) loss 0.8084 (0.9348) grad_norm 11.6323 (8.8742/2.0884) mem 68106MB [2022-12-19 14:01:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][500/1519] eta 0:17:05 lr 0.000030 time 0.9317 (1.0067) model_time 0.9316 (1.0055) loss 1.1398 (0.9378) grad_norm 7.5704 (8.8700/2.0725) mem 68106MB [2022-12-19 14:01:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][510/1519] eta 0:16:55 lr 0.000030 time 0.9067 (1.0067) model_time 0.9066 (1.0055) loss 0.9219 (0.9355) grad_norm 11.2897 (8.8800/2.0690) mem 68106MB [2022-12-19 14:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][520/1519] eta 0:16:45 lr 0.000030 time 0.9277 (1.0066) model_time 0.9276 (1.0054) loss 0.9916 (0.9341) grad_norm 8.7283 (8.8550/2.0599) mem 68106MB [2022-12-19 14:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][530/1519] eta 0:16:35 lr 0.000030 time 0.9369 (1.0066) model_time 0.9368 (1.0054) loss 0.8370 (0.9341) grad_norm 8.2039 (8.8467/2.0542) mem 68106MB [2022-12-19 14:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][540/1519] eta 0:16:25 lr 0.000030 time 0.9282 (1.0067) model_time 0.9281 (1.0055) loss 1.1544 (0.9340) grad_norm 11.4644 (8.8438/2.0495) mem 68106MB [2022-12-19 14:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][550/1519] eta 0:16:15 lr 0.000030 time 0.9331 (1.0067) model_time 0.9330 (1.0055) loss 1.2599 (0.9347) grad_norm 7.3168 (8.8488/2.0485) mem 68106MB [2022-12-19 14:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][560/1519] eta 0:16:05 lr 0.000030 time 0.9206 (1.0068) model_time 0.9204 (1.0056) loss 0.6997 (0.9352) grad_norm 6.8848 (8.8281/2.0424) mem 68106MB [2022-12-19 14:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][570/1519] eta 0:15:55 lr 0.000030 time 0.9307 (1.0069) model_time 0.9305 (1.0058) loss 1.0373 (0.9352) grad_norm 7.3500 (8.8078/2.0362) mem 68106MB [2022-12-19 14:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][580/1519] eta 0:15:45 lr 0.000030 time 0.9310 (1.0068) model_time 0.9308 (1.0057) loss 1.1936 (0.9355) grad_norm 6.9042 (8.8359/2.0848) mem 68106MB [2022-12-19 14:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][590/1519] eta 0:15:35 lr 0.000030 time 0.9239 (1.0067) model_time 0.9238 (1.0055) loss 0.7690 (0.9348) grad_norm 6.7902 (8.8422/2.1014) mem 68106MB [2022-12-19 14:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][600/1519] eta 0:15:25 lr 0.000030 time 0.9322 (1.0066) model_time 0.9321 (1.0055) loss 0.7255 (0.9346) grad_norm 9.1298 (8.8253/2.0969) mem 68106MB [2022-12-19 14:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][610/1519] eta 0:15:14 lr 0.000030 time 0.9315 (1.0064) model_time 0.9314 (1.0053) loss 0.8239 (0.9344) grad_norm 6.7590 (8.8149/2.0999) mem 68106MB [2022-12-19 14:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][620/1519] eta 0:15:04 lr 0.000030 time 0.9220 (1.0064) model_time 0.9219 (1.0053) loss 0.7090 (0.9342) grad_norm 8.0469 (8.8163/2.1239) mem 68106MB [2022-12-19 14:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][630/1519] eta 0:14:54 lr 0.000030 time 0.9269 (1.0062) model_time 0.9267 (1.0051) loss 0.8044 (0.9345) grad_norm 7.2944 (8.8428/2.1949) mem 68106MB [2022-12-19 14:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][640/1519] eta 0:14:44 lr 0.000030 time 0.9359 (1.0061) model_time 0.9358 (1.0051) loss 0.8304 (0.9339) grad_norm 10.8078 (8.8247/2.1485) mem 68106MB [2022-12-19 14:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][650/1519] eta 0:14:34 lr 0.000030 time 0.9236 (1.0060) model_time 0.9234 (1.0049) loss 0.7540 (0.9342) grad_norm 7.8206 (8.8170/2.1591) mem 68106MB [2022-12-19 14:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][660/1519] eta 0:14:24 lr 0.000030 time 0.9788 (1.0059) model_time 0.9786 (1.0049) loss 0.7748 (0.9340) grad_norm 10.1007 (8.8134/2.1787) mem 68106MB [2022-12-19 14:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][670/1519] eta 0:14:14 lr 0.000030 time 0.9246 (1.0059) model_time 0.9245 (1.0049) loss 1.0120 (0.9344) grad_norm 8.4649 (8.8027/2.1766) mem 68106MB [2022-12-19 14:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][680/1519] eta 0:14:03 lr 0.000030 time 0.9313 (1.0058) model_time 0.9311 (1.0048) loss 0.7143 (0.9338) grad_norm 6.6523 (8.8497/2.1981) mem 68106MB [2022-12-19 14:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][690/1519] eta 0:13:53 lr 0.000030 time 0.9295 (1.0057) model_time 0.9294 (1.0047) loss 1.0747 (0.9340) grad_norm 8.3451 (8.8643/2.2420) mem 68106MB [2022-12-19 14:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][700/1519] eta 0:13:43 lr 0.000030 time 0.9299 (1.0057) model_time 0.9297 (1.0047) loss 0.7355 (0.9344) grad_norm 7.7059 (8.8784/2.2448) mem 68106MB [2022-12-19 14:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][710/1519] eta 0:13:33 lr 0.000030 time 0.9292 (1.0057) model_time 0.9290 (1.0047) loss 1.5099 (0.9353) grad_norm 9.1487 (8.8889/2.2603) mem 68106MB [2022-12-19 14:04:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][720/1519] eta 0:13:23 lr 0.000030 time 0.9253 (1.0056) model_time 0.9251 (1.0046) loss 0.7776 (0.9349) grad_norm 10.4644 (8.8982/2.2710) mem 68106MB [2022-12-19 14:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][730/1519] eta 0:13:13 lr 0.000030 time 0.9335 (1.0056) model_time 0.9332 (1.0046) loss 0.7435 (0.9358) grad_norm 8.6938 (8.9176/2.2713) mem 68106MB [2022-12-19 14:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][740/1519] eta 0:13:03 lr 0.000030 time 0.9216 (1.0055) model_time 0.9215 (1.0046) loss 0.8908 (0.9358) grad_norm 14.8546 (8.9683/2.3010) mem 68106MB [2022-12-19 14:05:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][750/1519] eta 0:12:53 lr 0.000030 time 1.0245 (1.0056) model_time 1.0244 (1.0047) loss 0.9214 (0.9351) grad_norm 8.4800 (9.0149/2.3166) mem 68106MB [2022-12-19 14:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][760/1519] eta 0:12:43 lr 0.000030 time 0.9331 (1.0056) model_time 0.9330 (1.0047) loss 1.0001 (0.9344) grad_norm 5.5919 (9.0065/2.3245) mem 68106MB [2022-12-19 14:05:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][770/1519] eta 0:12:33 lr 0.000030 time 0.9284 (1.0055) model_time 0.9282 (1.0046) loss 0.7215 (0.9340) grad_norm 10.0221 (9.0235/2.3283) mem 68106MB [2022-12-19 14:05:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][780/1519] eta 0:12:23 lr 0.000030 time 0.9262 (1.0055) model_time 0.9260 (1.0046) loss 0.9053 (0.9338) grad_norm 6.0158 (8.9959/2.3444) mem 68106MB [2022-12-19 14:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][790/1519] eta 0:12:13 lr 0.000030 time 0.9318 (1.0055) model_time 0.9317 (1.0046) loss 0.9103 (0.9338) grad_norm 11.0661 (9.0274/2.3313) mem 68106MB [2022-12-19 14:06:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][800/1519] eta 0:12:02 lr 0.000030 time 0.9401 (1.0055) model_time 0.9400 (1.0046) loss 1.1668 (0.9339) grad_norm 7.6620 (9.0418/2.4284) mem 68106MB [2022-12-19 14:06:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][810/1519] eta 0:11:52 lr 0.000030 time 0.9460 (1.0055) model_time 0.9459 (1.0046) loss 0.7601 (0.9338) grad_norm 8.2173 (9.0202/2.4292) mem 68106MB [2022-12-19 14:06:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][820/1519] eta 0:11:42 lr 0.000030 time 0.9360 (1.0055) model_time 0.9358 (1.0046) loss 0.9594 (0.9336) grad_norm 5.6653 (9.0018/2.4242) mem 68106MB [2022-12-19 14:06:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][830/1519] eta 0:11:32 lr 0.000030 time 0.9287 (1.0055) model_time 0.9285 (1.0046) loss 0.8269 (0.9326) grad_norm 9.1083 (8.9746/2.4198) mem 68106MB [2022-12-19 14:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][840/1519] eta 0:11:22 lr 0.000030 time 0.9289 (1.0056) model_time 0.9288 (1.0048) loss 1.3837 (0.9328) grad_norm 9.5926 (8.9934/2.4223) mem 68106MB [2022-12-19 14:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][850/1519] eta 0:11:13 lr 0.000030 time 0.9261 (1.0061) model_time 0.9260 (1.0052) loss 0.9540 (0.9327) grad_norm 12.9765 (8.9644/2.4149) mem 68106MB [2022-12-19 14:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][860/1519] eta 0:11:03 lr 0.000030 time 0.9276 (1.0064) model_time 0.9274 (1.0055) loss 1.0694 (0.9328) grad_norm 7.8895 (8.9338/2.4078) mem 68106MB [2022-12-19 14:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][870/1519] eta 0:10:53 lr 0.000030 time 0.9289 (1.0064) model_time 0.9287 (1.0055) loss 1.0348 (0.9335) grad_norm 23.5399 (9.0086/2.5981) mem 68106MB [2022-12-19 14:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][880/1519] eta 0:10:43 lr 0.000030 time 0.9073 (1.0065) model_time 0.9071 (1.0056) loss 0.9093 (0.9330) grad_norm 5.9344 (9.0061/2.6074) mem 68106MB [2022-12-19 14:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][890/1519] eta 0:10:33 lr 0.000030 time 0.9237 (1.0065) model_time 0.9235 (1.0056) loss 1.1794 (0.9353) grad_norm 8.4632 (9.0306/2.6159) mem 68106MB [2022-12-19 14:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][900/1519] eta 0:10:22 lr 0.000030 time 0.9335 (1.0064) model_time 0.9334 (1.0056) loss 0.9309 (0.9350) grad_norm 6.3440 (9.0304/2.6188) mem 68106MB [2022-12-19 14:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][910/1519] eta 0:10:12 lr 0.000030 time 0.9255 (1.0064) model_time 0.9254 (1.0055) loss 0.7706 (0.9358) grad_norm 9.7649 (8.9944/2.6157) mem 68106MB [2022-12-19 14:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][920/1519] eta 0:10:02 lr 0.000030 time 0.9331 (1.0063) model_time 0.9330 (1.0054) loss 0.8090 (0.9355) grad_norm 7.8742 (9.0268/2.6351) mem 68106MB [2022-12-19 14:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][930/1519] eta 0:09:52 lr 0.000030 time 1.0302 (1.0065) model_time 1.0300 (1.0057) loss 1.1045 (0.9357) grad_norm 6.3069 (9.0471/2.6475) mem 68106MB [2022-12-19 14:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][940/1519] eta 0:09:42 lr 0.000030 time 0.9220 (1.0065) model_time 0.9217 (1.0057) loss 0.8946 (0.9365) grad_norm 7.0294 (9.0137/2.6182) mem 68106MB [2022-12-19 14:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][950/1519] eta 0:09:32 lr 0.000030 time 0.9350 (1.0065) model_time 0.9348 (1.0057) loss 1.0265 (0.9361) grad_norm 9.6469 (9.0209/2.6158) mem 68106MB [2022-12-19 14:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][960/1519] eta 0:09:22 lr 0.000030 time 0.9340 (1.0064) model_time 0.9338 (1.0056) loss 1.0504 (0.9359) grad_norm 8.1414 (9.0139/2.6124) mem 68106MB [2022-12-19 14:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][970/1519] eta 0:09:12 lr 0.000030 time 0.8958 (1.0064) model_time 0.8956 (1.0056) loss 1.2980 (0.9357) grad_norm 6.0984 (9.0205/2.6110) mem 68106MB [2022-12-19 14:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][980/1519] eta 0:09:02 lr 0.000030 time 0.9319 (1.0065) model_time 0.9317 (1.0057) loss 1.1160 (0.9367) grad_norm 7.6086 (9.0262/2.5896) mem 68106MB [2022-12-19 14:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][990/1519] eta 0:08:52 lr 0.000030 time 0.9316 (1.0065) model_time 0.9315 (1.0057) loss 1.0727 (0.9369) grad_norm 7.7567 (9.0082/2.5931) mem 68106MB [2022-12-19 14:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1000/1519] eta 0:08:42 lr 0.000030 time 0.9366 (1.0065) model_time 0.9365 (1.0057) loss 0.7545 (0.9360) grad_norm 5.8102 (9.0127/2.6010) mem 68106MB [2022-12-19 14:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1010/1519] eta 0:08:32 lr 0.000030 time 0.9750 (1.0065) model_time 0.9749 (1.0057) loss 1.0867 (0.9365) grad_norm 8.3279 (9.0478/2.6206) mem 68106MB [2022-12-19 14:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1020/1519] eta 0:08:22 lr 0.000030 time 0.9392 (1.0065) model_time 0.9390 (1.0057) loss 0.9330 (0.9365) grad_norm 7.7275 (9.0513/2.6170) mem 68106MB [2022-12-19 14:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1030/1519] eta 0:08:12 lr 0.000030 time 0.9253 (1.0065) model_time 0.9251 (1.0057) loss 0.6782 (0.9364) grad_norm 6.4620 (9.0563/2.6030) mem 68106MB [2022-12-19 14:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1040/1519] eta 0:08:02 lr 0.000030 time 0.9952 (1.0065) model_time 0.9950 (1.0057) loss 1.0240 (0.9366) grad_norm 8.1451 (9.0144/2.5798) mem 68106MB [2022-12-19 14:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1050/1519] eta 0:07:52 lr 0.000030 time 0.9329 (1.0065) model_time 0.9328 (1.0057) loss 0.8994 (0.9365) grad_norm 7.6461 (8.9713/2.5832) mem 68106MB [2022-12-19 14:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1060/1519] eta 0:07:41 lr 0.000030 time 0.9224 (1.0065) model_time 0.9223 (1.0057) loss 0.7026 (0.9367) grad_norm 7.1060 (8.9987/2.5860) mem 68106MB [2022-12-19 14:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1070/1519] eta 0:07:31 lr 0.000030 time 0.9276 (1.0064) model_time 0.9275 (1.0057) loss 0.8384 (0.9366) grad_norm 7.1581 (8.9994/2.6005) mem 68106MB [2022-12-19 14:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1080/1519] eta 0:07:21 lr 0.000030 time 0.9420 (1.0064) model_time 0.9417 (1.0056) loss 0.9788 (0.9366) grad_norm 9.6671 (8.9790/2.5882) mem 68106MB [2022-12-19 14:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1090/1519] eta 0:07:11 lr 0.000030 time 0.9254 (1.0064) model_time 0.9252 (1.0056) loss 1.1847 (0.9372) grad_norm 7.5616 (8.9486/2.5612) mem 68106MB [2022-12-19 14:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1100/1519] eta 0:07:01 lr 0.000030 time 0.9296 (1.0063) model_time 0.9295 (1.0056) loss 0.7972 (0.9381) grad_norm 12.9925 (8.9620/2.5719) mem 68106MB [2022-12-19 14:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1110/1519] eta 0:06:51 lr 0.000030 time 0.9227 (1.0063) model_time 0.9226 (1.0055) loss 0.8492 (0.9386) grad_norm 6.9577 (8.9564/2.5716) mem 68106MB [2022-12-19 14:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1120/1519] eta 0:06:41 lr 0.000030 time 0.9309 (1.0063) model_time 0.9308 (1.0055) loss 0.7302 (0.9373) grad_norm 6.7739 (8.9696/2.5719) mem 68106MB [2022-12-19 14:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1130/1519] eta 0:06:31 lr 0.000030 time 0.9221 (1.0063) model_time 0.9219 (1.0056) loss 0.8747 (0.9373) grad_norm 7.5594 (8.9706/2.5746) mem 68106MB [2022-12-19 14:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1140/1519] eta 0:06:21 lr 0.000030 time 0.9430 (1.0063) model_time 0.9428 (1.0055) loss 0.9157 (0.9372) grad_norm 7.0073 (8.9573/2.5764) mem 68106MB [2022-12-19 14:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1150/1519] eta 0:06:11 lr 0.000030 time 0.9336 (1.0062) model_time 0.9335 (1.0055) loss 0.7291 (0.9367) grad_norm 7.5698 (8.9322/2.5733) mem 68106MB [2022-12-19 14:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1160/1519] eta 0:06:01 lr 0.000030 time 0.8866 (1.0066) model_time 0.8865 (1.0058) loss 0.9516 (0.9369) grad_norm 9.3546 (8.9916/2.6190) mem 68106MB [2022-12-19 14:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1170/1519] eta 0:05:51 lr 0.000030 time 0.9644 (1.0065) model_time 0.9643 (1.0058) loss 1.0124 (0.9375) grad_norm 7.5908 (9.0187/2.6178) mem 68106MB [2022-12-19 14:12:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1180/1519] eta 0:05:41 lr 0.000030 time 0.9311 (1.0066) model_time 0.9310 (1.0059) loss 0.9243 (0.9372) grad_norm 7.6655 (9.0027/2.6159) mem 68106MB [2022-12-19 14:12:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1190/1519] eta 0:05:31 lr 0.000030 time 0.9852 (1.0066) model_time 0.9850 (1.0059) loss 0.8314 (0.9367) grad_norm 10.3921 (8.9921/2.5932) mem 68106MB [2022-12-19 14:13:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1200/1519] eta 0:05:21 lr 0.000030 time 0.9301 (1.0066) model_time 0.9299 (1.0059) loss 0.9221 (0.9384) grad_norm 7.6569 (9.0034/2.5957) mem 68106MB [2022-12-19 14:13:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1210/1519] eta 0:05:11 lr 0.000030 time 0.9318 (1.0066) model_time 0.9316 (1.0058) loss 1.0462 (0.9392) grad_norm 9.1499 (9.0338/2.5957) mem 68106MB [2022-12-19 14:13:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1220/1519] eta 0:05:00 lr 0.000030 time 0.9798 (1.0065) model_time 0.9797 (1.0058) loss 0.9313 (0.9388) grad_norm 12.3468 (9.0236/2.5843) mem 68106MB [2022-12-19 14:13:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1230/1519] eta 0:04:50 lr 0.000030 time 0.9275 (1.0065) model_time 0.9273 (1.0058) loss 0.7359 (0.9381) grad_norm 8.4539 (9.0177/2.5489) mem 68106MB [2022-12-19 14:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1240/1519] eta 0:04:40 lr 0.000030 time 0.9374 (1.0065) model_time 0.9372 (1.0058) loss 1.0487 (0.9392) grad_norm 9.3506 (9.0171/2.5487) mem 68106MB [2022-12-19 14:13:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1250/1519] eta 0:04:30 lr 0.000030 time 0.9300 (1.0066) model_time 0.9298 (1.0058) loss 0.6962 (0.9399) grad_norm 7.3365 (9.0374/2.5433) mem 68106MB [2022-12-19 14:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1260/1519] eta 0:04:20 lr 0.000030 time 0.9270 (1.0065) model_time 0.9269 (1.0058) loss 0.7168 (0.9395) grad_norm 8.5661 (9.0687/2.5806) mem 68106MB [2022-12-19 14:14:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1270/1519] eta 0:04:10 lr 0.000030 time 0.9337 (1.0065) model_time 0.9335 (1.0058) loss 0.9182 (0.9393) grad_norm 14.6375 (9.0975/2.5961) mem 68106MB [2022-12-19 14:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1280/1519] eta 0:04:00 lr 0.000030 time 0.9247 (1.0065) model_time 0.9245 (1.0058) loss 1.2009 (0.9393) grad_norm 8.2640 (9.0822/2.5801) mem 68106MB [2022-12-19 14:14:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1290/1519] eta 0:03:50 lr 0.000030 time 0.9378 (1.0064) model_time 0.9377 (1.0057) loss 0.9680 (0.9392) grad_norm 6.3023 (9.0512/2.5438) mem 68106MB [2022-12-19 14:14:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1300/1519] eta 0:03:40 lr 0.000030 time 0.9411 (1.0064) model_time 0.9409 (1.0057) loss 1.0596 (0.9400) grad_norm 6.7678 (9.0370/2.5386) mem 68106MB [2022-12-19 14:14:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1310/1519] eta 0:03:30 lr 0.000030 time 0.9243 (1.0064) model_time 0.9242 (1.0057) loss 1.1300 (0.9414) grad_norm 6.4695 (9.0076/2.5246) mem 68106MB [2022-12-19 14:15:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1320/1519] eta 0:03:20 lr 0.000030 time 0.9229 (1.0063) model_time 0.9227 (1.0056) loss 0.7951 (0.9414) grad_norm 6.3062 (8.9706/2.5155) mem 68106MB [2022-12-19 14:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1330/1519] eta 0:03:10 lr 0.000030 time 0.9292 (1.0063) model_time 0.9290 (1.0056) loss 1.1399 (0.9416) grad_norm 11.9296 (8.9657/2.5426) mem 68106MB [2022-12-19 14:15:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1340/1519] eta 0:03:00 lr 0.000030 time 0.9274 (1.0064) model_time 0.9272 (1.0057) loss 0.7922 (0.9416) grad_norm 12.7526 (8.9495/2.5267) mem 68106MB [2022-12-19 14:15:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1350/1519] eta 0:02:50 lr 0.000030 time 0.9237 (1.0063) model_time 0.9236 (1.0056) loss 1.2618 (0.9421) grad_norm 7.5167 (8.9251/2.5165) mem 68106MB [2022-12-19 14:15:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1360/1519] eta 0:02:39 lr 0.000030 time 0.9224 (1.0063) model_time 0.9223 (1.0056) loss 0.9776 (0.9414) grad_norm 7.3888 (8.9237/2.5157) mem 68106MB [2022-12-19 14:15:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1370/1519] eta 0:02:29 lr 0.000030 time 0.9499 (1.0062) model_time 0.9498 (1.0055) loss 1.2483 (0.9419) grad_norm 6.4996 (8.9283/2.5352) mem 68106MB [2022-12-19 14:16:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1380/1519] eta 0:02:19 lr 0.000030 time 0.9342 (1.0062) model_time 0.9340 (1.0055) loss 0.6921 (0.9417) grad_norm 8.5471 (8.8962/2.5263) mem 68106MB [2022-12-19 14:16:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1390/1519] eta 0:02:09 lr 0.000030 time 0.9431 (1.0062) model_time 0.9429 (1.0055) loss 0.9494 (0.9415) grad_norm 9.4661 (8.8731/2.5110) mem 68106MB [2022-12-19 14:16:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1400/1519] eta 0:01:59 lr 0.000030 time 0.9324 (1.0062) model_time 0.9323 (1.0055) loss 0.7326 (0.9420) grad_norm 14.7899 (8.8612/2.4329) mem 68106MB [2022-12-19 14:16:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1410/1519] eta 0:01:49 lr 0.000030 time 0.9296 (1.0062) model_time 0.9294 (1.0055) loss 0.9315 (0.9419) grad_norm 5.9966 (8.8558/2.4369) mem 68106MB [2022-12-19 14:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1420/1519] eta 0:01:39 lr 0.000030 time 0.9416 (1.0062) model_time 0.9415 (1.0055) loss 0.8634 (0.9419) grad_norm 11.2946 (8.8500/2.4225) mem 68106MB [2022-12-19 14:16:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1430/1519] eta 0:01:29 lr 0.000030 time 0.9285 (1.0061) model_time 0.9283 (1.0055) loss 0.7636 (0.9422) grad_norm 8.3973 (8.8249/2.4154) mem 68106MB [2022-12-19 14:17:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1440/1519] eta 0:01:19 lr 0.000030 time 0.9413 (1.0062) model_time 0.9412 (1.0055) loss 0.8428 (0.9426) grad_norm 6.5683 (8.8044/2.4241) mem 68106MB [2022-12-19 14:17:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1450/1519] eta 0:01:09 lr 0.000030 time 0.9299 (1.0062) model_time 0.9298 (1.0055) loss 1.0090 (0.9422) grad_norm 7.8194 (8.7801/2.4164) mem 68106MB [2022-12-19 14:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1460/1519] eta 0:00:59 lr 0.000030 time 0.9363 (1.0061) model_time 0.9361 (1.0055) loss 0.8924 (0.9421) grad_norm 8.1244 (8.8112/2.4218) mem 68106MB [2022-12-19 14:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1470/1519] eta 0:00:49 lr 0.000030 time 0.9300 (1.0061) model_time 0.9298 (1.0055) loss 0.9037 (0.9422) grad_norm 8.1580 (8.7296/2.1928) mem 68106MB [2022-12-19 14:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1480/1519] eta 0:00:39 lr 0.000030 time 0.9352 (1.0063) model_time 0.9351 (1.0056) loss 1.0432 (0.9421) grad_norm 10.0808 (8.7360/2.1675) mem 68106MB [2022-12-19 14:17:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1490/1519] eta 0:00:29 lr 0.000030 time 1.0129 (1.0064) model_time 1.0127 (1.0057) loss 1.1996 (0.9422) grad_norm 10.0596 (8.7069/2.1294) mem 68106MB [2022-12-19 14:18:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1500/1519] eta 0:00:19 lr 0.000030 time 0.9230 (1.0063) model_time 0.9229 (1.0057) loss 0.7267 (0.9422) grad_norm 6.9237 (8.7043/2.1242) mem 68106MB [2022-12-19 14:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [22/100][1510/1519] eta 0:00:09 lr 0.000030 time 0.9165 (1.0064) model_time 0.9165 (1.0057) loss 0.8741 (0.9422) grad_norm 7.0654 (8.7165/2.1323) mem 68106MB [2022-12-19 14:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 22 training takes 0:25:28 [2022-12-19 14:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_22.pth saving...... [2022-12-19 14:18:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_22.pth saved !!! [2022-12-19 14:18:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.705 (0.705) Loss 0.5053 (0.5053) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 14:18:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.335) Loss 0.5429 (0.5203) Acc@1 90.625 (91.004) Acc@5 97.917 (98.169) Mem 68106MB [2022-12-19 14:18:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.317) Loss 0.4799 (0.5215) Acc@1 89.931 (90.443) Acc@5 98.264 (98.065) Mem 68106MB [2022-12-19 14:18:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.311) Loss 0.6388 (0.5261) Acc@1 89.236 (90.345) Acc@5 96.528 (97.961) Mem 68106MB [2022-12-19 14:18:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.309) Loss 0.4985 (0.5182) Acc@1 89.931 (90.498) Acc@5 98.611 (98.128) Mem 68106MB [2022-12-19 14:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.307) Loss 0.5104 (0.5143) Acc@1 87.500 (90.502) Acc@5 99.653 (98.230) Mem 68106MB [2022-12-19 14:19:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.306) Loss 0.5312 (0.5145) Acc@1 87.500 (90.431) Acc@5 98.264 (98.218) Mem 68106MB [2022-12-19 14:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.301 (0.305) Loss 0.5797 (0.5169) Acc@1 89.583 (90.415) Acc@5 98.264 (98.200) Mem 68106MB [2022-12-19 14:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.304) Loss 0.4691 (0.5157) Acc@1 90.625 (90.394) Acc@5 98.264 (98.225) Mem 68106MB [2022-12-19 14:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:22] * Acc@1 90.377 Acc@5 98.220 [2022-12-19 14:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 90.4% [2022-12-19 14:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 14:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 14:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.38% [2022-12-19 14:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][0/1519] eta 0:34:28 lr 0.000030 time 1.3620 (1.3620) model_time 0.9510 (0.9510) loss 1.1265 (1.1265) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 14:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][10/1519] eta 0:25:58 lr 0.000030 time 0.9280 (1.0329) model_time 0.9278 (0.9951) loss 0.7957 (0.8911) grad_norm 6.9293 (8.5733/1.8322) mem 68106MB [2022-12-19 14:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][20/1519] eta 0:25:26 lr 0.000030 time 0.9313 (1.0182) model_time 0.9312 (0.9983) loss 0.8978 (0.9070) grad_norm 7.5343 (7.9463/1.6277) mem 68106MB [2022-12-19 14:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][30/1519] eta 0:25:09 lr 0.000030 time 0.9299 (1.0137) model_time 0.9297 (1.0001) loss 0.8314 (0.9059) grad_norm 7.7535 (8.1615/1.5430) mem 68106MB [2022-12-19 14:20:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][40/1519] eta 0:25:05 lr 0.000030 time 0.9234 (1.0179) model_time 0.9232 (1.0075) loss 0.8490 (0.8961) grad_norm 7.1422 (8.0835/1.4279) mem 68106MB [2022-12-19 14:20:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][50/1519] eta 0:24:52 lr 0.000030 time 0.9368 (1.0163) model_time 0.9366 (1.0078) loss 1.3699 (0.9114) grad_norm 9.1080 (8.4410/2.1332) mem 68106MB [2022-12-19 14:20:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][60/1519] eta 0:24:40 lr 0.000030 time 0.9402 (1.0144) model_time 0.9400 (1.0073) loss 0.8812 (0.8938) grad_norm 10.4069 (8.3647/2.0481) mem 68106MB [2022-12-19 14:20:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][70/1519] eta 0:24:26 lr 0.000030 time 0.9241 (1.0120) model_time 0.9240 (1.0058) loss 0.7856 (0.9071) grad_norm 9.7602 (8.3560/1.9857) mem 68106MB [2022-12-19 14:20:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][80/1519] eta 0:24:15 lr 0.000030 time 0.9189 (1.0118) model_time 0.9188 (1.0063) loss 1.1075 (0.9122) grad_norm 7.9976 (8.4286/1.9686) mem 68106MB [2022-12-19 14:21:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][90/1519] eta 0:24:03 lr 0.000030 time 0.9216 (1.0104) model_time 0.9214 (1.0055) loss 0.9250 (0.9054) grad_norm 9.4056 (8.4256/1.9460) mem 68106MB [2022-12-19 14:21:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][100/1519] eta 0:23:52 lr 0.000030 time 0.9298 (1.0095) model_time 0.9297 (1.0051) loss 1.0163 (0.9032) grad_norm 10.2856 (8.3067/1.9537) mem 68106MB [2022-12-19 14:21:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][110/1519] eta 0:23:41 lr 0.000030 time 0.9243 (1.0089) model_time 0.9242 (1.0048) loss 0.9338 (0.8973) grad_norm 10.5970 (8.3536/1.9199) mem 68106MB [2022-12-19 14:21:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][120/1519] eta 0:23:30 lr 0.000030 time 0.9219 (1.0082) model_time 0.9218 (1.0044) loss 0.7708 (0.8907) grad_norm 10.5656 (8.4228/1.9125) mem 68106MB [2022-12-19 14:21:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][130/1519] eta 0:23:20 lr 0.000030 time 0.9287 (1.0080) model_time 0.9286 (1.0045) loss 0.9024 (0.8952) grad_norm 14.1718 (8.5963/2.0191) mem 68106MB [2022-12-19 14:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][140/1519] eta 0:23:09 lr 0.000030 time 0.9254 (1.0079) model_time 0.9252 (1.0046) loss 0.8411 (0.8988) grad_norm 6.5502 (8.5664/1.9850) mem 68106MB [2022-12-19 14:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][150/1519] eta 0:22:59 lr 0.000030 time 0.9984 (1.0079) model_time 0.9983 (1.0048) loss 0.7637 (0.8956) grad_norm 7.1357 (8.5546/1.9453) mem 68106MB [2022-12-19 14:22:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][160/1519] eta 0:22:49 lr 0.000030 time 0.9206 (1.0077) model_time 0.9205 (1.0048) loss 0.8027 (0.8985) grad_norm 9.8613 (8.7264/2.0672) mem 68106MB [2022-12-19 14:22:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][170/1519] eta 0:22:39 lr 0.000030 time 0.9338 (1.0075) model_time 0.9337 (1.0047) loss 0.7826 (0.8967) grad_norm 6.6246 (8.6957/2.0711) mem 68106MB [2022-12-19 14:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][180/1519] eta 0:22:28 lr 0.000030 time 0.9304 (1.0069) model_time 0.9301 (1.0042) loss 1.0916 (0.8980) grad_norm 7.3242 (8.6915/2.0291) mem 68106MB [2022-12-19 14:22:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][190/1519] eta 0:22:17 lr 0.000030 time 0.9224 (1.0068) model_time 0.9222 (1.0042) loss 0.9540 (0.8971) grad_norm 7.7307 (8.6732/2.0258) mem 68106MB [2022-12-19 14:22:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][200/1519] eta 0:22:07 lr 0.000030 time 0.9243 (1.0064) model_time 0.9241 (1.0040) loss 1.0463 (0.9060) grad_norm 8.2894 (8.6590/1.9938) mem 68106MB [2022-12-19 14:23:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][210/1519] eta 0:21:57 lr 0.000030 time 0.9299 (1.0063) model_time 0.9296 (1.0040) loss 0.6955 (0.9036) grad_norm 8.6349 (8.6702/1.9569) mem 68106MB [2022-12-19 14:23:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][220/1519] eta 0:21:47 lr 0.000030 time 0.9219 (1.0067) model_time 0.9218 (1.0044) loss 1.0636 (0.9057) grad_norm 7.4573 (8.7097/1.9877) mem 68106MB [2022-12-19 14:23:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][230/1519] eta 0:21:37 lr 0.000030 time 0.9237 (1.0064) model_time 0.9236 (1.0043) loss 0.9272 (0.9064) grad_norm 5.6581 (8.6655/1.9743) mem 68106MB [2022-12-19 14:23:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][240/1519] eta 0:21:27 lr 0.000030 time 0.9268 (1.0063) model_time 0.9266 (1.0042) loss 0.8966 (0.9045) grad_norm 9.3827 (8.7094/2.0001) mem 68106MB [2022-12-19 14:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][250/1519] eta 0:21:17 lr 0.000030 time 0.9356 (1.0065) model_time 0.9354 (1.0045) loss 1.5024 (0.9061) grad_norm 13.7044 (8.7841/2.0431) mem 68106MB [2022-12-19 14:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][260/1519] eta 0:21:06 lr 0.000030 time 0.9186 (1.0062) model_time 0.9185 (1.0043) loss 0.8502 (0.9084) grad_norm 7.2139 (8.7607/2.0248) mem 68106MB [2022-12-19 14:24:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][270/1519] eta 0:20:57 lr 0.000030 time 1.0291 (1.0068) model_time 1.0290 (1.0050) loss 0.8118 (0.9084) grad_norm 7.3549 (8.7298/1.9998) mem 68106MB [2022-12-19 14:24:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][280/1519] eta 0:20:46 lr 0.000030 time 0.9240 (1.0064) model_time 0.9238 (1.0046) loss 0.8028 (0.9084) grad_norm 15.8545 (8.8420/2.0917) mem 68106MB [2022-12-19 14:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][290/1519] eta 0:20:37 lr 0.000030 time 0.9202 (1.0067) model_time 0.9200 (1.0050) loss 0.9457 (0.9104) grad_norm 5.6222 (8.7818/2.0911) mem 68106MB [2022-12-19 14:24:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][300/1519] eta 0:20:27 lr 0.000030 time 0.9221 (1.0069) model_time 0.9220 (1.0052) loss 1.0902 (0.9089) grad_norm 10.3269 (8.8197/2.0815) mem 68106MB [2022-12-19 14:24:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][310/1519] eta 0:20:16 lr 0.000030 time 0.9290 (1.0066) model_time 0.9288 (1.0049) loss 0.8058 (0.9117) grad_norm 8.2032 (8.8195/2.0815) mem 68106MB [2022-12-19 14:24:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][320/1519] eta 0:20:06 lr 0.000030 time 0.9247 (1.0063) model_time 0.9246 (1.0047) loss 1.3122 (0.9114) grad_norm 19.9061 (8.9190/2.2405) mem 68106MB [2022-12-19 14:25:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][330/1519] eta 0:19:56 lr 0.000029 time 0.9217 (1.0062) model_time 0.9216 (1.0046) loss 0.8021 (0.9101) grad_norm 9.8575 (8.9430/2.2223) mem 68106MB [2022-12-19 14:25:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][340/1519] eta 0:19:46 lr 0.000029 time 0.9282 (1.0060) model_time 0.9281 (1.0045) loss 0.8343 (0.9095) grad_norm 10.4010 (8.9397/2.1945) mem 68106MB [2022-12-19 14:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][350/1519] eta 0:19:36 lr 0.000029 time 1.1914 (1.0067) model_time 1.1913 (1.0051) loss 0.8701 (0.9091) grad_norm 8.6174 (8.9267/2.1680) mem 68106MB [2022-12-19 14:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][360/1519] eta 0:19:26 lr 0.000029 time 0.9251 (1.0065) model_time 0.9249 (1.0050) loss 1.4758 (0.9094) grad_norm 7.8314 (8.9140/2.1895) mem 68106MB [2022-12-19 14:25:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][370/1519] eta 0:19:16 lr 0.000029 time 0.9204 (1.0065) model_time 0.9202 (1.0051) loss 0.8522 (0.9098) grad_norm 6.6197 (8.8958/2.1781) mem 68106MB [2022-12-19 14:25:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][380/1519] eta 0:19:06 lr 0.000029 time 0.9170 (1.0064) model_time 0.9169 (1.0050) loss 1.0477 (0.9089) grad_norm 7.0691 (8.8809/2.1713) mem 68106MB [2022-12-19 14:26:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][390/1519] eta 0:18:56 lr 0.000029 time 0.9203 (1.0063) model_time 0.9202 (1.0049) loss 1.0286 (0.9070) grad_norm 9.8803 (8.9422/2.3733) mem 68106MB [2022-12-19 14:26:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][400/1519] eta 0:18:46 lr 0.000029 time 0.9196 (1.0064) model_time 0.9195 (1.0050) loss 0.7434 (0.9070) grad_norm 8.2495 (8.9351/2.3518) mem 68106MB [2022-12-19 14:26:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][410/1519] eta 0:18:35 lr 0.000029 time 0.9220 (1.0062) model_time 0.9217 (1.0048) loss 0.8357 (0.9065) grad_norm 12.5199 (8.9432/2.3457) mem 68106MB [2022-12-19 14:26:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][420/1519] eta 0:18:26 lr 0.000029 time 0.9660 (1.0064) model_time 0.9658 (1.0051) loss 0.7093 (0.9078) grad_norm 10.4308 (8.9464/2.3249) mem 68106MB [2022-12-19 14:26:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][430/1519] eta 0:18:15 lr 0.000029 time 0.9237 (1.0063) model_time 0.9236 (1.0050) loss 0.8617 (0.9084) grad_norm 12.3103 (8.9938/2.3370) mem 68106MB [2022-12-19 14:26:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][440/1519] eta 0:18:05 lr 0.000029 time 0.9246 (1.0061) model_time 0.9244 (1.0048) loss 0.8293 (0.9097) grad_norm 8.7666 (8.9706/2.3264) mem 68106MB [2022-12-19 14:27:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][450/1519] eta 0:17:56 lr 0.000029 time 1.0355 (1.0066) model_time 1.0354 (1.0053) loss 1.2361 (0.9110) grad_norm 6.0635 (8.9289/2.3206) mem 68106MB [2022-12-19 14:27:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][460/1519] eta 0:17:45 lr 0.000029 time 0.9195 (1.0064) model_time 0.9193 (1.0052) loss 0.8155 (0.9131) grad_norm 9.7171 (8.9614/2.3459) mem 68106MB [2022-12-19 14:27:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][470/1519] eta 0:17:35 lr 0.000029 time 0.9117 (1.0066) model_time 0.9114 (1.0054) loss 0.9473 (0.9142) grad_norm 7.8419 (8.9903/2.3886) mem 68106MB [2022-12-19 14:27:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][480/1519] eta 0:17:26 lr 0.000029 time 0.9398 (1.0068) model_time 0.9397 (1.0056) loss 1.0120 (0.9146) grad_norm 7.3244 (8.9790/2.3785) mem 68106MB [2022-12-19 14:27:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][490/1519] eta 0:17:15 lr 0.000029 time 0.9221 (1.0067) model_time 0.9220 (1.0055) loss 0.7247 (0.9155) grad_norm 5.5477 (8.9648/2.3675) mem 68106MB [2022-12-19 14:27:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][500/1519] eta 0:17:05 lr 0.000029 time 0.9202 (1.0065) model_time 0.9201 (1.0053) loss 0.8280 (0.9159) grad_norm 6.2066 (8.9320/2.3620) mem 68106MB [2022-12-19 14:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][510/1519] eta 0:16:55 lr 0.000029 time 0.9224 (1.0064) model_time 0.9222 (1.0053) loss 0.9631 (0.9159) grad_norm 13.0059 (8.9408/2.3574) mem 68106MB [2022-12-19 14:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][520/1519] eta 0:16:45 lr 0.000029 time 0.9301 (1.0063) model_time 0.9300 (1.0051) loss 0.9622 (0.9151) grad_norm 7.2262 (8.9096/2.3520) mem 68106MB [2022-12-19 14:28:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][530/1519] eta 0:16:35 lr 0.000029 time 1.0155 (1.0069) model_time 1.0154 (1.0058) loss 0.7569 (0.9160) grad_norm 6.6037 (8.9089/2.3377) mem 68106MB [2022-12-19 14:28:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][540/1519] eta 0:16:25 lr 0.000029 time 0.9182 (1.0067) model_time 0.9180 (1.0056) loss 1.1850 (0.9160) grad_norm 7.6421 (8.9079/2.3207) mem 68106MB [2022-12-19 14:28:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][550/1519] eta 0:16:15 lr 0.000029 time 0.9261 (1.0065) model_time 0.9260 (1.0054) loss 0.7881 (0.9169) grad_norm 9.0554 (8.9027/2.3030) mem 68106MB [2022-12-19 14:28:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][560/1519] eta 0:16:05 lr 0.000029 time 0.9170 (1.0065) model_time 0.9169 (1.0054) loss 1.0176 (0.9170) grad_norm 15.0311 (8.9180/2.3167) mem 68106MB [2022-12-19 14:29:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][570/1519] eta 0:15:55 lr 0.000029 time 0.9244 (1.0064) model_time 0.9242 (1.0053) loss 0.8209 (0.9183) grad_norm 8.3496 (8.9204/2.3000) mem 68106MB [2022-12-19 14:29:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][580/1519] eta 0:15:44 lr 0.000029 time 0.9247 (1.0064) model_time 0.9245 (1.0053) loss 1.1554 (0.9187) grad_norm 9.8517 (8.9108/2.2841) mem 68106MB [2022-12-19 14:29:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][590/1519] eta 0:15:34 lr 0.000029 time 0.9261 (1.0062) model_time 0.9260 (1.0052) loss 0.8088 (0.9204) grad_norm 7.3288 (8.9127/2.2917) mem 68106MB [2022-12-19 14:29:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][600/1519] eta 0:15:24 lr 0.000029 time 0.9871 (1.0065) model_time 0.9870 (1.0054) loss 1.0877 (0.9203) grad_norm 7.0478 (8.8927/2.2785) mem 68106MB [2022-12-19 14:29:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][610/1519] eta 0:15:14 lr 0.000029 time 0.8895 (1.0064) model_time 0.8894 (1.0054) loss 0.7157 (0.9203) grad_norm 7.1586 (8.8737/2.2780) mem 68106MB [2022-12-19 14:29:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][620/1519] eta 0:15:04 lr 0.000029 time 0.9277 (1.0063) model_time 0.9276 (1.0053) loss 1.2805 (0.9204) grad_norm 10.1668 (8.9210/2.2813) mem 68106MB [2022-12-19 14:30:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][630/1519] eta 0:14:54 lr 0.000029 time 0.9235 (1.0062) model_time 0.9233 (1.0052) loss 0.8668 (0.9189) grad_norm 7.1479 (8.9068/2.2874) mem 68106MB [2022-12-19 14:30:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][640/1519] eta 0:14:44 lr 0.000029 time 0.9280 (1.0061) model_time 0.9279 (1.0051) loss 0.7045 (0.9200) grad_norm 6.8159 (8.8937/2.2952) mem 68106MB [2022-12-19 14:30:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][650/1519] eta 0:14:34 lr 0.000029 time 0.9364 (1.0061) model_time 0.9363 (1.0051) loss 0.9012 (0.9230) grad_norm 8.0513 (8.8654/2.2518) mem 68106MB [2022-12-19 14:30:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][660/1519] eta 0:14:24 lr 0.000029 time 0.9245 (1.0060) model_time 0.9243 (1.0050) loss 1.0554 (0.9241) grad_norm 11.9867 (8.8770/2.2599) mem 68106MB [2022-12-19 14:30:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][670/1519] eta 0:14:14 lr 0.000029 time 0.9300 (1.0061) model_time 0.9298 (1.0051) loss 0.9768 (0.9245) grad_norm 8.1018 (8.8902/2.2546) mem 68106MB [2022-12-19 14:30:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][680/1519] eta 0:14:04 lr 0.000029 time 0.9326 (1.0060) model_time 0.9325 (1.0051) loss 0.8795 (0.9256) grad_norm 8.3914 (8.8820/2.2485) mem 68106MB [2022-12-19 14:31:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][690/1519] eta 0:13:53 lr 0.000029 time 0.9160 (1.0059) model_time 0.9158 (1.0050) loss 1.0461 (0.9261) grad_norm 5.7048 (8.8786/2.2604) mem 68106MB [2022-12-19 14:31:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][700/1519] eta 0:13:43 lr 0.000029 time 0.9293 (1.0059) model_time 0.9292 (1.0050) loss 1.1090 (0.9260) grad_norm 7.5601 (8.9011/2.2440) mem 68106MB [2022-12-19 14:31:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][710/1519] eta 0:13:33 lr 0.000029 time 0.9237 (1.0061) model_time 0.9236 (1.0052) loss 1.1610 (0.9253) grad_norm 6.6234 (8.8933/2.2448) mem 68106MB [2022-12-19 14:31:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][720/1519] eta 0:13:23 lr 0.000029 time 0.9225 (1.0060) model_time 0.9223 (1.0051) loss 1.0747 (0.9248) grad_norm 12.6518 (8.8882/2.2500) mem 68106MB [2022-12-19 14:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][730/1519] eta 0:13:13 lr 0.000029 time 0.9255 (1.0060) model_time 0.9254 (1.0051) loss 0.9379 (0.9258) grad_norm 8.0121 (8.8553/2.2297) mem 68106MB [2022-12-19 14:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][740/1519] eta 0:13:03 lr 0.000029 time 0.9292 (1.0059) model_time 0.9290 (1.0050) loss 1.0617 (0.9265) grad_norm 8.5566 (8.8557/2.2361) mem 68106MB [2022-12-19 14:32:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][750/1519] eta 0:12:53 lr 0.000029 time 0.9383 (1.0058) model_time 0.9381 (1.0049) loss 0.7069 (0.9269) grad_norm 11.2282 (8.8780/2.2345) mem 68106MB [2022-12-19 14:32:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][760/1519] eta 0:12:43 lr 0.000029 time 1.0091 (1.0059) model_time 1.0090 (1.0050) loss 1.0916 (0.9266) grad_norm 6.6634 (8.8237/2.2021) mem 68106MB [2022-12-19 14:32:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][770/1519] eta 0:12:33 lr 0.000029 time 0.9240 (1.0060) model_time 0.9239 (1.0051) loss 1.0483 (0.9263) grad_norm 5.4110 (8.8143/2.2023) mem 68106MB [2022-12-19 14:32:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][780/1519] eta 0:12:23 lr 0.000029 time 0.9270 (1.0060) model_time 0.9268 (1.0051) loss 0.7211 (0.9261) grad_norm 9.0931 (8.8308/2.2243) mem 68106MB [2022-12-19 14:32:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][790/1519] eta 0:12:13 lr 0.000029 time 0.9062 (1.0060) model_time 0.9061 (1.0052) loss 1.2665 (0.9260) grad_norm 7.9675 (8.8169/2.2191) mem 68106MB [2022-12-19 14:33:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][800/1519] eta 0:12:03 lr 0.000029 time 0.9192 (1.0060) model_time 0.9191 (1.0051) loss 0.8125 (0.9260) grad_norm 12.4347 (8.8331/2.2285) mem 68106MB [2022-12-19 14:33:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][810/1519] eta 0:11:53 lr 0.000029 time 0.9246 (1.0059) model_time 0.9245 (1.0051) loss 0.9099 (0.9248) grad_norm 9.4623 (8.8549/2.2396) mem 68106MB [2022-12-19 14:33:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][820/1519] eta 0:11:43 lr 0.000029 time 0.9086 (1.0060) model_time 0.9084 (1.0051) loss 0.8899 (0.9240) grad_norm 7.9102 (8.8623/2.2402) mem 68106MB [2022-12-19 14:33:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][830/1519] eta 0:11:33 lr 0.000029 time 0.9366 (1.0059) model_time 0.9364 (1.0051) loss 0.8832 (0.9238) grad_norm 6.7926 (8.8587/2.2389) mem 68106MB [2022-12-19 14:33:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][840/1519] eta 0:11:23 lr 0.000029 time 0.9298 (1.0059) model_time 0.9297 (1.0051) loss 1.1541 (0.9248) grad_norm 10.9631 (8.8522/2.2356) mem 68106MB [2022-12-19 14:33:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][850/1519] eta 0:11:13 lr 0.000029 time 0.9373 (1.0061) model_time 0.9372 (1.0052) loss 1.1902 (0.9254) grad_norm 6.1598 (8.8181/2.2132) mem 68106MB [2022-12-19 14:34:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][860/1519] eta 0:11:02 lr 0.000029 time 0.9351 (1.0060) model_time 0.9350 (1.0052) loss 0.9846 (0.9253) grad_norm 7.4749 (8.8105/2.2155) mem 68106MB [2022-12-19 14:34:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][870/1519] eta 0:10:52 lr 0.000029 time 0.9595 (1.0061) model_time 0.9594 (1.0053) loss 1.2015 (0.9260) grad_norm 7.6605 (8.8384/2.2203) mem 68106MB [2022-12-19 14:34:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][880/1519] eta 0:10:42 lr 0.000029 time 0.9374 (1.0060) model_time 0.9373 (1.0052) loss 0.7640 (0.9265) grad_norm 8.5911 (8.7810/2.1703) mem 68106MB [2022-12-19 14:34:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][890/1519] eta 0:10:32 lr 0.000029 time 1.0073 (1.0060) model_time 1.0072 (1.0052) loss 0.7156 (0.9263) grad_norm 11.9707 (8.8145/2.1696) mem 68106MB [2022-12-19 14:34:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][900/1519] eta 0:10:22 lr 0.000029 time 0.9227 (1.0060) model_time 0.9225 (1.0052) loss 0.7266 (0.9254) grad_norm 6.3342 (8.7987/2.1821) mem 68106MB [2022-12-19 14:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][910/1519] eta 0:10:12 lr 0.000029 time 1.0087 (1.0062) model_time 1.0086 (1.0054) loss 0.8679 (0.9250) grad_norm 8.9601 (8.8042/2.1763) mem 68106MB [2022-12-19 14:35:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][920/1519] eta 0:10:02 lr 0.000029 time 0.9266 (1.0061) model_time 0.9265 (1.0054) loss 1.1187 (0.9256) grad_norm 6.4483 (8.7466/2.0773) mem 68106MB [2022-12-19 14:35:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][930/1519] eta 0:09:52 lr 0.000029 time 0.9240 (1.0061) model_time 0.9239 (1.0054) loss 1.0153 (0.9268) grad_norm 8.7958 (8.7556/2.1789) mem 68106MB [2022-12-19 14:35:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][940/1519] eta 0:09:42 lr 0.000029 time 0.9245 (1.0061) model_time 0.9243 (1.0053) loss 0.9636 (0.9260) grad_norm 12.4538 (8.7431/2.1985) mem 68106MB [2022-12-19 14:35:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][950/1519] eta 0:09:32 lr 0.000029 time 0.9246 (1.0060) model_time 0.9245 (1.0052) loss 0.7780 (0.9250) grad_norm 8.8478 (8.7316/2.2056) mem 68106MB [2022-12-19 14:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][960/1519] eta 0:09:22 lr 0.000029 time 0.9290 (1.0060) model_time 0.9289 (1.0052) loss 0.9258 (0.9260) grad_norm 6.9479 (8.7359/2.1865) mem 68106MB [2022-12-19 14:35:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][970/1519] eta 0:09:12 lr 0.000029 time 0.9255 (1.0059) model_time 0.9254 (1.0051) loss 1.1158 (0.9261) grad_norm 7.2825 (8.7382/2.1856) mem 68106MB [2022-12-19 14:36:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][980/1519] eta 0:09:02 lr 0.000029 time 0.9282 (1.0058) model_time 0.9281 (1.0051) loss 1.3021 (0.9263) grad_norm 12.0015 (8.7438/2.1848) mem 68106MB [2022-12-19 14:36:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][990/1519] eta 0:08:52 lr 0.000029 time 0.9215 (1.0058) model_time 0.9214 (1.0050) loss 1.2867 (0.9266) grad_norm 9.2641 (8.7094/2.0374) mem 68106MB [2022-12-19 14:36:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1000/1519] eta 0:08:41 lr 0.000029 time 0.9185 (1.0057) model_time 0.9183 (1.0049) loss 0.8366 (0.9265) grad_norm 8.5022 (8.7098/2.0534) mem 68106MB [2022-12-19 14:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1010/1519] eta 0:08:31 lr 0.000029 time 0.9341 (1.0056) model_time 0.9340 (1.0049) loss 0.7357 (0.9262) grad_norm 6.8203 (8.6991/2.0427) mem 68106MB [2022-12-19 14:36:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1020/1519] eta 0:08:21 lr 0.000029 time 0.9352 (1.0056) model_time 0.9350 (1.0049) loss 1.0151 (0.9255) grad_norm 9.8485 (8.7135/2.0476) mem 68106MB [2022-12-19 14:36:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1030/1519] eta 0:08:11 lr 0.000029 time 0.9300 (1.0055) model_time 0.9298 (1.0048) loss 0.6969 (0.9255) grad_norm 6.8807 (8.6882/2.0358) mem 68106MB [2022-12-19 14:37:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1040/1519] eta 0:08:01 lr 0.000029 time 0.9202 (1.0055) model_time 0.9201 (1.0047) loss 0.7713 (0.9258) grad_norm 8.4035 (8.6943/2.0303) mem 68106MB [2022-12-19 14:37:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1050/1519] eta 0:07:51 lr 0.000029 time 0.9767 (1.0054) model_time 0.9766 (1.0047) loss 0.9038 (0.9257) grad_norm 6.2482 (8.6960/2.0302) mem 68106MB [2022-12-19 14:37:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1060/1519] eta 0:07:41 lr 0.000029 time 0.9305 (1.0054) model_time 0.9304 (1.0047) loss 0.7971 (0.9264) grad_norm 6.3897 (8.6493/1.9936) mem 68106MB [2022-12-19 14:37:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1070/1519] eta 0:07:31 lr 0.000029 time 1.0252 (1.0054) model_time 1.0250 (1.0047) loss 0.8445 (0.9263) grad_norm 8.4867 (8.6352/1.9503) mem 68106MB [2022-12-19 14:37:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1080/1519] eta 0:07:21 lr 0.000029 time 0.9212 (1.0056) model_time 0.9211 (1.0048) loss 0.8008 (0.9256) grad_norm 8.7971 (8.6346/1.9374) mem 68106MB [2022-12-19 14:37:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1090/1519] eta 0:07:11 lr 0.000029 time 1.0007 (1.0056) model_time 1.0005 (1.0049) loss 0.6919 (0.9249) grad_norm 7.2649 (8.6574/1.9550) mem 68106MB [2022-12-19 14:38:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1100/1519] eta 0:07:01 lr 0.000029 time 1.0005 (1.0059) model_time 1.0003 (1.0052) loss 0.6989 (0.9248) grad_norm 9.0996 (8.7009/1.9522) mem 68106MB [2022-12-19 14:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1110/1519] eta 0:06:51 lr 0.000029 time 0.9386 (1.0059) model_time 0.9384 (1.0052) loss 0.8545 (0.9251) grad_norm 14.7480 (8.7276/2.0020) mem 68106MB [2022-12-19 14:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1120/1519] eta 0:06:41 lr 0.000029 time 0.9273 (1.0059) model_time 0.9272 (1.0052) loss 0.7505 (0.9242) grad_norm 7.0897 (8.7444/1.9946) mem 68106MB [2022-12-19 14:38:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1130/1519] eta 0:06:31 lr 0.000029 time 0.9288 (1.0058) model_time 0.9287 (1.0051) loss 1.3254 (0.9244) grad_norm 7.1717 (8.7447/2.0001) mem 68106MB [2022-12-19 14:38:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1140/1519] eta 0:06:21 lr 0.000029 time 0.9322 (1.0058) model_time 0.9320 (1.0051) loss 1.2048 (0.9244) grad_norm 7.9589 (8.7429/2.0045) mem 68106MB [2022-12-19 14:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1150/1519] eta 0:06:11 lr 0.000029 time 0.9487 (1.0058) model_time 0.9485 (1.0051) loss 0.9126 (0.9243) grad_norm 9.4354 (8.7444/2.0032) mem 68106MB [2022-12-19 14:39:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1160/1519] eta 0:06:01 lr 0.000029 time 0.9332 (1.0061) model_time 0.9330 (1.0054) loss 0.8306 (0.9247) grad_norm 6.4449 (8.7249/1.9709) mem 68106MB [2022-12-19 14:39:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1170/1519] eta 0:05:51 lr 0.000029 time 0.9255 (1.0061) model_time 0.9254 (1.0054) loss 0.8154 (0.9248) grad_norm 6.4575 (8.6913/1.9802) mem 68106MB [2022-12-19 14:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1180/1519] eta 0:05:41 lr 0.000029 time 0.9269 (1.0063) model_time 0.9268 (1.0056) loss 1.1161 (0.9246) grad_norm 8.8522 (8.6882/1.9834) mem 68106MB [2022-12-19 14:39:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1190/1519] eta 0:05:31 lr 0.000029 time 0.9241 (1.0062) model_time 0.9240 (1.0055) loss 0.7525 (0.9248) grad_norm 6.8622 (8.6836/1.9595) mem 68106MB [2022-12-19 14:39:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1200/1519] eta 0:05:20 lr 0.000029 time 0.9194 (1.0062) model_time 0.9193 (1.0055) loss 1.0438 (0.9253) grad_norm 6.7858 (8.6842/1.9601) mem 68106MB [2022-12-19 14:39:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1210/1519] eta 0:05:10 lr 0.000029 time 0.9317 (1.0062) model_time 0.9315 (1.0055) loss 1.1837 (0.9254) grad_norm 8.8322 (8.7242/1.9562) mem 68106MB [2022-12-19 14:40:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1220/1519] eta 0:05:00 lr 0.000029 time 0.9683 (1.0061) model_time 0.9681 (1.0055) loss 1.1300 (0.9257) grad_norm 9.8071 (8.7143/1.9487) mem 68106MB [2022-12-19 14:40:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1230/1519] eta 0:04:50 lr 0.000029 time 0.9249 (1.0061) model_time 0.9248 (1.0054) loss 0.9413 (0.9254) grad_norm 8.1435 (8.7218/1.9521) mem 68106MB [2022-12-19 14:40:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1240/1519] eta 0:04:40 lr 0.000029 time 0.9301 (1.0060) model_time 0.9300 (1.0054) loss 1.3434 (0.9256) grad_norm 12.5230 (8.7701/1.9587) mem 68106MB [2022-12-19 14:40:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1250/1519] eta 0:04:30 lr 0.000029 time 0.9326 (1.0060) model_time 0.9323 (1.0053) loss 0.8833 (0.9253) grad_norm 6.5832 (8.7746/1.9632) mem 68106MB [2022-12-19 14:40:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1260/1519] eta 0:04:20 lr 0.000029 time 0.9316 (1.0062) model_time 0.9314 (1.0055) loss 1.1127 (0.9258) grad_norm 8.3316 (8.7798/1.9445) mem 68106MB [2022-12-19 14:40:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1270/1519] eta 0:04:10 lr 0.000029 time 0.9302 (1.0061) model_time 0.9301 (1.0055) loss 0.7912 (0.9252) grad_norm 5.2655 (8.7624/1.9542) mem 68106MB [2022-12-19 14:41:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1280/1519] eta 0:04:00 lr 0.000029 time 0.9898 (1.0062) model_time 0.9896 (1.0055) loss 0.8627 (0.9260) grad_norm 7.9999 (8.7458/1.9691) mem 68106MB [2022-12-19 14:41:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1290/1519] eta 0:03:50 lr 0.000029 time 0.9287 (1.0061) model_time 0.9286 (1.0055) loss 0.8676 (0.9254) grad_norm 8.6220 (8.7595/1.9656) mem 68106MB [2022-12-19 14:41:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1300/1519] eta 0:03:40 lr 0.000029 time 0.9345 (1.0061) model_time 0.9343 (1.0054) loss 0.7547 (0.9255) grad_norm 13.7334 (8.8042/2.0095) mem 68106MB [2022-12-19 14:41:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1310/1519] eta 0:03:30 lr 0.000029 time 0.9312 (1.0060) model_time 0.9311 (1.0054) loss 0.7309 (0.9251) grad_norm 6.1457 (8.7890/2.0130) mem 68106MB [2022-12-19 14:41:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1320/1519] eta 0:03:20 lr 0.000029 time 0.9328 (1.0060) model_time 0.9327 (1.0053) loss 0.9335 (0.9253) grad_norm 7.6004 (8.7980/2.0227) mem 68106MB [2022-12-19 14:41:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1330/1519] eta 0:03:10 lr 0.000029 time 0.9275 (1.0059) model_time 0.9274 (1.0053) loss 1.1522 (0.9258) grad_norm 8.1788 (8.8005/2.0233) mem 68106MB [2022-12-19 14:42:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1340/1519] eta 0:03:00 lr 0.000029 time 0.9310 (1.0060) model_time 0.9308 (1.0053) loss 0.8816 (0.9254) grad_norm 7.6461 (8.8115/2.0080) mem 68106MB [2022-12-19 14:42:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1350/1519] eta 0:02:50 lr 0.000029 time 0.9338 (1.0059) model_time 0.9337 (1.0053) loss 0.7418 (0.9249) grad_norm 8.1222 (8.8045/2.0196) mem 68106MB [2022-12-19 14:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1360/1519] eta 0:02:39 lr 0.000029 time 0.9325 (1.0059) model_time 0.9324 (1.0052) loss 0.8574 (0.9250) grad_norm 9.0121 (8.8445/2.0314) mem 68106MB [2022-12-19 14:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1370/1519] eta 0:02:29 lr 0.000029 time 0.9250 (1.0059) model_time 0.9249 (1.0053) loss 0.7430 (0.9255) grad_norm 6.7049 (8.8948/2.1510) mem 68106MB [2022-12-19 14:42:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1380/1519] eta 0:02:19 lr 0.000029 time 0.9306 (1.0059) model_time 0.9304 (1.0052) loss 0.7469 (0.9253) grad_norm 8.2808 (8.9118/2.1828) mem 68106MB [2022-12-19 14:42:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1390/1519] eta 0:02:09 lr 0.000029 time 0.9451 (1.0060) model_time 0.9449 (1.0054) loss 0.7950 (0.9250) grad_norm 9.1027 (8.9193/2.1787) mem 68106MB [2022-12-19 14:43:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1400/1519] eta 0:01:59 lr 0.000029 time 0.9237 (1.0060) model_time 0.9235 (1.0054) loss 1.0778 (0.9255) grad_norm 6.8328 (8.9077/2.1764) mem 68106MB [2022-12-19 14:43:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1410/1519] eta 0:01:49 lr 0.000029 time 0.9339 (1.0061) model_time 0.9337 (1.0054) loss 0.7750 (0.9257) grad_norm 9.9204 (8.9169/2.1956) mem 68106MB [2022-12-19 14:43:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1420/1519] eta 0:01:39 lr 0.000029 time 0.9296 (1.0061) model_time 0.9293 (1.0055) loss 0.7590 (0.9258) grad_norm 8.2682 (8.8862/2.1764) mem 68106MB [2022-12-19 14:43:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1430/1519] eta 0:01:29 lr 0.000029 time 0.9278 (1.0061) model_time 0.9277 (1.0054) loss 0.9692 (0.9256) grad_norm 9.0224 (8.8989/2.1813) mem 68106MB [2022-12-19 14:43:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1440/1519] eta 0:01:19 lr 0.000029 time 0.9230 (1.0060) model_time 0.9229 (1.0054) loss 1.0258 (0.9251) grad_norm 7.7816 (8.8831/2.1648) mem 68106MB [2022-12-19 14:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1450/1519] eta 0:01:09 lr 0.000029 time 0.9313 (1.0060) model_time 0.9311 (1.0053) loss 0.6870 (0.9249) grad_norm 10.2823 (8.8954/2.1691) mem 68106MB [2022-12-19 14:44:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1460/1519] eta 0:00:59 lr 0.000029 time 0.9278 (1.0060) model_time 0.9276 (1.0053) loss 0.8634 (0.9249) grad_norm 7.5357 (8.9081/2.1655) mem 68106MB [2022-12-19 14:44:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1470/1519] eta 0:00:49 lr 0.000029 time 0.9374 (1.0060) model_time 0.9372 (1.0054) loss 0.7331 (0.9248) grad_norm 8.0922 (8.8836/2.1583) mem 68106MB [2022-12-19 14:44:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1480/1519] eta 0:00:39 lr 0.000029 time 0.9080 (1.0060) model_time 0.9078 (1.0054) loss 0.9223 (0.9251) grad_norm 8.1519 (8.8814/2.1592) mem 68106MB [2022-12-19 14:44:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1490/1519] eta 0:00:29 lr 0.000029 time 0.9734 (1.0060) model_time 0.9732 (1.0054) loss 1.1293 (0.9249) grad_norm 11.6707 (8.8814/2.1577) mem 68106MB [2022-12-19 14:44:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1500/1519] eta 0:00:19 lr 0.000029 time 0.9195 (1.0060) model_time 0.9194 (1.0053) loss 0.8369 (0.9253) grad_norm 9.4389 (8.9070/2.1617) mem 68106MB [2022-12-19 14:44:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [23/100][1510/1519] eta 0:00:09 lr 0.000029 time 0.9068 (1.0061) model_time 0.9067 (1.0055) loss 0.8071 (0.9255) grad_norm 6.8179 (8.9079/2.1662) mem 68106MB [2022-12-19 14:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 23 training takes 0:25:28 [2022-12-19 14:45:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_23.pth saving...... [2022-12-19 14:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_23.pth saved !!! [2022-12-19 14:45:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.643 (0.643) Loss 0.5073 (0.5073) Acc@1 90.278 (90.278) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 14:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.327) Loss 0.5255 (0.5075) Acc@1 90.972 (91.004) Acc@5 97.917 (98.201) Mem 68106MB [2022-12-19 14:45:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.313) Loss 0.4624 (0.5103) Acc@1 92.361 (90.642) Acc@5 98.958 (98.082) Mem 68106MB [2022-12-19 14:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.308) Loss 0.6043 (0.5152) Acc@1 89.236 (90.535) Acc@5 97.569 (97.995) Mem 68106MB [2022-12-19 14:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.305 (0.306) Loss 0.4788 (0.5054) Acc@1 90.625 (90.777) Acc@5 98.264 (98.086) Mem 68106MB [2022-12-19 14:45:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.306 (0.305) Loss 0.5276 (0.5019) Acc@1 86.458 (90.761) Acc@5 98.611 (98.189) Mem 68106MB [2022-12-19 14:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.304) Loss 0.5371 (0.5024) Acc@1 88.542 (90.739) Acc@5 98.264 (98.230) Mem 68106MB [2022-12-19 14:45:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5627 (0.5050) Acc@1 90.625 (90.674) Acc@5 98.264 (98.220) Mem 68106MB [2022-12-19 14:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4284 (0.5028) Acc@1 91.667 (90.685) Acc@5 98.611 (98.230) Mem 68106MB [2022-12-19 14:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:23] * Acc@1 90.668 Acc@5 98.228 [2022-12-19 14:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 90.7% [2022-12-19 14:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 14:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 14:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.67% [2022-12-19 14:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][0/1519] eta 0:35:54 lr 0.000029 time 1.4183 (1.4183) model_time 0.9757 (0.9757) loss 0.8210 (0.8210) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 14:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][10/1519] eta 0:26:06 lr 0.000029 time 0.9262 (1.0379) model_time 0.9261 (0.9974) loss 0.9528 (0.9256) grad_norm 13.3160 (8.9645/2.5539) mem 68106MB [2022-12-19 14:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][20/1519] eta 0:25:34 lr 0.000029 time 0.9355 (1.0237) model_time 0.9353 (1.0024) loss 0.8488 (0.8927) grad_norm 6.3092 (8.6188/2.1852) mem 68106MB [2022-12-19 14:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][30/1519] eta 0:25:13 lr 0.000029 time 0.9279 (1.0166) model_time 0.9276 (1.0020) loss 1.2757 (0.9196) grad_norm 7.0933 (8.2991/1.9399) mem 68106MB [2022-12-19 14:47:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][40/1519] eta 0:24:58 lr 0.000029 time 0.9361 (1.0130) model_time 0.9360 (1.0019) loss 0.7064 (0.9315) grad_norm 8.8674 (9.0942/2.7247) mem 68106MB [2022-12-19 14:47:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][50/1519] eta 0:24:46 lr 0.000029 time 0.9248 (1.0121) model_time 0.9247 (1.0031) loss 0.8410 (0.9223) grad_norm 10.6578 (9.1980/2.4913) mem 68106MB [2022-12-19 14:47:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][60/1519] eta 0:24:33 lr 0.000029 time 0.9317 (1.0098) model_time 0.9316 (1.0022) loss 0.9762 (0.9212) grad_norm 12.4966 (9.2694/2.6064) mem 68106MB [2022-12-19 14:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][70/1519] eta 0:24:23 lr 0.000029 time 0.9241 (1.0099) model_time 0.9239 (1.0033) loss 1.0179 (0.9245) grad_norm 7.9269 (9.0359/2.4848) mem 68106MB [2022-12-19 14:47:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][80/1519] eta 0:24:13 lr 0.000029 time 0.9240 (1.0098) model_time 0.9239 (1.0040) loss 1.0634 (0.9297) grad_norm 8.0265 (9.0742/2.4641) mem 68106MB [2022-12-19 14:47:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][90/1519] eta 0:24:02 lr 0.000029 time 0.9327 (1.0097) model_time 0.9325 (1.0045) loss 0.7596 (0.9374) grad_norm 12.4538 (9.0615/2.4187) mem 68106MB [2022-12-19 14:48:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][100/1519] eta 0:23:51 lr 0.000029 time 0.9320 (1.0086) model_time 0.9318 (1.0038) loss 0.9203 (0.9435) grad_norm 6.8955 (8.9905/2.3531) mem 68106MB [2022-12-19 14:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][110/1519] eta 0:23:40 lr 0.000029 time 0.9350 (1.0080) model_time 0.9348 (1.0037) loss 1.0357 (0.9364) grad_norm 9.3854 (9.0502/2.2780) mem 68106MB [2022-12-19 14:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][120/1519] eta 0:23:30 lr 0.000029 time 0.9321 (1.0079) model_time 0.9320 (1.0039) loss 1.2377 (0.9349) grad_norm 6.6579 (8.9082/2.2428) mem 68106MB [2022-12-19 14:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][130/1519] eta 0:23:21 lr 0.000029 time 0.9173 (1.0090) model_time 0.9171 (1.0052) loss 0.7455 (0.9270) grad_norm 13.0994 (9.0207/2.2878) mem 68106MB [2022-12-19 14:48:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][140/1519] eta 0:23:10 lr 0.000029 time 0.9229 (1.0083) model_time 0.9227 (1.0048) loss 0.7876 (0.9272) grad_norm 16.5800 (9.1214/2.3945) mem 68106MB [2022-12-19 14:48:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][150/1519] eta 0:23:00 lr 0.000029 time 0.9435 (1.0083) model_time 0.9434 (1.0050) loss 0.7800 (0.9272) grad_norm 9.2196 (9.0746/2.3376) mem 68106MB [2022-12-19 14:49:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][160/1519] eta 0:22:49 lr 0.000029 time 0.9257 (1.0080) model_time 0.9255 (1.0049) loss 0.9760 (0.9248) grad_norm 7.6011 (8.9541/2.3352) mem 68106MB [2022-12-19 14:49:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][170/1519] eta 0:22:39 lr 0.000029 time 0.9338 (1.0077) model_time 0.9337 (1.0047) loss 0.8607 (0.9252) grad_norm 8.2156 (8.9422/2.2730) mem 68106MB [2022-12-19 14:49:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][180/1519] eta 0:22:30 lr 0.000029 time 0.9316 (1.0084) model_time 0.9315 (1.0056) loss 1.0911 (0.9247) grad_norm 8.4617 (8.8593/2.2423) mem 68106MB [2022-12-19 14:49:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][190/1519] eta 0:22:19 lr 0.000029 time 0.9276 (1.0082) model_time 0.9274 (1.0055) loss 0.8017 (0.9220) grad_norm 8.7283 (8.8525/2.1940) mem 68106MB [2022-12-19 14:49:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][200/1519] eta 0:22:10 lr 0.000029 time 0.9340 (1.0084) model_time 0.9339 (1.0059) loss 0.8035 (0.9204) grad_norm 6.8043 (8.8520/2.1594) mem 68106MB [2022-12-19 14:49:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][210/1519] eta 0:21:59 lr 0.000029 time 0.9367 (1.0082) model_time 0.9365 (1.0058) loss 0.9262 (0.9230) grad_norm 6.3387 (8.8889/2.1778) mem 68106MB [2022-12-19 14:50:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][220/1519] eta 0:21:49 lr 0.000029 time 0.9306 (1.0080) model_time 0.9303 (1.0057) loss 1.0798 (0.9230) grad_norm 8.7338 (8.8779/2.1637) mem 68106MB [2022-12-19 14:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][230/1519] eta 0:21:39 lr 0.000029 time 0.9326 (1.0081) model_time 0.9324 (1.0059) loss 0.8496 (0.9208) grad_norm 8.2477 (8.8170/2.1409) mem 68106MB [2022-12-19 14:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][240/1519] eta 0:21:28 lr 0.000029 time 0.9171 (1.0077) model_time 0.9170 (1.0056) loss 0.8159 (0.9241) grad_norm 7.9315 (8.7873/2.1066) mem 68106MB [2022-12-19 14:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][250/1519] eta 0:21:18 lr 0.000029 time 0.9408 (1.0075) model_time 0.9406 (1.0053) loss 1.0680 (0.9250) grad_norm 9.6023 (8.7989/2.0905) mem 68106MB [2022-12-19 14:50:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][260/1519] eta 0:21:08 lr 0.000029 time 0.9325 (1.0072) model_time 0.9324 (1.0051) loss 1.0001 (0.9240) grad_norm 7.9249 (8.7697/2.0616) mem 68106MB [2022-12-19 14:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][270/1519] eta 0:20:58 lr 0.000029 time 0.9252 (1.0072) model_time 0.9251 (1.0053) loss 0.7178 (0.9228) grad_norm 8.3634 (8.7721/2.0356) mem 68106MB [2022-12-19 14:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][280/1519] eta 0:20:47 lr 0.000029 time 0.9315 (1.0071) model_time 0.9314 (1.0052) loss 0.9269 (0.9219) grad_norm 12.1879 (8.7957/2.0248) mem 68106MB [2022-12-19 14:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][290/1519] eta 0:20:37 lr 0.000029 time 0.9217 (1.0070) model_time 0.9216 (1.0051) loss 1.0424 (0.9225) grad_norm 7.8579 (8.7752/2.0282) mem 68106MB [2022-12-19 14:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][300/1519] eta 0:20:27 lr 0.000029 time 0.9223 (1.0067) model_time 0.9222 (1.0049) loss 0.7998 (0.9249) grad_norm 7.1888 (8.8010/2.0145) mem 68106MB [2022-12-19 14:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][310/1519] eta 0:20:17 lr 0.000029 time 0.9204 (1.0068) model_time 0.9203 (1.0050) loss 0.7433 (0.9267) grad_norm 11.7545 (8.8232/2.0116) mem 68106MB [2022-12-19 14:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][320/1519] eta 0:20:07 lr 0.000029 time 0.9272 (1.0068) model_time 0.9271 (1.0051) loss 0.7045 (0.9237) grad_norm 8.7907 (8.7784/2.0006) mem 68106MB [2022-12-19 14:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][330/1519] eta 0:19:57 lr 0.000029 time 0.9236 (1.0069) model_time 0.9235 (1.0052) loss 0.9575 (0.9244) grad_norm 7.0644 (8.7467/1.9900) mem 68106MB [2022-12-19 14:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][340/1519] eta 0:19:47 lr 0.000029 time 0.9265 (1.0069) model_time 0.9263 (1.0053) loss 1.0733 (0.9241) grad_norm 8.0205 (8.7413/2.0041) mem 68106MB [2022-12-19 14:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][350/1519] eta 0:19:36 lr 0.000029 time 0.9211 (1.0067) model_time 0.9209 (1.0051) loss 0.7274 (0.9256) grad_norm 7.9341 (8.7211/1.9881) mem 68106MB [2022-12-19 14:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][360/1519] eta 0:19:27 lr 0.000029 time 0.9228 (1.0070) model_time 0.9227 (1.0054) loss 1.0271 (0.9269) grad_norm 7.3418 (8.7515/1.9890) mem 68106MB [2022-12-19 14:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][370/1519] eta 0:19:16 lr 0.000029 time 0.9222 (1.0068) model_time 0.9221 (1.0053) loss 0.7394 (0.9267) grad_norm 6.3040 (8.7386/1.9771) mem 68106MB [2022-12-19 14:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][380/1519] eta 0:19:07 lr 0.000029 time 0.9985 (1.0072) model_time 0.9984 (1.0057) loss 0.7069 (0.9266) grad_norm 9.2081 (8.7219/1.9590) mem 68106MB [2022-12-19 14:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][390/1519] eta 0:18:57 lr 0.000029 time 0.9258 (1.0072) model_time 0.9256 (1.0057) loss 0.9080 (0.9271) grad_norm 7.0971 (8.6878/1.9486) mem 68106MB [2022-12-19 14:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][400/1519] eta 0:18:47 lr 0.000029 time 0.9271 (1.0072) model_time 0.9270 (1.0058) loss 0.7580 (0.9278) grad_norm 7.1442 (8.6555/1.9436) mem 68106MB [2022-12-19 14:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][410/1519] eta 0:18:36 lr 0.000029 time 0.9314 (1.0071) model_time 0.9312 (1.0057) loss 0.8371 (0.9254) grad_norm 7.1033 (8.6500/1.9460) mem 68106MB [2022-12-19 14:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][420/1519] eta 0:18:26 lr 0.000029 time 0.9289 (1.0069) model_time 0.9287 (1.0055) loss 0.8509 (0.9269) grad_norm 10.5552 (8.6257/1.9577) mem 68106MB [2022-12-19 14:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][430/1519] eta 0:18:17 lr 0.000029 time 0.8936 (1.0074) model_time 0.8934 (1.0060) loss 1.0771 (0.9256) grad_norm 7.0728 (8.6321/1.9987) mem 68106MB [2022-12-19 14:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][440/1519] eta 0:18:07 lr 0.000029 time 1.2285 (1.0079) model_time 1.2283 (1.0066) loss 1.0498 (0.9257) grad_norm 8.4018 (8.6294/1.9829) mem 68106MB [2022-12-19 14:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][450/1519] eta 0:17:57 lr 0.000029 time 0.9385 (1.0079) model_time 0.9383 (1.0066) loss 1.1673 (0.9256) grad_norm 12.7737 (8.6513/1.9919) mem 68106MB [2022-12-19 14:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][460/1519] eta 0:17:47 lr 0.000029 time 0.9163 (1.0079) model_time 0.9162 (1.0066) loss 0.8071 (0.9260) grad_norm 13.0366 (8.6654/2.0177) mem 68106MB [2022-12-19 14:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][470/1519] eta 0:17:37 lr 0.000029 time 0.9257 (1.0079) model_time 0.9255 (1.0066) loss 0.9837 (0.9249) grad_norm 8.5927 (8.6713/2.0089) mem 68106MB [2022-12-19 14:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][480/1519] eta 0:17:27 lr 0.000029 time 0.9307 (1.0078) model_time 0.9304 (1.0065) loss 0.7001 (0.9230) grad_norm 8.4309 (8.6784/2.0049) mem 68106MB [2022-12-19 14:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][490/1519] eta 0:17:16 lr 0.000029 time 0.9319 (1.0077) model_time 0.9317 (1.0065) loss 0.8317 (0.9213) grad_norm 8.5926 (8.6658/1.9893) mem 68106MB [2022-12-19 14:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][500/1519] eta 0:17:06 lr 0.000029 time 0.9328 (1.0076) model_time 0.9327 (1.0064) loss 0.8983 (0.9203) grad_norm 7.3914 (8.6548/1.9798) mem 68106MB [2022-12-19 14:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][510/1519] eta 0:16:56 lr 0.000029 time 0.9384 (1.0074) model_time 0.9383 (1.0062) loss 0.7452 (0.9187) grad_norm 6.8881 (8.6418/1.9707) mem 68106MB [2022-12-19 14:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][520/1519] eta 0:16:46 lr 0.000029 time 0.9316 (1.0073) model_time 0.9315 (1.0061) loss 1.1825 (0.9189) grad_norm 9.5446 (8.6562/1.9789) mem 68106MB [2022-12-19 14:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][530/1519] eta 0:16:36 lr 0.000029 time 0.9275 (1.0072) model_time 0.9271 (1.0060) loss 0.9189 (0.9191) grad_norm 6.9167 (8.6302/1.9737) mem 68106MB [2022-12-19 14:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][540/1519] eta 0:16:26 lr 0.000029 time 0.9291 (1.0072) model_time 0.9290 (1.0060) loss 0.9039 (0.9191) grad_norm 7.1861 (8.6570/1.9851) mem 68106MB [2022-12-19 14:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][550/1519] eta 0:16:15 lr 0.000029 time 0.9309 (1.0070) model_time 0.9307 (1.0059) loss 0.7908 (0.9183) grad_norm 9.4564 (8.6538/1.9709) mem 68106MB [2022-12-19 14:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][560/1519] eta 0:16:05 lr 0.000029 time 1.0067 (1.0071) model_time 1.0066 (1.0060) loss 1.4513 (0.9189) grad_norm 7.0987 (8.6467/1.9749) mem 68106MB [2022-12-19 14:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][570/1519] eta 0:15:55 lr 0.000029 time 0.9340 (1.0071) model_time 0.9338 (1.0060) loss 1.0021 (0.9192) grad_norm 10.7311 (8.7131/2.0965) mem 68106MB [2022-12-19 14:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][580/1519] eta 0:15:45 lr 0.000029 time 0.9195 (1.0074) model_time 0.9192 (1.0063) loss 0.7881 (0.9182) grad_norm 10.6757 (8.7060/2.0909) mem 68106MB [2022-12-19 14:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][590/1519] eta 0:15:35 lr 0.000029 time 0.9295 (1.0073) model_time 0.9294 (1.0062) loss 0.9303 (0.9177) grad_norm 7.1706 (8.6998/2.0762) mem 68106MB [2022-12-19 14:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][600/1519] eta 0:15:25 lr 0.000029 time 0.9357 (1.0072) model_time 0.9355 (1.0061) loss 0.7934 (0.9177) grad_norm 7.3886 (8.6889/2.0668) mem 68106MB [2022-12-19 14:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][610/1519] eta 0:15:15 lr 0.000029 time 0.9326 (1.0071) model_time 0.9324 (1.0060) loss 1.0465 (0.9163) grad_norm 7.1973 (8.6877/2.0576) mem 68106MB [2022-12-19 14:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][620/1519] eta 0:15:05 lr 0.000029 time 1.0233 (1.0072) model_time 1.0232 (1.0061) loss 0.9373 (0.9170) grad_norm 7.8752 (8.6914/2.0490) mem 68106MB [2022-12-19 14:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][630/1519] eta 0:14:55 lr 0.000029 time 0.9296 (1.0070) model_time 0.9295 (1.0060) loss 0.8652 (0.9165) grad_norm 11.2415 (8.7253/2.0483) mem 68106MB [2022-12-19 14:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][640/1519] eta 0:14:45 lr 0.000029 time 0.9358 (1.0070) model_time 0.9357 (1.0060) loss 0.8916 (0.9164) grad_norm 11.6627 (8.6777/1.9837) mem 68106MB [2022-12-19 14:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][650/1519] eta 0:14:35 lr 0.000029 time 0.9099 (1.0070) model_time 0.9097 (1.0059) loss 1.2774 (0.9175) grad_norm 12.3778 (8.6680/1.9893) mem 68106MB [2022-12-19 14:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][660/1519] eta 0:14:24 lr 0.000029 time 0.9256 (1.0069) model_time 0.9255 (1.0059) loss 1.0216 (0.9163) grad_norm 12.5967 (8.6563/1.9718) mem 68106MB [2022-12-19 14:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][670/1519] eta 0:14:15 lr 0.000029 time 0.9288 (1.0071) model_time 0.9286 (1.0061) loss 1.2645 (0.9167) grad_norm 7.2718 (8.6807/1.9759) mem 68106MB [2022-12-19 14:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][680/1519] eta 0:14:04 lr 0.000029 time 0.9318 (1.0070) model_time 0.9316 (1.0060) loss 0.7651 (0.9178) grad_norm 10.6815 (8.6822/1.9615) mem 68106MB [2022-12-19 14:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][690/1519] eta 0:13:54 lr 0.000029 time 0.9076 (1.0071) model_time 0.9075 (1.0061) loss 1.1171 (0.9177) grad_norm 16.4235 (8.7146/2.0027) mem 68106MB [2022-12-19 14:58:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][700/1519] eta 0:13:44 lr 0.000029 time 0.9344 (1.0070) model_time 0.9343 (1.0061) loss 0.9792 (0.9178) grad_norm 9.2758 (8.7334/1.9970) mem 68106MB [2022-12-19 14:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][710/1519] eta 0:13:34 lr 0.000029 time 1.0218 (1.0071) model_time 1.0216 (1.0061) loss 0.6885 (0.9186) grad_norm 11.3202 (8.7235/2.0029) mem 68106MB [2022-12-19 14:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][720/1519] eta 0:13:24 lr 0.000029 time 0.9218 (1.0070) model_time 0.9216 (1.0061) loss 0.8287 (0.9181) grad_norm 15.4857 (8.7774/2.0387) mem 68106MB [2022-12-19 14:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][730/1519] eta 0:13:14 lr 0.000029 time 0.9344 (1.0070) model_time 0.9342 (1.0060) loss 1.1341 (0.9185) grad_norm 6.9497 (8.7331/2.0111) mem 68106MB [2022-12-19 14:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][740/1519] eta 0:13:04 lr 0.000029 time 0.9397 (1.0069) model_time 0.9396 (1.0060) loss 0.8459 (0.9187) grad_norm 8.0295 (8.7078/1.9632) mem 68106MB [2022-12-19 14:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][750/1519] eta 0:12:54 lr 0.000029 time 0.9328 (1.0069) model_time 0.9327 (1.0060) loss 0.8330 (0.9196) grad_norm 11.6388 (8.7352/1.9743) mem 68106MB [2022-12-19 14:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][760/1519] eta 0:12:44 lr 0.000029 time 0.9328 (1.0072) model_time 0.9325 (1.0062) loss 1.1627 (0.9197) grad_norm 8.3594 (8.7922/1.9809) mem 68106MB [2022-12-19 14:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][770/1519] eta 0:12:34 lr 0.000029 time 0.9307 (1.0072) model_time 0.9305 (1.0062) loss 0.8428 (0.9202) grad_norm 8.5505 (8.7952/1.9839) mem 68106MB [2022-12-19 14:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][780/1519] eta 0:12:24 lr 0.000029 time 0.9339 (1.0072) model_time 0.9337 (1.0063) loss 0.7555 (0.9191) grad_norm 7.9028 (8.8200/2.0027) mem 68106MB [2022-12-19 14:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][790/1519] eta 0:12:14 lr 0.000029 time 0.9367 (1.0071) model_time 0.9365 (1.0062) loss 1.4796 (0.9196) grad_norm 9.5334 (8.8455/2.0147) mem 68106MB [2022-12-19 14:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][800/1519] eta 0:12:04 lr 0.000029 time 0.9322 (1.0070) model_time 0.9321 (1.0061) loss 0.9499 (0.9187) grad_norm 6.6109 (8.8258/2.0210) mem 68106MB [2022-12-19 14:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][810/1519] eta 0:11:53 lr 0.000029 time 0.9365 (1.0070) model_time 0.9364 (1.0061) loss 0.7929 (0.9184) grad_norm 11.2647 (8.8046/2.0067) mem 68106MB [2022-12-19 15:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][820/1519] eta 0:11:43 lr 0.000029 time 0.9379 (1.0070) model_time 0.9377 (1.0061) loss 0.7585 (0.9193) grad_norm 9.2681 (8.8009/1.9953) mem 68106MB [2022-12-19 15:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][830/1519] eta 0:11:33 lr 0.000029 time 0.9329 (1.0068) model_time 0.9327 (1.0060) loss 0.7089 (0.9197) grad_norm 9.3268 (8.8138/1.9900) mem 68106MB [2022-12-19 15:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][840/1519] eta 0:11:23 lr 0.000029 time 0.9339 (1.0068) model_time 0.9337 (1.0059) loss 1.0938 (0.9208) grad_norm 5.6915 (8.7829/2.0134) mem 68106MB [2022-12-19 15:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][850/1519] eta 0:11:13 lr 0.000029 time 1.0075 (1.0068) model_time 1.0074 (1.0059) loss 0.9134 (0.9200) grad_norm 8.3393 (8.7883/2.0061) mem 68106MB [2022-12-19 15:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][860/1519] eta 0:11:03 lr 0.000029 time 0.9270 (1.0068) model_time 0.9268 (1.0059) loss 0.7386 (0.9190) grad_norm 9.3641 (8.8148/2.0166) mem 68106MB [2022-12-19 15:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][870/1519] eta 0:10:53 lr 0.000029 time 0.9344 (1.0068) model_time 0.9339 (1.0059) loss 0.8189 (0.9191) grad_norm 10.6176 (8.8141/2.0219) mem 68106MB [2022-12-19 15:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][880/1519] eta 0:10:43 lr 0.000029 time 0.9077 (1.0070) model_time 0.9075 (1.0061) loss 1.2006 (0.9212) grad_norm 7.0323 (8.8077/2.0208) mem 68106MB [2022-12-19 15:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][890/1519] eta 0:10:33 lr 0.000029 time 1.0367 (1.0070) model_time 1.0365 (1.0061) loss 0.7010 (0.9216) grad_norm 8.9502 (8.8052/2.0074) mem 68106MB [2022-12-19 15:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][900/1519] eta 0:10:23 lr 0.000029 time 0.9321 (1.0069) model_time 0.9320 (1.0060) loss 0.9725 (0.9218) grad_norm 6.6720 (8.7772/2.0105) mem 68106MB [2022-12-19 15:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][910/1519] eta 0:10:13 lr 0.000029 time 0.9307 (1.0068) model_time 0.9306 (1.0059) loss 0.9319 (0.9221) grad_norm 8.0606 (8.7750/2.0118) mem 68106MB [2022-12-19 15:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][920/1519] eta 0:10:03 lr 0.000029 time 0.9384 (1.0067) model_time 0.9383 (1.0059) loss 1.0078 (0.9215) grad_norm 7.9546 (8.8225/2.0478) mem 68106MB [2022-12-19 15:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][930/1519] eta 0:09:52 lr 0.000029 time 0.9346 (1.0067) model_time 0.9344 (1.0058) loss 0.9225 (0.9216) grad_norm 7.9555 (8.8259/2.0409) mem 68106MB [2022-12-19 15:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][940/1519] eta 0:09:42 lr 0.000029 time 0.9487 (1.0068) model_time 0.9484 (1.0060) loss 0.9283 (0.9213) grad_norm 11.0243 (8.8361/2.0330) mem 68106MB [2022-12-19 15:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][950/1519] eta 0:09:32 lr 0.000029 time 1.0379 (1.0068) model_time 1.0378 (1.0060) loss 0.7828 (0.9216) grad_norm 7.3253 (8.8640/2.0481) mem 68106MB [2022-12-19 15:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][960/1519] eta 0:09:22 lr 0.000029 time 0.9281 (1.0068) model_time 0.9280 (1.0059) loss 1.1254 (0.9209) grad_norm 6.4419 (8.8261/2.0425) mem 68106MB [2022-12-19 15:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][970/1519] eta 0:09:12 lr 0.000029 time 0.9210 (1.0067) model_time 0.9207 (1.0058) loss 1.0220 (0.9204) grad_norm 12.7143 (8.8383/2.0482) mem 68106MB [2022-12-19 15:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][980/1519] eta 0:09:02 lr 0.000029 time 0.9281 (1.0066) model_time 0.9280 (1.0058) loss 0.8835 (0.9204) grad_norm 10.9719 (8.8499/2.0505) mem 68106MB [2022-12-19 15:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][990/1519] eta 0:08:52 lr 0.000029 time 0.9378 (1.0067) model_time 0.9377 (1.0059) loss 0.7358 (0.9199) grad_norm 10.5482 (8.8792/2.0472) mem 68106MB [2022-12-19 15:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1000/1519] eta 0:08:42 lr 0.000029 time 0.9548 (1.0067) model_time 0.9546 (1.0059) loss 0.8259 (0.9202) grad_norm 8.7711 (8.8990/2.0525) mem 68106MB [2022-12-19 15:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1010/1519] eta 0:08:32 lr 0.000029 time 0.9377 (1.0067) model_time 0.9376 (1.0059) loss 0.7617 (0.9200) grad_norm 6.8144 (8.9194/2.0518) mem 68106MB [2022-12-19 15:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1020/1519] eta 0:08:22 lr 0.000029 time 0.9204 (1.0067) model_time 0.9203 (1.0059) loss 1.1691 (0.9210) grad_norm 7.9141 (8.9442/2.0396) mem 68106MB [2022-12-19 15:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1030/1519] eta 0:08:12 lr 0.000029 time 1.2666 (1.0069) model_time 1.2664 (1.0061) loss 1.0032 (0.9214) grad_norm 8.9014 (8.9198/2.0097) mem 68106MB [2022-12-19 15:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1040/1519] eta 0:08:02 lr 0.000029 time 0.9304 (1.0069) model_time 0.9303 (1.0061) loss 1.0161 (0.9214) grad_norm 5.5099 (8.8953/2.0272) mem 68106MB [2022-12-19 15:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1050/1519] eta 0:07:52 lr 0.000029 time 0.9279 (1.0068) model_time 0.9278 (1.0060) loss 1.1267 (0.9216) grad_norm 11.0656 (8.8805/2.0463) mem 68106MB [2022-12-19 15:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1060/1519] eta 0:07:42 lr 0.000029 time 0.9251 (1.0070) model_time 0.9249 (1.0062) loss 0.7512 (0.9215) grad_norm 7.0501 (8.8775/2.0886) mem 68106MB [2022-12-19 15:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1070/1519] eta 0:07:32 lr 0.000029 time 0.9877 (1.0070) model_time 0.9875 (1.0062) loss 0.8386 (0.9211) grad_norm 8.5250 (8.8599/2.0902) mem 68106MB [2022-12-19 15:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1080/1519] eta 0:07:22 lr 0.000029 time 0.9287 (1.0070) model_time 0.9286 (1.0062) loss 0.9509 (0.9216) grad_norm 8.7972 (8.8583/2.0790) mem 68106MB [2022-12-19 15:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1090/1519] eta 0:07:12 lr 0.000029 time 0.9170 (1.0070) model_time 0.9168 (1.0062) loss 1.0554 (0.9221) grad_norm 11.6390 (8.8869/2.0978) mem 68106MB [2022-12-19 15:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1100/1519] eta 0:07:01 lr 0.000029 time 0.9298 (1.0069) model_time 0.9296 (1.0061) loss 0.9577 (0.9223) grad_norm 9.6310 (8.9003/2.0978) mem 68106MB [2022-12-19 15:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1110/1519] eta 0:06:51 lr 0.000029 time 0.9324 (1.0069) model_time 0.9322 (1.0061) loss 0.6830 (0.9223) grad_norm 7.5485 (8.9164/2.0951) mem 68106MB [2022-12-19 15:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1120/1519] eta 0:06:41 lr 0.000029 time 0.9262 (1.0068) model_time 0.9261 (1.0061) loss 0.7297 (0.9230) grad_norm 8.4410 (8.9160/2.0935) mem 68106MB [2022-12-19 15:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1130/1519] eta 0:06:31 lr 0.000029 time 1.0094 (1.0068) model_time 1.0091 (1.0061) loss 0.7073 (0.9218) grad_norm 5.4157 (8.9117/2.0964) mem 68106MB [2022-12-19 15:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1140/1519] eta 0:06:21 lr 0.000029 time 0.9271 (1.0069) model_time 0.9270 (1.0061) loss 1.0565 (0.9218) grad_norm 14.0042 (8.9042/2.0973) mem 68106MB [2022-12-19 15:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1150/1519] eta 0:06:11 lr 0.000029 time 0.9334 (1.0068) model_time 0.9332 (1.0060) loss 1.0269 (0.9223) grad_norm 9.1956 (8.8995/2.0966) mem 68106MB [2022-12-19 15:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1160/1519] eta 0:06:01 lr 0.000029 time 0.9397 (1.0067) model_time 0.9394 (1.0060) loss 1.4084 (0.9229) grad_norm 8.1128 (8.9040/2.0784) mem 68106MB [2022-12-19 15:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1170/1519] eta 0:05:51 lr 0.000029 time 0.9395 (1.0067) model_time 0.9394 (1.0060) loss 1.1031 (0.9229) grad_norm 6.8300 (8.8553/1.9883) mem 68106MB [2022-12-19 15:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1180/1519] eta 0:05:41 lr 0.000029 time 0.9593 (1.0067) model_time 0.9592 (1.0060) loss 0.9098 (0.9227) grad_norm 8.7598 (8.8608/1.9850) mem 68106MB [2022-12-19 15:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1190/1519] eta 0:05:31 lr 0.000029 time 0.9314 (1.0067) model_time 0.9313 (1.0059) loss 1.1254 (0.9231) grad_norm 6.6920 (8.8558/1.9869) mem 68106MB [2022-12-19 15:06:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1200/1519] eta 0:05:21 lr 0.000029 time 0.9340 (1.0067) model_time 0.9339 (1.0060) loss 0.8806 (0.9228) grad_norm 8.9943 (8.8614/1.9837) mem 68106MB [2022-12-19 15:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1210/1519] eta 0:05:11 lr 0.000029 time 0.9304 (1.0068) model_time 0.9302 (1.0060) loss 0.8247 (0.9230) grad_norm 12.4823 (8.8572/1.9816) mem 68106MB [2022-12-19 15:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1220/1519] eta 0:05:01 lr 0.000029 time 0.9387 (1.0067) model_time 0.9386 (1.0060) loss 0.7051 (0.9227) grad_norm 8.4107 (8.8536/1.9849) mem 68106MB [2022-12-19 15:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1230/1519] eta 0:04:50 lr 0.000029 time 0.9314 (1.0067) model_time 0.9312 (1.0060) loss 1.1283 (0.9230) grad_norm 9.3501 (8.8309/1.9825) mem 68106MB [2022-12-19 15:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1240/1519] eta 0:04:40 lr 0.000029 time 0.9348 (1.0067) model_time 0.9347 (1.0060) loss 0.8136 (0.9223) grad_norm 8.5446 (8.8369/1.9779) mem 68106MB [2022-12-19 15:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1250/1519] eta 0:04:30 lr 0.000029 time 0.9171 (1.0071) model_time 0.9168 (1.0064) loss 0.7172 (0.9226) grad_norm 6.4671 (8.8112/1.9800) mem 68106MB [2022-12-19 15:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1260/1519] eta 0:04:20 lr 0.000029 time 0.9162 (1.0071) model_time 0.9160 (1.0063) loss 1.1936 (0.9228) grad_norm 10.7059 (8.7904/1.9689) mem 68106MB [2022-12-19 15:07:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1270/1519] eta 0:04:10 lr 0.000029 time 0.9342 (1.0071) model_time 0.9340 (1.0064) loss 1.4844 (0.9237) grad_norm 8.5269 (8.8033/2.0069) mem 68106MB [2022-12-19 15:07:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1280/1519] eta 0:04:00 lr 0.000029 time 0.9377 (1.0071) model_time 0.9375 (1.0064) loss 0.9210 (0.9233) grad_norm 9.8320 (8.7733/2.0121) mem 68106MB [2022-12-19 15:07:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1290/1519] eta 0:03:50 lr 0.000029 time 0.9451 (1.0071) model_time 0.9448 (1.0063) loss 0.9833 (0.9238) grad_norm 7.9904 (8.7549/1.9973) mem 68106MB [2022-12-19 15:08:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1300/1519] eta 0:03:40 lr 0.000029 time 0.9390 (1.0071) model_time 0.9389 (1.0064) loss 1.2771 (0.9244) grad_norm 5.2887 (8.7311/2.0089) mem 68106MB [2022-12-19 15:08:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1310/1519] eta 0:03:30 lr 0.000029 time 0.9295 (1.0072) model_time 0.9293 (1.0064) loss 0.8161 (0.9244) grad_norm 10.1506 (8.7148/2.0094) mem 68106MB [2022-12-19 15:08:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1320/1519] eta 0:03:20 lr 0.000029 time 0.9314 (1.0072) model_time 0.9313 (1.0065) loss 1.2373 (0.9244) grad_norm 8.8414 (8.7044/1.9768) mem 68106MB [2022-12-19 15:08:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1330/1519] eta 0:03:10 lr 0.000029 time 0.9461 (1.0072) model_time 0.9460 (1.0065) loss 1.1170 (0.9243) grad_norm 10.1755 (8.7483/1.9838) mem 68106MB [2022-12-19 15:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1340/1519] eta 0:03:00 lr 0.000029 time 0.9568 (1.0071) model_time 0.9567 (1.0064) loss 0.7679 (0.9240) grad_norm 7.3123 (8.7351/1.9876) mem 68106MB [2022-12-19 15:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1350/1519] eta 0:02:50 lr 0.000029 time 0.9290 (1.0072) model_time 0.9288 (1.0064) loss 0.8329 (0.9237) grad_norm 8.1368 (8.7290/2.0194) mem 68106MB [2022-12-19 15:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1360/1519] eta 0:02:40 lr 0.000029 time 0.9173 (1.0071) model_time 0.9171 (1.0064) loss 0.7276 (0.9240) grad_norm 7.7919 (8.6920/2.0103) mem 68106MB [2022-12-19 15:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1370/1519] eta 0:02:30 lr 0.000029 time 0.9299 (1.0070) model_time 0.9297 (1.0063) loss 0.8123 (0.9235) grad_norm 9.7376 (8.6878/2.0077) mem 68106MB [2022-12-19 15:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1380/1519] eta 0:02:19 lr 0.000029 time 0.9303 (1.0070) model_time 0.9300 (1.0063) loss 0.9067 (0.9235) grad_norm 6.7200 (8.6842/2.0034) mem 68106MB [2022-12-19 15:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1390/1519] eta 0:02:09 lr 0.000029 time 1.0381 (1.0070) model_time 1.0380 (1.0063) loss 0.8176 (0.9231) grad_norm 13.1322 (8.6758/2.0063) mem 68106MB [2022-12-19 15:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1400/1519] eta 0:01:59 lr 0.000029 time 0.9316 (1.0070) model_time 0.9314 (1.0063) loss 0.8866 (0.9231) grad_norm 7.4827 (8.6860/2.0040) mem 68106MB [2022-12-19 15:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1410/1519] eta 0:01:49 lr 0.000029 time 0.9297 (1.0070) model_time 0.9290 (1.0063) loss 0.9050 (0.9236) grad_norm 7.0645 (8.6614/2.0067) mem 68106MB [2022-12-19 15:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1420/1519] eta 0:01:39 lr 0.000029 time 0.9856 (1.0070) model_time 0.9852 (1.0063) loss 0.7912 (0.9236) grad_norm 9.3233 (8.6889/2.0275) mem 68106MB [2022-12-19 15:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1430/1519] eta 0:01:29 lr 0.000029 time 0.9368 (1.0069) model_time 0.9366 (1.0062) loss 0.8110 (0.9234) grad_norm 11.5215 (8.6989/2.0438) mem 68106MB [2022-12-19 15:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1440/1519] eta 0:01:19 lr 0.000029 time 0.9391 (1.0069) model_time 0.9389 (1.0062) loss 1.0974 (0.9232) grad_norm 13.5333 (8.7476/2.0467) mem 68106MB [2022-12-19 15:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1450/1519] eta 0:01:09 lr 0.000029 time 0.9380 (1.0069) model_time 0.9378 (1.0062) loss 0.9724 (0.9231) grad_norm 9.8253 (8.7319/2.0462) mem 68106MB [2022-12-19 15:10:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1460/1519] eta 0:00:59 lr 0.000029 time 0.9380 (1.0069) model_time 0.9378 (1.0062) loss 0.9531 (0.9230) grad_norm 8.9161 (8.7303/2.0493) mem 68106MB [2022-12-19 15:11:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1470/1519] eta 0:00:49 lr 0.000029 time 0.9282 (1.0068) model_time 0.9280 (1.0061) loss 0.7318 (0.9233) grad_norm 8.7078 (8.7266/2.0453) mem 68106MB [2022-12-19 15:11:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1480/1519] eta 0:00:39 lr 0.000029 time 0.9340 (1.0069) model_time 0.9339 (1.0062) loss 0.7920 (0.9241) grad_norm 6.4816 (8.7053/2.0447) mem 68106MB [2022-12-19 15:11:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1490/1519] eta 0:00:29 lr 0.000029 time 0.9307 (1.0069) model_time 0.9305 (1.0062) loss 0.9601 (0.9243) grad_norm 6.5734 (8.7083/2.0510) mem 68106MB [2022-12-19 15:11:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1500/1519] eta 0:00:19 lr 0.000029 time 0.9295 (1.0069) model_time 0.9293 (1.0062) loss 1.3990 (0.9250) grad_norm 9.0107 (8.7483/2.0566) mem 68106MB [2022-12-19 15:11:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [24/100][1510/1519] eta 0:00:09 lr 0.000029 time 0.9323 (1.0069) model_time 0.9322 (1.0062) loss 1.3692 (0.9257) grad_norm 9.7916 (8.7443/2.0521) mem 68106MB [2022-12-19 15:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 24 training takes 0:25:29 [2022-12-19 15:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_24.pth saving...... [2022-12-19 15:12:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_24.pth saved !!! [2022-12-19 15:12:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.728 (0.728) Loss 0.4748 (0.4748) Acc@1 92.708 (92.708) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 15:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.337) Loss 0.5299 (0.4907) Acc@1 90.625 (91.477) Acc@5 98.264 (98.359) Mem 68106MB [2022-12-19 15:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.319) Loss 0.4459 (0.4905) Acc@1 92.014 (91.088) Acc@5 98.611 (98.280) Mem 68106MB [2022-12-19 15:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.312) Loss 0.6200 (0.4963) Acc@1 88.542 (90.871) Acc@5 97.222 (98.185) Mem 68106MB [2022-12-19 15:12:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.309) Loss 0.4651 (0.4872) Acc@1 90.278 (90.947) Acc@5 98.958 (98.264) Mem 68106MB [2022-12-19 15:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.307) Loss 0.4942 (0.4850) Acc@1 89.583 (90.945) Acc@5 98.958 (98.305) Mem 68106MB [2022-12-19 15:12:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.306) Loss 0.5239 (0.4855) Acc@1 87.847 (90.927) Acc@5 98.264 (98.349) Mem 68106MB [2022-12-19 15:12:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.305) Loss 0.5716 (0.4875) Acc@1 89.236 (90.865) Acc@5 98.264 (98.352) Mem 68106MB [2022-12-19 15:12:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.304) Loss 0.4166 (0.4856) Acc@1 93.056 (90.959) Acc@5 98.958 (98.371) Mem 68106MB [2022-12-19 15:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:24] * Acc@1 90.934 Acc@5 98.379 [2022-12-19 15:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 90.9% [2022-12-19 15:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 15:13:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 15:13:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.93% [2022-12-19 15:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][0/1519] eta 0:37:19 lr 0.000029 time 1.4744 (1.4744) model_time 0.9553 (0.9553) loss 1.0352 (1.0352) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 15:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][10/1519] eta 0:26:15 lr 0.000029 time 0.9349 (1.0439) model_time 0.9348 (0.9963) loss 0.9126 (0.8953) grad_norm 9.0721 (7.9849/1.3455) mem 68106MB [2022-12-19 15:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][20/1519] eta 0:25:40 lr 0.000029 time 1.0377 (1.0278) model_time 1.0375 (1.0026) loss 0.7457 (0.9165) grad_norm 7.5775 (8.1067/1.0585) mem 68106MB [2022-12-19 15:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][30/1519] eta 0:25:16 lr 0.000029 time 0.9407 (1.0184) model_time 0.9403 (1.0012) loss 1.0188 (0.9397) grad_norm 6.9607 (8.0992/0.9463) mem 68106MB [2022-12-19 15:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][40/1519] eta 0:25:01 lr 0.000029 time 0.9924 (1.0150) model_time 0.9923 (1.0018) loss 0.8720 (0.9309) grad_norm 9.5624 (8.2654/0.9100) mem 68106MB [2022-12-19 15:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][50/1519] eta 0:24:48 lr 0.000029 time 0.9376 (1.0130) model_time 0.9375 (1.0023) loss 1.3704 (0.9266) grad_norm 8.7960 (8.1343/0.9451) mem 68106MB [2022-12-19 15:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][60/1519] eta 0:24:37 lr 0.000029 time 0.9518 (1.0129) model_time 0.9516 (1.0039) loss 0.7982 (0.9312) grad_norm 10.5616 (8.3121/1.1198) mem 68106MB [2022-12-19 15:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][70/1519] eta 0:24:25 lr 0.000029 time 0.9377 (1.0111) model_time 0.9375 (1.0033) loss 0.7670 (0.9220) grad_norm 7.4160 (8.3436/1.0783) mem 68106MB [2022-12-19 15:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][80/1519] eta 0:24:13 lr 0.000029 time 0.9342 (1.0098) model_time 0.9341 (1.0029) loss 0.7970 (0.9136) grad_norm 7.1947 (8.3236/1.1984) mem 68106MB [2022-12-19 15:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][90/1519] eta 0:24:02 lr 0.000029 time 0.9314 (1.0094) model_time 0.9313 (1.0033) loss 0.8829 (0.9079) grad_norm 8.1884 (8.3469/1.2152) mem 68106MB [2022-12-19 15:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][100/1519] eta 0:23:50 lr 0.000029 time 0.9384 (1.0083) model_time 0.9382 (1.0028) loss 1.1416 (0.9182) grad_norm 6.7247 (8.3933/1.3377) mem 68106MB [2022-12-19 15:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][110/1519] eta 0:23:39 lr 0.000029 time 0.9309 (1.0077) model_time 0.9307 (1.0026) loss 0.9258 (0.9137) grad_norm 8.8391 (8.3717/1.3213) mem 68106MB [2022-12-19 15:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][120/1519] eta 0:23:28 lr 0.000029 time 0.9584 (1.0071) model_time 0.9579 (1.0024) loss 1.0320 (0.9066) grad_norm 9.0524 (8.7195/1.8746) mem 68106MB [2022-12-19 15:15:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][130/1519] eta 0:23:18 lr 0.000029 time 0.9290 (1.0068) model_time 0.9289 (1.0024) loss 0.7527 (0.9034) grad_norm 11.9780 (8.7687/1.8693) mem 68106MB [2022-12-19 15:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][140/1519] eta 0:23:08 lr 0.000029 time 0.9369 (1.0066) model_time 0.9366 (1.0025) loss 0.8635 (0.9039) grad_norm 9.3177 (8.8476/1.8608) mem 68106MB [2022-12-19 15:15:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][150/1519] eta 0:22:58 lr 0.000029 time 0.9369 (1.0069) model_time 0.9367 (1.0030) loss 0.7779 (0.9058) grad_norm 5.9492 (8.8014/1.8366) mem 68106MB [2022-12-19 15:15:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][160/1519] eta 0:22:47 lr 0.000029 time 0.9358 (1.0063) model_time 0.9356 (1.0027) loss 0.8925 (0.9057) grad_norm 10.3646 (8.7424/1.8213) mem 68106MB [2022-12-19 15:15:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][170/1519] eta 0:22:37 lr 0.000029 time 0.9867 (1.0062) model_time 0.9865 (1.0028) loss 1.1314 (0.9092) grad_norm 8.5985 (8.7277/1.7766) mem 68106MB [2022-12-19 15:16:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][180/1519] eta 0:22:27 lr 0.000029 time 0.9290 (1.0065) model_time 0.9288 (1.0032) loss 0.6957 (0.9094) grad_norm 11.7241 (8.7151/1.7834) mem 68106MB [2022-12-19 15:16:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][190/1519] eta 0:22:17 lr 0.000029 time 0.9364 (1.0061) model_time 0.9363 (1.0030) loss 0.8929 (0.9067) grad_norm 14.1464 (8.7869/1.8530) mem 68106MB [2022-12-19 15:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][200/1519] eta 0:22:06 lr 0.000029 time 0.9281 (1.0057) model_time 0.9279 (1.0027) loss 1.0338 (0.9059) grad_norm 11.2437 (8.7718/1.8497) mem 68106MB [2022-12-19 15:16:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][210/1519] eta 0:21:56 lr 0.000029 time 0.9265 (1.0055) model_time 0.9262 (1.0027) loss 1.1970 (0.9098) grad_norm 10.2421 (8.8170/1.8553) mem 68106MB [2022-12-19 15:16:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][220/1519] eta 0:21:46 lr 0.000029 time 0.9806 (1.0054) model_time 0.9804 (1.0027) loss 1.2075 (0.9118) grad_norm 9.0781 (8.8364/1.8254) mem 68106MB [2022-12-19 15:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][230/1519] eta 0:21:35 lr 0.000029 time 0.9265 (1.0051) model_time 0.9263 (1.0024) loss 1.1027 (0.9112) grad_norm 7.9661 (8.7932/1.8175) mem 68106MB [2022-12-19 15:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][240/1519] eta 0:21:26 lr 0.000029 time 0.9155 (1.0059) model_time 0.9154 (1.0033) loss 0.9525 (0.9125) grad_norm 16.9301 (8.8961/1.9412) mem 68106MB [2022-12-19 15:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][250/1519] eta 0:21:16 lr 0.000029 time 0.9340 (1.0057) model_time 0.9338 (1.0033) loss 1.3975 (0.9147) grad_norm 8.4594 (8.8974/1.9353) mem 68106MB [2022-12-19 15:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][260/1519] eta 0:21:06 lr 0.000029 time 0.9315 (1.0056) model_time 0.9314 (1.0032) loss 0.7974 (0.9137) grad_norm 6.4619 (8.8629/1.9370) mem 68106MB [2022-12-19 15:17:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][270/1519] eta 0:20:56 lr 0.000029 time 0.9320 (1.0057) model_time 0.9318 (1.0034) loss 0.8883 (0.9103) grad_norm 8.5935 (8.8477/1.9592) mem 68106MB [2022-12-19 15:17:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][280/1519] eta 0:20:46 lr 0.000029 time 0.9132 (1.0057) model_time 0.9131 (1.0035) loss 0.7837 (0.9080) grad_norm 8.5257 (8.8339/1.9492) mem 68106MB [2022-12-19 15:17:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][290/1519] eta 0:20:36 lr 0.000029 time 0.9369 (1.0060) model_time 0.9368 (1.0038) loss 0.8613 (0.9075) grad_norm 8.7238 (8.8190/1.9340) mem 68106MB [2022-12-19 15:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][300/1519] eta 0:20:26 lr 0.000029 time 0.9266 (1.0057) model_time 0.9265 (1.0036) loss 1.0436 (0.9074) grad_norm 9.1120 (8.7850/1.9180) mem 68106MB [2022-12-19 15:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][310/1519] eta 0:20:15 lr 0.000029 time 0.9272 (1.0056) model_time 0.9270 (1.0036) loss 0.8178 (0.9079) grad_norm 6.5422 (8.8059/1.9227) mem 68106MB [2022-12-19 15:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][320/1519] eta 0:20:05 lr 0.000029 time 0.9319 (1.0057) model_time 0.9318 (1.0036) loss 0.9853 (0.9098) grad_norm 12.8300 (8.8524/1.9599) mem 68106MB [2022-12-19 15:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][330/1519] eta 0:19:55 lr 0.000029 time 0.9403 (1.0055) model_time 0.9401 (1.0035) loss 0.8833 (0.9098) grad_norm 6.1145 (8.8365/1.9519) mem 68106MB [2022-12-19 15:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][340/1519] eta 0:19:45 lr 0.000029 time 0.9360 (1.0056) model_time 0.9359 (1.0037) loss 1.0328 (0.9085) grad_norm 9.3729 (8.8168/1.9309) mem 68106MB [2022-12-19 15:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][350/1519] eta 0:19:35 lr 0.000029 time 0.9315 (1.0055) model_time 0.9313 (1.0036) loss 0.8015 (0.9101) grad_norm 8.8750 (8.8301/1.9289) mem 68106MB [2022-12-19 15:19:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][360/1519] eta 0:19:25 lr 0.000029 time 0.9337 (1.0057) model_time 0.9335 (1.0038) loss 0.9308 (0.9114) grad_norm 9.1125 (8.8596/2.0067) mem 68106MB [2022-12-19 15:19:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][370/1519] eta 0:19:15 lr 0.000029 time 0.9203 (1.0057) model_time 0.9201 (1.0039) loss 1.0331 (0.9118) grad_norm 8.1416 (8.8849/2.0101) mem 68106MB [2022-12-19 15:19:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][380/1519] eta 0:19:05 lr 0.000029 time 0.9354 (1.0055) model_time 0.9350 (1.0038) loss 1.2388 (0.9095) grad_norm 6.8202 (8.8705/1.9948) mem 68106MB [2022-12-19 15:19:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][390/1519] eta 0:18:55 lr 0.000029 time 0.9218 (1.0054) model_time 0.9216 (1.0037) loss 0.7147 (0.9103) grad_norm 17.5999 (8.9094/2.0756) mem 68106MB [2022-12-19 15:19:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][400/1519] eta 0:18:44 lr 0.000029 time 0.9814 (1.0054) model_time 0.9812 (1.0037) loss 1.0721 (0.9098) grad_norm 8.2473 (8.8683/2.0674) mem 68106MB [2022-12-19 15:19:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][410/1519] eta 0:18:34 lr 0.000029 time 0.9325 (1.0052) model_time 0.9322 (1.0035) loss 0.8812 (0.9107) grad_norm 8.3331 (8.8570/2.0573) mem 68106MB [2022-12-19 15:20:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][420/1519] eta 0:18:24 lr 0.000029 time 0.9228 (1.0052) model_time 0.9227 (1.0036) loss 0.7157 (0.9093) grad_norm 5.8472 (8.8162/2.0568) mem 68106MB [2022-12-19 15:20:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][430/1519] eta 0:18:14 lr 0.000029 time 0.9278 (1.0050) model_time 0.9276 (1.0034) loss 0.9653 (0.9107) grad_norm 10.7261 (8.8381/2.0456) mem 68106MB [2022-12-19 15:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][440/1519] eta 0:18:04 lr 0.000029 time 0.9289 (1.0050) model_time 0.9288 (1.0034) loss 1.1350 (0.9103) grad_norm 6.8502 (8.8126/2.0387) mem 68106MB [2022-12-19 15:20:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][450/1519] eta 0:17:54 lr 0.000029 time 0.9206 (1.0051) model_time 0.9204 (1.0036) loss 0.9583 (0.9103) grad_norm 9.0846 (8.8643/2.1010) mem 68106MB [2022-12-19 15:20:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][460/1519] eta 0:17:44 lr 0.000029 time 0.9294 (1.0051) model_time 0.9293 (1.0036) loss 0.9151 (0.9109) grad_norm 9.2380 (8.8325/2.0944) mem 68106MB [2022-12-19 15:20:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][470/1519] eta 0:17:34 lr 0.000029 time 0.9222 (1.0052) model_time 0.9221 (1.0037) loss 0.9116 (0.9115) grad_norm 8.1947 (8.8312/2.0775) mem 68106MB [2022-12-19 15:21:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][480/1519] eta 0:17:24 lr 0.000029 time 0.9256 (1.0051) model_time 0.9254 (1.0036) loss 0.8036 (0.9106) grad_norm 7.2913 (8.8218/2.0656) mem 68106MB [2022-12-19 15:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][490/1519] eta 0:17:14 lr 0.000029 time 0.9362 (1.0053) model_time 0.9360 (1.0039) loss 0.7532 (0.9105) grad_norm 6.1348 (8.8255/2.0968) mem 68106MB [2022-12-19 15:21:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][500/1519] eta 0:17:04 lr 0.000029 time 0.9352 (1.0052) model_time 0.9351 (1.0038) loss 0.7778 (0.9109) grad_norm 7.9869 (8.8229/2.0883) mem 68106MB [2022-12-19 15:21:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][510/1519] eta 0:16:54 lr 0.000029 time 0.9277 (1.0051) model_time 0.9276 (1.0037) loss 0.7957 (0.9111) grad_norm 9.6665 (8.8286/2.0729) mem 68106MB [2022-12-19 15:21:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][520/1519] eta 0:16:43 lr 0.000029 time 0.9276 (1.0049) model_time 0.9275 (1.0036) loss 1.3639 (0.9122) grad_norm 6.9324 (8.8257/2.0584) mem 68106MB [2022-12-19 15:21:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][530/1519] eta 0:16:33 lr 0.000029 time 0.9307 (1.0048) model_time 0.9306 (1.0034) loss 0.8123 (0.9142) grad_norm 6.9100 (8.8195/2.0597) mem 68106MB [2022-12-19 15:22:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][540/1519] eta 0:16:23 lr 0.000029 time 0.9268 (1.0050) model_time 0.9267 (1.0036) loss 0.7465 (0.9138) grad_norm 11.5465 (8.8273/2.0539) mem 68106MB [2022-12-19 15:22:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][550/1519] eta 0:16:13 lr 0.000029 time 0.9327 (1.0048) model_time 0.9325 (1.0035) loss 1.1810 (0.9137) grad_norm 8.6465 (8.8327/2.0470) mem 68106MB [2022-12-19 15:22:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][560/1519] eta 0:16:03 lr 0.000029 time 1.0057 (1.0052) model_time 1.0056 (1.0038) loss 1.2401 (0.9142) grad_norm 10.3518 (8.8404/2.0425) mem 68106MB [2022-12-19 15:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][570/1519] eta 0:15:53 lr 0.000029 time 0.9270 (1.0051) model_time 0.9269 (1.0038) loss 0.7863 (0.9149) grad_norm 6.4422 (8.8286/2.0550) mem 68106MB [2022-12-19 15:22:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][580/1519] eta 0:15:43 lr 0.000029 time 0.9868 (1.0052) model_time 0.9865 (1.0039) loss 0.9745 (0.9154) grad_norm 6.8148 (8.8003/2.0502) mem 68106MB [2022-12-19 15:22:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][590/1519] eta 0:15:33 lr 0.000029 time 0.9300 (1.0051) model_time 0.9299 (1.0038) loss 1.0565 (0.9151) grad_norm 8.6027 (8.8007/2.0452) mem 68106MB [2022-12-19 15:23:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][600/1519] eta 0:15:23 lr 0.000029 time 0.9314 (1.0052) model_time 0.9311 (1.0039) loss 0.9474 (0.9164) grad_norm 8.1257 (8.7840/2.0412) mem 68106MB [2022-12-19 15:23:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][610/1519] eta 0:15:13 lr 0.000029 time 0.9321 (1.0052) model_time 0.9319 (1.0039) loss 1.2114 (0.9174) grad_norm 6.9709 (8.7807/2.0478) mem 68106MB [2022-12-19 15:23:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][620/1519] eta 0:15:03 lr 0.000029 time 0.9203 (1.0052) model_time 0.9201 (1.0040) loss 0.8347 (0.9171) grad_norm 8.0693 (8.7985/2.0574) mem 68106MB [2022-12-19 15:23:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][630/1519] eta 0:14:53 lr 0.000029 time 0.9325 (1.0053) model_time 0.9323 (1.0041) loss 1.0683 (0.9169) grad_norm 8.1780 (8.8009/2.0635) mem 68106MB [2022-12-19 15:23:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][640/1519] eta 0:14:43 lr 0.000029 time 0.9267 (1.0052) model_time 0.9266 (1.0040) loss 0.6952 (0.9161) grad_norm 9.3637 (8.8031/2.0634) mem 68106MB [2022-12-19 15:23:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][650/1519] eta 0:14:33 lr 0.000029 time 0.9342 (1.0053) model_time 0.9341 (1.0041) loss 1.4149 (0.9157) grad_norm 10.0628 (8.8282/2.0780) mem 68106MB [2022-12-19 15:24:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][660/1519] eta 0:14:23 lr 0.000029 time 0.9215 (1.0053) model_time 0.9214 (1.0041) loss 0.7411 (0.9157) grad_norm 7.7538 (8.8181/2.0706) mem 68106MB [2022-12-19 15:24:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][670/1519] eta 0:14:13 lr 0.000029 time 0.9124 (1.0053) model_time 0.9123 (1.0041) loss 1.0250 (0.9171) grad_norm 6.8824 (8.8048/2.0784) mem 68106MB [2022-12-19 15:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][680/1519] eta 0:14:03 lr 0.000029 time 0.9352 (1.0053) model_time 0.9350 (1.0041) loss 0.7229 (0.9174) grad_norm 11.1270 (8.8034/2.0777) mem 68106MB [2022-12-19 15:24:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][690/1519] eta 0:13:53 lr 0.000029 time 0.9195 (1.0052) model_time 0.9194 (1.0040) loss 1.4615 (0.9179) grad_norm 7.7256 (8.8040/2.0819) mem 68106MB [2022-12-19 15:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][700/1519] eta 0:13:43 lr 0.000029 time 0.9330 (1.0050) model_time 0.9328 (1.0039) loss 0.7584 (0.9173) grad_norm 10.8159 (8.8027/2.0789) mem 68106MB [2022-12-19 15:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][710/1519] eta 0:13:32 lr 0.000029 time 0.9236 (1.0049) model_time 0.9233 (1.0038) loss 0.7308 (0.9185) grad_norm 8.8435 (8.7961/2.0842) mem 68106MB [2022-12-19 15:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][720/1519] eta 0:13:22 lr 0.000029 time 0.9315 (1.0049) model_time 0.9313 (1.0038) loss 0.9467 (0.9170) grad_norm 10.7105 (8.7380/2.0066) mem 68106MB [2022-12-19 15:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][730/1519] eta 0:13:12 lr 0.000029 time 0.9104 (1.0048) model_time 0.9103 (1.0037) loss 0.8739 (0.9170) grad_norm 8.5488 (8.7246/1.9957) mem 68106MB [2022-12-19 15:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][740/1519] eta 0:13:02 lr 0.000029 time 1.1956 (1.0051) model_time 1.1955 (1.0040) loss 0.8274 (0.9171) grad_norm 10.6304 (8.7022/1.9876) mem 68106MB [2022-12-19 15:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][750/1519] eta 0:12:52 lr 0.000029 time 0.9333 (1.0050) model_time 0.9331 (1.0039) loss 0.8227 (0.9165) grad_norm 6.5579 (8.6891/1.9902) mem 68106MB [2022-12-19 15:25:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][760/1519] eta 0:12:42 lr 0.000029 time 0.9876 (1.0050) model_time 0.9874 (1.0040) loss 1.0729 (0.9166) grad_norm 8.0833 (8.7008/1.9812) mem 68106MB [2022-12-19 15:25:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][770/1519] eta 0:12:32 lr 0.000029 time 0.9454 (1.0052) model_time 0.9452 (1.0041) loss 1.6677 (0.9178) grad_norm 7.4689 (8.6845/2.0016) mem 68106MB [2022-12-19 15:26:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][780/1519] eta 0:12:22 lr 0.000029 time 0.9448 (1.0052) model_time 0.9446 (1.0042) loss 1.0729 (0.9185) grad_norm 8.0813 (8.7071/1.9970) mem 68106MB [2022-12-19 15:26:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][790/1519] eta 0:12:12 lr 0.000029 time 0.9355 (1.0052) model_time 0.9353 (1.0042) loss 1.0832 (0.9188) grad_norm 9.3440 (8.7276/2.0262) mem 68106MB [2022-12-19 15:26:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][800/1519] eta 0:12:02 lr 0.000029 time 0.9427 (1.0052) model_time 0.9425 (1.0041) loss 0.8023 (0.9192) grad_norm 7.5070 (8.7113/2.0229) mem 68106MB [2022-12-19 15:26:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][810/1519] eta 0:11:52 lr 0.000029 time 0.9317 (1.0053) model_time 0.9315 (1.0042) loss 0.7375 (0.9183) grad_norm 7.4017 (8.6971/2.0095) mem 68106MB [2022-12-19 15:26:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][820/1519] eta 0:11:42 lr 0.000029 time 0.9274 (1.0052) model_time 0.9272 (1.0041) loss 0.9049 (0.9186) grad_norm 13.3062 (8.6997/2.0259) mem 68106MB [2022-12-19 15:26:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][830/1519] eta 0:11:32 lr 0.000029 time 0.9321 (1.0051) model_time 0.9318 (1.0041) loss 1.2522 (0.9185) grad_norm 6.5980 (8.7019/2.0261) mem 68106MB [2022-12-19 15:27:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][840/1519] eta 0:11:22 lr 0.000029 time 0.9317 (1.0050) model_time 0.9315 (1.0040) loss 0.9149 (0.9191) grad_norm 8.0004 (8.6736/1.9843) mem 68106MB [2022-12-19 15:27:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][850/1519] eta 0:11:12 lr 0.000029 time 0.9302 (1.0053) model_time 0.9301 (1.0043) loss 0.7588 (0.9189) grad_norm 7.4207 (8.6582/1.9744) mem 68106MB [2022-12-19 15:27:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][860/1519] eta 0:11:02 lr 0.000029 time 0.9300 (1.0052) model_time 0.9299 (1.0042) loss 0.8230 (0.9195) grad_norm 6.6266 (8.6636/1.9839) mem 68106MB [2022-12-19 15:27:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][870/1519] eta 0:10:52 lr 0.000029 time 0.9224 (1.0052) model_time 0.9223 (1.0042) loss 1.1515 (0.9201) grad_norm 6.7297 (8.6570/1.9684) mem 68106MB [2022-12-19 15:27:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][880/1519] eta 0:10:42 lr 0.000029 time 0.9333 (1.0052) model_time 0.9332 (1.0042) loss 1.0104 (0.9206) grad_norm 9.8235 (8.6650/1.9614) mem 68106MB [2022-12-19 15:27:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][890/1519] eta 0:10:32 lr 0.000029 time 0.9277 (1.0051) model_time 0.9275 (1.0041) loss 0.8234 (0.9204) grad_norm 6.5170 (8.6710/1.9680) mem 68106MB [2022-12-19 15:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][900/1519] eta 0:10:22 lr 0.000029 time 0.9261 (1.0051) model_time 0.9259 (1.0041) loss 0.7775 (0.9202) grad_norm 5.8587 (8.6605/1.9737) mem 68106MB [2022-12-19 15:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][910/1519] eta 0:10:12 lr 0.000029 time 1.1950 (1.0054) model_time 1.1949 (1.0044) loss 1.0669 (0.9205) grad_norm 14.0778 (8.6678/2.0012) mem 68106MB [2022-12-19 15:28:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][920/1519] eta 0:10:02 lr 0.000029 time 0.9340 (1.0054) model_time 0.9339 (1.0044) loss 0.8208 (0.9198) grad_norm 10.8392 (8.6626/1.9725) mem 68106MB [2022-12-19 15:28:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][930/1519] eta 0:09:52 lr 0.000029 time 0.9921 (1.0054) model_time 0.9919 (1.0044) loss 1.1798 (0.9192) grad_norm 7.2440 (8.6842/1.9856) mem 68106MB [2022-12-19 15:28:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][940/1519] eta 0:09:42 lr 0.000029 time 0.9332 (1.0053) model_time 0.9330 (1.0043) loss 1.0365 (0.9197) grad_norm 7.4421 (8.6743/1.9927) mem 68106MB [2022-12-19 15:29:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][950/1519] eta 0:09:32 lr 0.000029 time 0.9337 (1.0054) model_time 0.9335 (1.0044) loss 0.7399 (0.9196) grad_norm 13.5132 (8.7017/2.0282) mem 68106MB [2022-12-19 15:29:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][960/1519] eta 0:09:22 lr 0.000029 time 0.9321 (1.0054) model_time 0.9320 (1.0044) loss 1.0393 (0.9198) grad_norm 6.3029 (8.6849/1.9978) mem 68106MB [2022-12-19 15:29:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][970/1519] eta 0:09:11 lr 0.000029 time 0.9308 (1.0053) model_time 0.9307 (1.0044) loss 0.7173 (0.9195) grad_norm 6.9131 (8.6429/1.9863) mem 68106MB [2022-12-19 15:29:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][980/1519] eta 0:09:01 lr 0.000029 time 0.9311 (1.0052) model_time 0.9310 (1.0043) loss 0.9878 (0.9190) grad_norm 10.2888 (8.6388/1.9876) mem 68106MB [2022-12-19 15:29:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][990/1519] eta 0:08:51 lr 0.000029 time 0.9787 (1.0053) model_time 0.9785 (1.0043) loss 1.2457 (0.9200) grad_norm 9.4768 (8.6041/1.9297) mem 68106MB [2022-12-19 15:29:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1000/1519] eta 0:08:41 lr 0.000029 time 0.9291 (1.0052) model_time 0.9290 (1.0043) loss 1.1603 (0.9209) grad_norm 7.2378 (8.6263/1.9347) mem 68106MB [2022-12-19 15:30:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1010/1519] eta 0:08:31 lr 0.000029 time 0.9322 (1.0051) model_time 0.9320 (1.0042) loss 0.7112 (0.9213) grad_norm 6.6287 (8.6079/1.9372) mem 68106MB [2022-12-19 15:30:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1020/1519] eta 0:08:21 lr 0.000029 time 0.9282 (1.0051) model_time 0.9280 (1.0042) loss 0.8300 (0.9212) grad_norm 9.3878 (8.6299/1.9242) mem 68106MB [2022-12-19 15:30:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1030/1519] eta 0:08:11 lr 0.000029 time 0.9322 (1.0050) model_time 0.9319 (1.0041) loss 0.7992 (0.9212) grad_norm 5.9237 (8.6078/1.9505) mem 68106MB [2022-12-19 15:30:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1040/1519] eta 0:08:01 lr 0.000029 time 0.9314 (1.0049) model_time 0.9313 (1.0041) loss 1.3496 (0.9211) grad_norm 7.8995 (8.6109/1.9570) mem 68106MB [2022-12-19 15:30:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1050/1519] eta 0:07:51 lr 0.000029 time 0.9360 (1.0051) model_time 0.9358 (1.0042) loss 1.0374 (0.9212) grad_norm 8.1117 (8.5694/1.8944) mem 68106MB [2022-12-19 15:30:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1060/1519] eta 0:07:41 lr 0.000029 time 0.9333 (1.0051) model_time 0.9332 (1.0042) loss 1.0025 (0.9207) grad_norm 9.6474 (8.6146/1.9026) mem 68106MB [2022-12-19 15:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1070/1519] eta 0:07:31 lr 0.000029 time 0.9230 (1.0050) model_time 0.9229 (1.0041) loss 0.9516 (0.9202) grad_norm 7.4309 (8.6079/1.9009) mem 68106MB [2022-12-19 15:31:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1080/1519] eta 0:07:21 lr 0.000029 time 0.9329 (1.0051) model_time 0.9328 (1.0042) loss 0.9800 (0.9205) grad_norm 5.7486 (8.5865/1.9079) mem 68106MB [2022-12-19 15:31:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1090/1519] eta 0:07:11 lr 0.000029 time 1.0539 (1.0052) model_time 1.0537 (1.0043) loss 0.9463 (0.9210) grad_norm 8.0115 (8.5840/1.8622) mem 68106MB [2022-12-19 15:31:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1100/1519] eta 0:07:01 lr 0.000029 time 0.9295 (1.0052) model_time 0.9294 (1.0044) loss 1.0237 (0.9214) grad_norm 10.1769 (8.5665/1.8624) mem 68106MB [2022-12-19 15:31:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1110/1519] eta 0:06:51 lr 0.000029 time 0.9347 (1.0052) model_time 0.9345 (1.0043) loss 0.8670 (0.9209) grad_norm 8.8102 (8.5469/1.8615) mem 68106MB [2022-12-19 15:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1120/1519] eta 0:06:41 lr 0.000029 time 0.9285 (1.0052) model_time 0.9283 (1.0044) loss 0.7569 (0.9207) grad_norm 7.7420 (8.5297/1.8613) mem 68106MB [2022-12-19 15:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1130/1519] eta 0:06:31 lr 0.000029 time 0.9456 (1.0052) model_time 0.9454 (1.0043) loss 0.7454 (0.9203) grad_norm 9.2423 (8.5270/1.8533) mem 68106MB [2022-12-19 15:32:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1140/1519] eta 0:06:20 lr 0.000029 time 0.9300 (1.0052) model_time 0.9299 (1.0044) loss 0.8610 (0.9205) grad_norm 8.9352 (8.5123/1.8399) mem 68106MB [2022-12-19 15:32:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1150/1519] eta 0:06:10 lr 0.000029 time 0.9406 (1.0052) model_time 0.9405 (1.0043) loss 0.7289 (0.9215) grad_norm 12.7280 (8.5260/1.8484) mem 68106MB [2022-12-19 15:32:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1160/1519] eta 0:06:00 lr 0.000029 time 0.8828 (1.0052) model_time 0.8827 (1.0044) loss 0.7717 (0.9210) grad_norm 7.2945 (8.5078/1.8420) mem 68106MB [2022-12-19 15:32:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1170/1519] eta 0:05:50 lr 0.000029 time 0.9875 (1.0052) model_time 0.9873 (1.0043) loss 0.8020 (0.9209) grad_norm 6.6955 (8.5317/1.8349) mem 68106MB [2022-12-19 15:32:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1180/1519] eta 0:05:40 lr 0.000029 time 0.9435 (1.0051) model_time 0.9433 (1.0043) loss 0.7268 (0.9206) grad_norm 7.5118 (8.5210/1.8453) mem 68106MB [2022-12-19 15:33:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1190/1519] eta 0:05:30 lr 0.000029 time 0.9352 (1.0051) model_time 0.9350 (1.0043) loss 0.9253 (0.9215) grad_norm 7.9230 (8.5181/1.8432) mem 68106MB [2022-12-19 15:33:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1200/1519] eta 0:05:20 lr 0.000029 time 0.9339 (1.0051) model_time 0.9337 (1.0042) loss 0.8197 (0.9223) grad_norm 8.1306 (8.5411/1.8498) mem 68106MB [2022-12-19 15:33:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1210/1519] eta 0:05:10 lr 0.000029 time 0.9355 (1.0050) model_time 0.9353 (1.0042) loss 0.8703 (0.9223) grad_norm 6.2448 (8.5726/1.8852) mem 68106MB [2022-12-19 15:33:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1220/1519] eta 0:05:00 lr 0.000029 time 0.9325 (1.0050) model_time 0.9324 (1.0042) loss 0.8580 (0.9223) grad_norm 12.5416 (8.5873/1.8971) mem 68106MB [2022-12-19 15:33:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1230/1519] eta 0:04:50 lr 0.000029 time 0.9273 (1.0049) model_time 0.9272 (1.0041) loss 0.9362 (0.9227) grad_norm 10.6665 (8.6163/1.9078) mem 68106MB [2022-12-19 15:33:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1240/1519] eta 0:04:40 lr 0.000029 time 0.9243 (1.0049) model_time 0.9242 (1.0041) loss 0.7598 (0.9226) grad_norm 10.3251 (8.6048/1.9130) mem 68106MB [2022-12-19 15:34:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1250/1519] eta 0:04:30 lr 0.000029 time 0.9320 (1.0049) model_time 0.9318 (1.0041) loss 1.3193 (0.9226) grad_norm 7.6561 (8.5767/1.9002) mem 68106MB [2022-12-19 15:34:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1260/1519] eta 0:04:20 lr 0.000029 time 0.9179 (1.0051) model_time 0.9178 (1.0043) loss 0.7312 (0.9220) grad_norm 8.4891 (8.5654/1.9062) mem 68106MB [2022-12-19 15:34:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1270/1519] eta 0:04:10 lr 0.000029 time 0.9311 (1.0050) model_time 0.9310 (1.0042) loss 0.7601 (0.9223) grad_norm 5.8706 (8.5812/1.9077) mem 68106MB [2022-12-19 15:34:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1280/1519] eta 0:04:00 lr 0.000029 time 0.9341 (1.0050) model_time 0.9340 (1.0042) loss 0.9310 (0.9219) grad_norm 9.0814 (8.6139/1.9086) mem 68106MB [2022-12-19 15:34:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1290/1519] eta 0:03:50 lr 0.000029 time 0.9319 (1.0049) model_time 0.9318 (1.0042) loss 0.7968 (0.9219) grad_norm 8.0583 (8.6324/1.9535) mem 68106MB [2022-12-19 15:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1300/1519] eta 0:03:40 lr 0.000029 time 0.9601 (1.0050) model_time 0.9599 (1.0042) loss 1.3284 (0.9220) grad_norm 8.0212 (8.6599/2.0438) mem 68106MB [2022-12-19 15:35:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1310/1519] eta 0:03:30 lr 0.000029 time 0.9301 (1.0049) model_time 0.9300 (1.0042) loss 0.8090 (0.9223) grad_norm 9.5990 (8.6800/2.0368) mem 68106MB [2022-12-19 15:35:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1320/1519] eta 0:03:19 lr 0.000029 time 0.9341 (1.0049) model_time 0.9340 (1.0041) loss 1.0303 (0.9225) grad_norm 9.1714 (8.6703/2.0334) mem 68106MB [2022-12-19 15:35:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1330/1519] eta 0:03:09 lr 0.000029 time 0.9328 (1.0048) model_time 0.9327 (1.0041) loss 0.7478 (0.9222) grad_norm 8.6052 (8.6838/2.0463) mem 68106MB [2022-12-19 15:35:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1340/1519] eta 0:02:59 lr 0.000029 time 0.9330 (1.0048) model_time 0.9329 (1.0041) loss 1.4070 (0.9225) grad_norm 10.8279 (8.6891/2.0634) mem 68106MB [2022-12-19 15:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1350/1519] eta 0:02:49 lr 0.000029 time 0.9278 (1.0048) model_time 0.9276 (1.0040) loss 0.7886 (0.9223) grad_norm 6.9465 (8.7269/2.0764) mem 68106MB [2022-12-19 15:35:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1360/1519] eta 0:02:39 lr 0.000029 time 0.9319 (1.0050) model_time 0.9318 (1.0043) loss 0.6778 (0.9223) grad_norm 11.4571 (8.7462/2.0951) mem 68106MB [2022-12-19 15:36:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1370/1519] eta 0:02:29 lr 0.000029 time 0.9350 (1.0051) model_time 0.9348 (1.0043) loss 0.7993 (0.9218) grad_norm 7.3984 (8.7732/2.0901) mem 68106MB [2022-12-19 15:36:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1380/1519] eta 0:02:19 lr 0.000029 time 0.9319 (1.0050) model_time 0.9318 (1.0042) loss 0.7238 (0.9217) grad_norm 14.1717 (8.7678/2.1112) mem 68106MB [2022-12-19 15:36:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1390/1519] eta 0:02:09 lr 0.000029 time 0.9447 (1.0050) model_time 0.9446 (1.0043) loss 1.0892 (0.9216) grad_norm 6.7227 (8.6995/2.0638) mem 68106MB [2022-12-19 15:36:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1400/1519] eta 0:01:59 lr 0.000029 time 0.9398 (1.0050) model_time 0.9397 (1.0043) loss 1.0633 (0.9217) grad_norm 7.0857 (8.7036/2.0615) mem 68106MB [2022-12-19 15:36:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1410/1519] eta 0:01:49 lr 0.000029 time 0.9440 (1.0054) model_time 0.9438 (1.0046) loss 0.6894 (0.9209) grad_norm 9.2193 (8.6982/2.0632) mem 68106MB [2022-12-19 15:36:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1420/1519] eta 0:01:39 lr 0.000029 time 0.9366 (1.0054) model_time 0.9365 (1.0046) loss 1.0556 (0.9207) grad_norm 11.6447 (8.7198/2.0738) mem 68106MB [2022-12-19 15:37:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1430/1519] eta 0:01:29 lr 0.000029 time 0.9268 (1.0054) model_time 0.9267 (1.0046) loss 0.7396 (0.9207) grad_norm 8.0643 (8.7349/2.0669) mem 68106MB [2022-12-19 15:37:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1440/1519] eta 0:01:19 lr 0.000029 time 0.9309 (1.0054) model_time 0.9307 (1.0046) loss 0.8230 (0.9206) grad_norm 7.7869 (8.7218/2.0485) mem 68106MB [2022-12-19 15:37:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1450/1519] eta 0:01:09 lr 0.000028 time 0.9338 (1.0053) model_time 0.9337 (1.0046) loss 1.0018 (0.9209) grad_norm 7.3619 (8.7280/2.0517) mem 68106MB [2022-12-19 15:37:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1460/1519] eta 0:00:59 lr 0.000028 time 0.9362 (1.0053) model_time 0.9361 (1.0046) loss 0.8013 (0.9206) grad_norm 10.6940 (8.7548/2.0422) mem 68106MB [2022-12-19 15:37:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1470/1519] eta 0:00:49 lr 0.000028 time 0.9272 (1.0053) model_time 0.9271 (1.0046) loss 1.2557 (0.9210) grad_norm 9.2795 (8.7765/2.0468) mem 68106MB [2022-12-19 15:37:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1480/1519] eta 0:00:39 lr 0.000028 time 0.9256 (1.0053) model_time 0.9254 (1.0046) loss 1.3679 (0.9211) grad_norm 8.4758 (8.7710/2.0450) mem 68106MB [2022-12-19 15:38:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1490/1519] eta 0:00:29 lr 0.000028 time 0.9272 (1.0054) model_time 0.9271 (1.0046) loss 0.9044 (0.9211) grad_norm 9.7123 (8.7895/2.0477) mem 68106MB [2022-12-19 15:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1500/1519] eta 0:00:19 lr 0.000028 time 0.9309 (1.0053) model_time 0.9308 (1.0046) loss 0.9907 (0.9207) grad_norm 8.4849 (8.8016/2.0434) mem 68106MB [2022-12-19 15:38:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [25/100][1510/1519] eta 0:00:09 lr 0.000028 time 0.9275 (1.0053) model_time 0.9274 (1.0046) loss 0.7964 (0.9210) grad_norm 9.3980 (8.8319/2.0800) mem 68106MB [2022-12-19 15:38:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 25 training takes 0:25:27 [2022-12-19 15:38:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_25.pth saving...... [2022-12-19 15:38:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_25.pth saved !!! [2022-12-19 15:38:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.673 (0.673) Loss 0.5258 (0.5258) Acc@1 90.625 (90.625) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 15:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.331) Loss 0.5143 (0.5015) Acc@1 90.972 (91.540) Acc@5 97.917 (98.359) Mem 68106MB [2022-12-19 15:39:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4673 (0.5057) Acc@1 92.014 (91.038) Acc@5 98.264 (98.165) Mem 68106MB [2022-12-19 15:39:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6128 (0.5125) Acc@1 89.236 (90.916) Acc@5 97.917 (98.118) Mem 68106MB [2022-12-19 15:39:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.311 (0.307) Loss 0.4864 (0.5046) Acc@1 90.278 (90.938) Acc@5 98.264 (98.196) Mem 68106MB [2022-12-19 15:39:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.4905 (0.5013) Acc@1 89.236 (90.884) Acc@5 99.653 (98.291) Mem 68106MB [2022-12-19 15:39:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.301 (0.305) Loss 0.4969 (0.5014) Acc@1 89.931 (90.915) Acc@5 97.569 (98.304) Mem 68106MB [2022-12-19 15:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5748 (0.5031) Acc@1 89.583 (90.865) Acc@5 98.264 (98.288) Mem 68106MB [2022-12-19 15:39:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4382 (0.5020) Acc@1 92.014 (90.912) Acc@5 98.611 (98.345) Mem 68106MB [2022-12-19 15:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:25] * Acc@1 90.954 Acc@5 98.355 [2022-12-19 15:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.0% [2022-12-19 15:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 15:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 15:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 90.95% [2022-12-19 15:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][0/1519] eta 0:36:45 lr 0.000028 time 1.4520 (1.4520) model_time 0.9922 (0.9922) loss 1.2195 (1.2195) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 15:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][10/1519] eta 0:26:19 lr 0.000028 time 0.9411 (1.0468) model_time 0.9410 (1.0046) loss 0.7284 (0.8746) grad_norm 8.5675 (9.1073/1.2175) mem 68106MB [2022-12-19 15:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][20/1519] eta 0:25:36 lr 0.000028 time 0.9241 (1.0248) model_time 0.9240 (1.0026) loss 0.9831 (0.8567) grad_norm 9.2189 (8.8361/1.2551) mem 68106MB [2022-12-19 15:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][30/1519] eta 0:25:16 lr 0.000028 time 0.9258 (1.0185) model_time 0.9257 (1.0033) loss 0.7536 (0.8764) grad_norm 14.2560 (9.2387/1.9992) mem 68106MB [2022-12-19 15:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][40/1519] eta 0:24:58 lr 0.000028 time 0.9310 (1.0135) model_time 0.9308 (1.0019) loss 0.8208 (0.8730) grad_norm 7.8200 (9.0940/1.9756) mem 68106MB [2022-12-19 15:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][50/1519] eta 0:24:46 lr 0.000028 time 0.9358 (1.0117) model_time 0.9356 (1.0024) loss 0.8823 (0.8710) grad_norm 9.5313 (9.0868/1.9092) mem 68106MB [2022-12-19 15:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][60/1519] eta 0:24:33 lr 0.000028 time 0.9347 (1.0097) model_time 0.9345 (1.0019) loss 0.7301 (0.8681) grad_norm 9.9493 (9.0828/1.8046) mem 68106MB [2022-12-19 15:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][70/1519] eta 0:24:20 lr 0.000028 time 0.9255 (1.0080) model_time 0.9254 (1.0012) loss 1.0779 (0.8727) grad_norm 8.2838 (9.3258/1.9632) mem 68106MB [2022-12-19 15:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][80/1519] eta 0:24:10 lr 0.000028 time 0.9295 (1.0077) model_time 0.9293 (1.0017) loss 1.1369 (0.8743) grad_norm 5.5148 (9.0953/1.9958) mem 68106MB [2022-12-19 15:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][90/1519] eta 0:24:02 lr 0.000028 time 1.0178 (1.0094) model_time 1.0176 (1.0040) loss 0.7920 (0.8850) grad_norm 6.2293 (9.0352/1.9633) mem 68106MB [2022-12-19 15:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][100/1519] eta 0:23:50 lr 0.000028 time 0.9262 (1.0083) model_time 0.9261 (1.0034) loss 0.7334 (0.8798) grad_norm 8.6539 (8.9434/1.8908) mem 68106MB [2022-12-19 15:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][110/1519] eta 0:23:40 lr 0.000028 time 0.9427 (1.0082) model_time 0.9424 (1.0037) loss 1.0755 (0.8827) grad_norm 7.2538 (8.9310/1.8211) mem 68106MB [2022-12-19 15:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][120/1519] eta 0:23:30 lr 0.000028 time 0.9351 (1.0085) model_time 0.9350 (1.0044) loss 0.8290 (0.8864) grad_norm 8.6708 (8.8579/1.7641) mem 68106MB [2022-12-19 15:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][130/1519] eta 0:23:19 lr 0.000028 time 0.9381 (1.0079) model_time 0.9379 (1.0040) loss 1.3449 (0.8912) grad_norm 7.2162 (8.8482/1.7325) mem 68106MB [2022-12-19 15:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][140/1519] eta 0:23:10 lr 0.000028 time 0.9308 (1.0080) model_time 0.9305 (1.0044) loss 0.9371 (0.8904) grad_norm 9.0262 (8.9892/1.8821) mem 68106MB [2022-12-19 15:42:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][150/1519] eta 0:22:59 lr 0.000028 time 0.9815 (1.0079) model_time 0.9814 (1.0045) loss 0.7637 (0.8934) grad_norm 7.7995 (8.9349/1.8429) mem 68106MB [2022-12-19 15:42:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][160/1519] eta 0:22:49 lr 0.000028 time 0.9146 (1.0079) model_time 0.9145 (1.0047) loss 1.0500 (0.8946) grad_norm 10.4489 (8.9679/1.8479) mem 68106MB [2022-12-19 15:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][170/1519] eta 0:22:39 lr 0.000028 time 0.9310 (1.0079) model_time 0.9309 (1.0048) loss 0.9028 (0.8919) grad_norm 7.4507 (8.9052/1.8317) mem 68106MB [2022-12-19 15:42:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][180/1519] eta 0:22:29 lr 0.000028 time 0.9117 (1.0079) model_time 0.9115 (1.0050) loss 0.7534 (0.8902) grad_norm 6.8953 (8.8200/1.8236) mem 68106MB [2022-12-19 15:42:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][190/1519] eta 0:22:19 lr 0.000028 time 0.9482 (1.0076) model_time 0.9481 (1.0048) loss 0.7636 (0.8939) grad_norm 7.8580 (8.7801/1.8025) mem 68106MB [2022-12-19 15:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][200/1519] eta 0:22:08 lr 0.000028 time 0.9320 (1.0075) model_time 0.9319 (1.0049) loss 1.0816 (0.8945) grad_norm 7.1195 (8.7722/1.8461) mem 68106MB [2022-12-19 15:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][210/1519] eta 0:21:58 lr 0.000028 time 0.9342 (1.0073) model_time 0.9340 (1.0047) loss 0.8737 (0.8953) grad_norm 9.1376 (8.7836/2.0001) mem 68106MB [2022-12-19 15:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][220/1519] eta 0:21:48 lr 0.000028 time 0.9309 (1.0071) model_time 0.9307 (1.0047) loss 1.0112 (0.8981) grad_norm 7.8926 (8.7185/1.9801) mem 68106MB [2022-12-19 15:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][230/1519] eta 0:21:37 lr 0.000028 time 0.9227 (1.0070) model_time 0.9224 (1.0046) loss 1.0161 (0.8996) grad_norm 10.4604 (8.7666/1.9965) mem 68106MB [2022-12-19 15:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][240/1519] eta 0:21:27 lr 0.000028 time 0.9289 (1.0068) model_time 0.9288 (1.0045) loss 1.1589 (0.9013) grad_norm 8.5348 (8.7788/2.0571) mem 68106MB [2022-12-19 15:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][250/1519] eta 0:21:17 lr 0.000028 time 0.9308 (1.0069) model_time 0.9306 (1.0047) loss 0.9684 (0.9030) grad_norm 10.5078 (8.7833/2.0328) mem 68106MB [2022-12-19 15:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][260/1519] eta 0:21:07 lr 0.000028 time 0.9314 (1.0067) model_time 0.9312 (1.0046) loss 1.0174 (0.9008) grad_norm 6.2825 (8.7298/2.0217) mem 68106MB [2022-12-19 15:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][270/1519] eta 0:20:57 lr 0.000028 time 0.9252 (1.0070) model_time 0.9251 (1.0049) loss 0.9825 (0.9021) grad_norm 6.8490 (8.6913/1.9990) mem 68106MB [2022-12-19 15:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][280/1519] eta 0:20:47 lr 0.000028 time 0.9530 (1.0067) model_time 0.9529 (1.0047) loss 0.9802 (0.9063) grad_norm 7.1539 (8.7067/1.9839) mem 68106MB [2022-12-19 15:44:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][290/1519] eta 0:20:36 lr 0.000028 time 0.9269 (1.0065) model_time 0.9267 (1.0045) loss 1.0256 (0.9043) grad_norm 5.7338 (8.6766/1.9764) mem 68106MB [2022-12-19 15:44:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][300/1519] eta 0:20:26 lr 0.000028 time 0.9335 (1.0064) model_time 0.9334 (1.0046) loss 0.8259 (0.9054) grad_norm 8.1148 (8.6530/1.9531) mem 68106MB [2022-12-19 15:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][310/1519] eta 0:20:16 lr 0.000028 time 0.9180 (1.0062) model_time 0.9178 (1.0043) loss 0.7188 (0.9040) grad_norm 16.7978 (8.6646/2.0468) mem 68106MB [2022-12-19 15:45:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][320/1519] eta 0:20:07 lr 0.000028 time 0.9280 (1.0069) model_time 0.9279 (1.0051) loss 1.1157 (0.9044) grad_norm 16.2006 (8.6807/2.1160) mem 68106MB [2022-12-19 15:45:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][330/1519] eta 0:19:57 lr 0.000028 time 0.9929 (1.0068) model_time 0.9927 (1.0050) loss 0.8638 (0.9063) grad_norm 7.7157 (8.6661/2.1056) mem 68106MB [2022-12-19 15:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][340/1519] eta 0:19:47 lr 0.000028 time 1.1893 (1.0073) model_time 1.1891 (1.0056) loss 1.2862 (0.9072) grad_norm 9.8515 (8.6848/2.1147) mem 68106MB [2022-12-19 15:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][350/1519] eta 0:19:37 lr 0.000028 time 0.9300 (1.0075) model_time 0.9299 (1.0058) loss 0.9129 (0.9059) grad_norm 8.2328 (8.6609/2.0934) mem 68106MB [2022-12-19 15:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][360/1519] eta 0:19:27 lr 0.000028 time 0.9322 (1.0072) model_time 0.9320 (1.0056) loss 0.8397 (0.9051) grad_norm 10.6135 (8.7031/2.1600) mem 68106MB [2022-12-19 15:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][370/1519] eta 0:19:17 lr 0.000028 time 0.9287 (1.0070) model_time 0.9286 (1.0054) loss 0.9112 (0.9065) grad_norm 9.0760 (8.6703/2.1452) mem 68106MB [2022-12-19 15:46:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][380/1519] eta 0:19:06 lr 0.000028 time 0.9352 (1.0067) model_time 0.9350 (1.0052) loss 1.0308 (0.9054) grad_norm 7.9251 (8.6593/2.1273) mem 68106MB [2022-12-19 15:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][390/1519] eta 0:18:56 lr 0.000028 time 1.0242 (1.0068) model_time 1.0241 (1.0052) loss 0.8996 (0.9053) grad_norm 8.1832 (8.6423/2.1051) mem 68106MB [2022-12-19 15:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][400/1519] eta 0:18:46 lr 0.000028 time 0.9731 (1.0068) model_time 0.9729 (1.0053) loss 0.8279 (0.9044) grad_norm 14.3269 (8.6864/2.1321) mem 68106MB [2022-12-19 15:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][410/1519] eta 0:18:37 lr 0.000028 time 0.9249 (1.0075) model_time 0.9248 (1.0060) loss 0.9932 (0.9059) grad_norm 8.7858 (8.6859/2.1282) mem 68106MB [2022-12-19 15:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][420/1519] eta 0:18:27 lr 0.000028 time 0.9066 (1.0079) model_time 0.9064 (1.0064) loss 0.9443 (0.9048) grad_norm 7.3754 (8.6691/2.1102) mem 68106MB [2022-12-19 15:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][430/1519] eta 0:18:17 lr 0.000028 time 0.9366 (1.0082) model_time 0.9365 (1.0067) loss 0.9660 (0.9067) grad_norm 5.8829 (8.6437/2.1020) mem 68106MB [2022-12-19 15:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][440/1519] eta 0:18:07 lr 0.000028 time 0.9292 (1.0079) model_time 0.9290 (1.0065) loss 0.9365 (0.9071) grad_norm 8.4536 (8.6458/2.0804) mem 68106MB [2022-12-19 15:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][450/1519] eta 0:17:57 lr 0.000028 time 0.9276 (1.0079) model_time 0.9275 (1.0065) loss 0.7937 (0.9059) grad_norm 11.2694 (8.6683/2.0671) mem 68106MB [2022-12-19 15:47:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][460/1519] eta 0:17:47 lr 0.000028 time 0.9312 (1.0077) model_time 0.9309 (1.0063) loss 0.9068 (0.9040) grad_norm 7.8307 (8.6429/2.0620) mem 68106MB [2022-12-19 15:47:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][470/1519] eta 0:17:36 lr 0.000028 time 0.9217 (1.0074) model_time 0.9215 (1.0061) loss 0.9007 (0.9035) grad_norm 10.7374 (8.6497/2.0564) mem 68106MB [2022-12-19 15:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][480/1519] eta 0:17:26 lr 0.000028 time 0.9334 (1.0074) model_time 0.9332 (1.0061) loss 0.7342 (0.9050) grad_norm 7.4395 (8.6301/2.0487) mem 68106MB [2022-12-19 15:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][490/1519] eta 0:17:16 lr 0.000028 time 0.9257 (1.0072) model_time 0.9255 (1.0059) loss 0.9163 (0.9072) grad_norm 8.9631 (8.6561/2.0747) mem 68106MB [2022-12-19 15:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][500/1519] eta 0:17:06 lr 0.000028 time 0.9293 (1.0072) model_time 0.9291 (1.0059) loss 1.2117 (0.9088) grad_norm 8.6563 (8.6394/2.0587) mem 68106MB [2022-12-19 15:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][510/1519] eta 0:16:56 lr 0.000028 time 1.0249 (1.0072) model_time 1.0247 (1.0060) loss 0.9430 (0.9089) grad_norm 8.1817 (8.6214/2.0483) mem 68106MB [2022-12-19 15:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][520/1519] eta 0:16:46 lr 0.000028 time 0.9839 (1.0072) model_time 0.9837 (1.0059) loss 0.9074 (0.9102) grad_norm 9.0290 (8.6464/2.0495) mem 68106MB [2022-12-19 15:48:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][530/1519] eta 0:16:35 lr 0.000028 time 0.9245 (1.0070) model_time 0.9244 (1.0057) loss 0.9592 (0.9117) grad_norm 6.8467 (8.6377/2.0418) mem 68106MB [2022-12-19 15:48:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][540/1519] eta 0:16:25 lr 0.000028 time 0.9313 (1.0068) model_time 0.9310 (1.0056) loss 1.0127 (0.9111) grad_norm 6.3866 (8.6427/2.0409) mem 68106MB [2022-12-19 15:49:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][550/1519] eta 0:16:15 lr 0.000028 time 0.9341 (1.0067) model_time 0.9340 (1.0055) loss 0.7110 (0.9097) grad_norm 13.5154 (8.6644/2.0744) mem 68106MB [2022-12-19 15:49:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][560/1519] eta 0:16:05 lr 0.000028 time 0.9265 (1.0066) model_time 0.9263 (1.0054) loss 0.7678 (0.9113) grad_norm 7.0181 (8.6724/2.0679) mem 68106MB [2022-12-19 15:49:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][570/1519] eta 0:15:55 lr 0.000028 time 1.0099 (1.0066) model_time 1.0097 (1.0055) loss 1.0447 (0.9110) grad_norm 7.8545 (8.6564/2.0542) mem 68106MB [2022-12-19 15:49:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][580/1519] eta 0:15:45 lr 0.000028 time 0.9925 (1.0068) model_time 0.9924 (1.0056) loss 0.7568 (0.9110) grad_norm 9.9754 (8.6567/2.0487) mem 68106MB [2022-12-19 15:49:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][590/1519] eta 0:15:35 lr 0.000028 time 0.9272 (1.0069) model_time 0.9270 (1.0057) loss 0.7585 (0.9101) grad_norm 7.9655 (8.6630/2.0361) mem 68106MB [2022-12-19 15:49:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][600/1519] eta 0:15:25 lr 0.000028 time 0.9251 (1.0068) model_time 0.9250 (1.0056) loss 0.9794 (0.9103) grad_norm 10.2539 (8.6608/2.0388) mem 68106MB [2022-12-19 15:50:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][610/1519] eta 0:15:15 lr 0.000028 time 0.8863 (1.0072) model_time 0.8861 (1.0061) loss 0.9358 (0.9106) grad_norm 11.1457 (8.6589/2.0434) mem 68106MB [2022-12-19 15:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][620/1519] eta 0:15:05 lr 0.000028 time 0.9274 (1.0070) model_time 0.9272 (1.0059) loss 0.7270 (0.9120) grad_norm 6.6300 (8.6382/2.0488) mem 68106MB [2022-12-19 15:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][630/1519] eta 0:14:55 lr 0.000028 time 0.9288 (1.0070) model_time 0.9286 (1.0059) loss 0.7329 (0.9125) grad_norm 8.3330 (8.6343/2.0414) mem 68106MB [2022-12-19 15:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][640/1519] eta 0:14:45 lr 0.000028 time 0.9314 (1.0069) model_time 0.9312 (1.0058) loss 0.8188 (0.9122) grad_norm 6.0569 (8.6034/2.0095) mem 68106MB [2022-12-19 15:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][650/1519] eta 0:14:34 lr 0.000028 time 0.9348 (1.0069) model_time 0.9347 (1.0058) loss 0.7652 (0.9116) grad_norm 8.3346 (8.6507/2.0698) mem 68106MB [2022-12-19 15:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][660/1519] eta 0:14:24 lr 0.000028 time 0.9360 (1.0069) model_time 0.9358 (1.0058) loss 0.6919 (0.9106) grad_norm 6.4654 (8.6197/2.0771) mem 68106MB [2022-12-19 15:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][670/1519] eta 0:14:14 lr 0.000028 time 0.9291 (1.0068) model_time 0.9289 (1.0057) loss 0.9035 (0.9112) grad_norm 7.6174 (8.5718/2.0443) mem 68106MB [2022-12-19 15:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][680/1519] eta 0:14:04 lr 0.000028 time 0.9296 (1.0066) model_time 0.9294 (1.0056) loss 0.8366 (0.9109) grad_norm 10.3925 (8.5775/2.0448) mem 68106MB [2022-12-19 15:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][690/1519] eta 0:13:54 lr 0.000028 time 0.9291 (1.0065) model_time 0.9289 (1.0055) loss 0.8703 (0.9097) grad_norm 9.6432 (8.5737/2.0352) mem 68106MB [2022-12-19 15:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][700/1519] eta 0:13:44 lr 0.000028 time 0.9388 (1.0064) model_time 0.9386 (1.0054) loss 0.7353 (0.9103) grad_norm 8.3537 (8.6066/2.0620) mem 68106MB [2022-12-19 15:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][710/1519] eta 0:13:34 lr 0.000028 time 0.9189 (1.0064) model_time 0.9187 (1.0054) loss 1.2270 (0.9101) grad_norm 9.8722 (8.6040/2.0725) mem 68106MB [2022-12-19 15:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][720/1519] eta 0:13:24 lr 0.000028 time 0.9309 (1.0064) model_time 0.9307 (1.0054) loss 0.6849 (0.9097) grad_norm 7.3183 (8.6288/2.0846) mem 68106MB [2022-12-19 15:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][730/1519] eta 0:13:14 lr 0.000028 time 0.9273 (1.0064) model_time 0.9272 (1.0054) loss 0.8136 (0.9101) grad_norm 9.2346 (8.6167/2.0875) mem 68106MB [2022-12-19 15:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][740/1519] eta 0:13:04 lr 0.000028 time 0.9272 (1.0065) model_time 0.9270 (1.0055) loss 0.8722 (0.9098) grad_norm 8.4039 (8.5677/2.0461) mem 68106MB [2022-12-19 15:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][750/1519] eta 0:12:53 lr 0.000028 time 0.9307 (1.0065) model_time 0.9305 (1.0055) loss 1.2083 (0.9100) grad_norm 6.8687 (8.5417/2.0565) mem 68106MB [2022-12-19 15:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][760/1519] eta 0:12:43 lr 0.000028 time 0.9330 (1.0063) model_time 0.9329 (1.0054) loss 0.7743 (0.9095) grad_norm 7.2574 (8.5355/2.0633) mem 68106MB [2022-12-19 15:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][770/1519] eta 0:12:33 lr 0.000028 time 0.9289 (1.0063) model_time 0.9287 (1.0054) loss 1.0537 (0.9101) grad_norm 12.3593 (8.5657/2.0702) mem 68106MB [2022-12-19 15:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][780/1519] eta 0:12:23 lr 0.000028 time 0.9299 (1.0063) model_time 0.9297 (1.0053) loss 0.7346 (0.9100) grad_norm 9.4756 (8.6095/2.0802) mem 68106MB [2022-12-19 15:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][790/1519] eta 0:12:13 lr 0.000028 time 0.9256 (1.0062) model_time 0.9250 (1.0053) loss 0.7589 (0.9104) grad_norm 13.8592 (8.6268/2.0985) mem 68106MB [2022-12-19 15:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][800/1519] eta 0:12:03 lr 0.000028 time 0.9316 (1.0061) model_time 0.9314 (1.0052) loss 1.0998 (0.9106) grad_norm 9.0666 (8.6423/2.0826) mem 68106MB [2022-12-19 15:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][810/1519] eta 0:11:53 lr 0.000028 time 0.9352 (1.0061) model_time 0.9351 (1.0052) loss 0.6905 (0.9108) grad_norm 8.6165 (8.6453/2.0368) mem 68106MB [2022-12-19 15:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][820/1519] eta 0:11:43 lr 0.000028 time 0.9306 (1.0060) model_time 0.9304 (1.0051) loss 0.7985 (0.9107) grad_norm 7.0212 (8.6587/2.0370) mem 68106MB [2022-12-19 15:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][830/1519] eta 0:11:33 lr 0.000028 time 0.9282 (1.0060) model_time 0.9280 (1.0051) loss 1.0614 (0.9103) grad_norm 7.0802 (8.6438/2.0579) mem 68106MB [2022-12-19 15:53:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][840/1519] eta 0:11:23 lr 0.000028 time 0.9315 (1.0061) model_time 0.9313 (1.0052) loss 1.1661 (0.9107) grad_norm 9.9404 (8.6260/2.0252) mem 68106MB [2022-12-19 15:54:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][850/1519] eta 0:11:13 lr 0.000028 time 0.9259 (1.0060) model_time 0.9257 (1.0051) loss 1.2420 (0.9106) grad_norm 6.5441 (8.5991/2.0287) mem 68106MB [2022-12-19 15:54:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][860/1519] eta 0:11:02 lr 0.000028 time 0.9254 (1.0059) model_time 0.9251 (1.0050) loss 0.6964 (0.9109) grad_norm 6.1629 (8.6014/2.0264) mem 68106MB [2022-12-19 15:54:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][870/1519] eta 0:10:52 lr 0.000028 time 0.9301 (1.0059) model_time 0.9299 (1.0050) loss 0.9180 (0.9124) grad_norm 6.7176 (8.6063/2.0273) mem 68106MB [2022-12-19 15:54:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][880/1519] eta 0:10:42 lr 0.000028 time 0.9282 (1.0058) model_time 0.9280 (1.0049) loss 0.9561 (0.9122) grad_norm 8.0721 (8.5859/2.0268) mem 68106MB [2022-12-19 15:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][890/1519] eta 0:10:32 lr 0.000028 time 0.9200 (1.0059) model_time 0.9199 (1.0050) loss 1.2875 (0.9123) grad_norm 5.8125 (8.6203/2.1301) mem 68106MB [2022-12-19 15:54:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][900/1519] eta 0:10:22 lr 0.000028 time 0.9404 (1.0060) model_time 0.9402 (1.0051) loss 1.0415 (0.9123) grad_norm 9.0633 (8.6178/2.1321) mem 68106MB [2022-12-19 15:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][910/1519] eta 0:10:12 lr 0.000028 time 0.9323 (1.0059) model_time 0.9322 (1.0050) loss 0.9661 (0.9119) grad_norm 8.3708 (8.6216/2.1062) mem 68106MB [2022-12-19 15:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][920/1519] eta 0:10:02 lr 0.000028 time 0.9317 (1.0059) model_time 0.9316 (1.0050) loss 1.3028 (0.9135) grad_norm 8.3947 (8.6118/2.0624) mem 68106MB [2022-12-19 15:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][930/1519] eta 0:09:52 lr 0.000028 time 0.9312 (1.0059) model_time 0.9311 (1.0051) loss 0.8406 (0.9131) grad_norm 7.6934 (8.6291/2.0570) mem 68106MB [2022-12-19 15:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][940/1519] eta 0:09:42 lr 0.000028 time 0.9308 (1.0061) model_time 0.9307 (1.0053) loss 0.7299 (0.9135) grad_norm 6.8150 (8.5924/2.0436) mem 68106MB [2022-12-19 15:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][950/1519] eta 0:09:32 lr 0.000028 time 0.9317 (1.0061) model_time 0.9316 (1.0052) loss 0.8870 (0.9131) grad_norm 7.3242 (8.5896/2.0460) mem 68106MB [2022-12-19 15:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][960/1519] eta 0:09:22 lr 0.000028 time 0.9305 (1.0060) model_time 0.9304 (1.0052) loss 0.9380 (0.9128) grad_norm 9.0709 (8.6127/2.0887) mem 68106MB [2022-12-19 15:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][970/1519] eta 0:09:12 lr 0.000028 time 0.9284 (1.0061) model_time 0.9282 (1.0052) loss 0.8963 (0.9128) grad_norm 10.0096 (8.6533/2.0875) mem 68106MB [2022-12-19 15:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][980/1519] eta 0:09:02 lr 0.000028 time 0.9317 (1.0060) model_time 0.9315 (1.0051) loss 0.8192 (0.9132) grad_norm 8.8217 (8.6820/2.1086) mem 68106MB [2022-12-19 15:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][990/1519] eta 0:08:52 lr 0.000028 time 0.9360 (1.0060) model_time 0.9359 (1.0052) loss 0.9000 (0.9133) grad_norm 7.7751 (8.7170/2.1453) mem 68106MB [2022-12-19 15:56:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1000/1519] eta 0:08:42 lr 0.000028 time 0.9348 (1.0059) model_time 0.9347 (1.0051) loss 0.6955 (0.9127) grad_norm 7.4137 (8.7047/2.1359) mem 68106MB [2022-12-19 15:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1010/1519] eta 0:08:32 lr 0.000028 time 0.9318 (1.0059) model_time 0.9317 (1.0051) loss 0.7226 (0.9129) grad_norm 10.8448 (8.7300/2.1329) mem 68106MB [2022-12-19 15:56:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1020/1519] eta 0:08:21 lr 0.000028 time 0.9295 (1.0059) model_time 0.9294 (1.0050) loss 0.9762 (0.9129) grad_norm 26.6841 (8.7816/2.3785) mem 68106MB [2022-12-19 15:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1030/1519] eta 0:08:11 lr 0.000028 time 0.9299 (1.0059) model_time 0.9297 (1.0050) loss 0.7324 (0.9124) grad_norm 13.7778 (8.8188/2.3925) mem 68106MB [2022-12-19 15:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1040/1519] eta 0:08:01 lr 0.000028 time 0.9344 (1.0059) model_time 0.9342 (1.0050) loss 0.7628 (0.9119) grad_norm 13.4002 (8.8291/2.4086) mem 68106MB [2022-12-19 15:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1050/1519] eta 0:07:51 lr 0.000028 time 1.0773 (1.0060) model_time 1.0772 (1.0051) loss 1.0115 (0.9121) grad_norm 12.7198 (8.8114/2.4214) mem 68106MB [2022-12-19 15:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1060/1519] eta 0:07:41 lr 0.000028 time 0.9241 (1.0059) model_time 0.9239 (1.0051) loss 0.8753 (0.9122) grad_norm 8.4776 (8.8475/2.4273) mem 68106MB [2022-12-19 15:57:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1070/1519] eta 0:07:31 lr 0.000028 time 0.9260 (1.0059) model_time 0.9259 (1.0050) loss 0.7131 (0.9126) grad_norm 7.7111 (8.8583/2.4345) mem 68106MB [2022-12-19 15:57:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1080/1519] eta 0:07:21 lr 0.000028 time 0.9334 (1.0058) model_time 0.9333 (1.0050) loss 0.8374 (0.9119) grad_norm 9.9378 (8.8830/2.4498) mem 68106MB [2022-12-19 15:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1090/1519] eta 0:07:11 lr 0.000028 time 0.9323 (1.0057) model_time 0.9322 (1.0049) loss 0.7146 (0.9116) grad_norm 9.6277 (8.8837/2.4294) mem 68106MB [2022-12-19 15:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1100/1519] eta 0:07:01 lr 0.000028 time 0.9765 (1.0057) model_time 0.9764 (1.0049) loss 0.8388 (0.9117) grad_norm 10.2073 (8.9304/2.4507) mem 68106MB [2022-12-19 15:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1110/1519] eta 0:06:51 lr 0.000028 time 0.9348 (1.0059) model_time 0.9347 (1.0051) loss 1.0656 (0.9121) grad_norm 10.1608 (8.9327/2.4536) mem 68106MB [2022-12-19 15:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1120/1519] eta 0:06:41 lr 0.000028 time 0.9292 (1.0059) model_time 0.9291 (1.0051) loss 1.0732 (0.9128) grad_norm 6.8459 (8.9250/2.4529) mem 68106MB [2022-12-19 15:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1130/1519] eta 0:06:31 lr 0.000028 time 0.9300 (1.0058) model_time 0.9299 (1.0050) loss 0.7726 (0.9128) grad_norm 11.4989 (8.9488/2.4537) mem 68106MB [2022-12-19 15:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1140/1519] eta 0:06:21 lr 0.000028 time 0.9315 (1.0057) model_time 0.9313 (1.0049) loss 1.1635 (0.9134) grad_norm 7.3252 (8.9487/2.4472) mem 68106MB [2022-12-19 15:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1150/1519] eta 0:06:11 lr 0.000028 time 0.9352 (1.0057) model_time 0.9350 (1.0049) loss 1.0419 (0.9133) grad_norm 17.3109 (8.9628/2.4642) mem 68106MB [2022-12-19 15:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1160/1519] eta 0:06:01 lr 0.000028 time 0.9385 (1.0057) model_time 0.9384 (1.0049) loss 0.7561 (0.9140) grad_norm 6.7324 (8.9567/2.4811) mem 68106MB [2022-12-19 15:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1170/1519] eta 0:05:50 lr 0.000028 time 0.9377 (1.0057) model_time 0.9376 (1.0049) loss 0.9869 (0.9147) grad_norm 7.8005 (8.9670/2.4865) mem 68106MB [2022-12-19 15:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1180/1519] eta 0:05:40 lr 0.000028 time 0.9321 (1.0056) model_time 0.9319 (1.0049) loss 0.8286 (0.9147) grad_norm 20.8612 (9.0032/2.5721) mem 68106MB [2022-12-19 15:59:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1190/1519] eta 0:05:30 lr 0.000028 time 0.9421 (1.0056) model_time 0.9420 (1.0048) loss 0.8219 (0.9152) grad_norm 6.5096 (8.9796/2.5763) mem 68106MB [2022-12-19 15:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1200/1519] eta 0:05:20 lr 0.000028 time 1.0003 (1.0057) model_time 1.0002 (1.0049) loss 0.9350 (0.9157) grad_norm 7.9617 (8.9900/2.5615) mem 68106MB [2022-12-19 16:00:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1210/1519] eta 0:05:10 lr 0.000028 time 0.9307 (1.0058) model_time 0.9306 (1.0050) loss 0.9177 (0.9162) grad_norm 7.8853 (8.9812/2.5534) mem 68106MB [2022-12-19 16:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1220/1519] eta 0:05:00 lr 0.000028 time 0.9315 (1.0057) model_time 0.9314 (1.0049) loss 0.7935 (0.9153) grad_norm 8.8686 (8.9846/2.5579) mem 68106MB [2022-12-19 16:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1230/1519] eta 0:04:50 lr 0.000028 time 1.2093 (1.0059) model_time 1.2092 (1.0052) loss 1.2267 (0.9163) grad_norm 8.2636 (8.9767/2.5598) mem 68106MB [2022-12-19 16:00:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1240/1519] eta 0:04:40 lr 0.000028 time 0.9306 (1.0059) model_time 0.9305 (1.0051) loss 0.9514 (0.9160) grad_norm 18.1724 (9.0086/2.6147) mem 68106MB [2022-12-19 16:00:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1250/1519] eta 0:04:30 lr 0.000028 time 0.9770 (1.0059) model_time 0.9768 (1.0051) loss 0.8581 (0.9166) grad_norm 8.8048 (8.9391/2.5751) mem 68106MB [2022-12-19 16:00:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1260/1519] eta 0:04:20 lr 0.000028 time 0.9380 (1.0058) model_time 0.9379 (1.0051) loss 0.7914 (0.9166) grad_norm 8.5637 (8.9539/2.5657) mem 68106MB [2022-12-19 16:01:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1270/1519] eta 0:04:10 lr 0.000028 time 0.9298 (1.0058) model_time 0.9296 (1.0050) loss 0.9299 (0.9163) grad_norm 8.5225 (8.9448/2.5658) mem 68106MB [2022-12-19 16:01:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1280/1519] eta 0:04:00 lr 0.000028 time 0.9845 (1.0058) model_time 0.9843 (1.0050) loss 0.9071 (0.9162) grad_norm 10.3286 (9.0035/2.5987) mem 68106MB [2022-12-19 16:01:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1290/1519] eta 0:03:50 lr 0.000028 time 0.9302 (1.0057) model_time 0.9300 (1.0050) loss 1.1803 (0.9162) grad_norm 9.7033 (9.0399/2.6199) mem 68106MB [2022-12-19 16:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1300/1519] eta 0:03:40 lr 0.000028 time 0.9314 (1.0058) model_time 0.9312 (1.0050) loss 0.9936 (0.9165) grad_norm 13.4442 (9.0577/2.6145) mem 68106MB [2022-12-19 16:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1310/1519] eta 0:03:30 lr 0.000028 time 0.9321 (1.0057) model_time 0.9319 (1.0050) loss 0.9881 (0.9167) grad_norm 7.8436 (9.0493/2.6097) mem 68106MB [2022-12-19 16:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1320/1519] eta 0:03:20 lr 0.000028 time 0.9314 (1.0058) model_time 0.9313 (1.0051) loss 1.0444 (0.9163) grad_norm 5.9943 (9.0377/2.6133) mem 68106MB [2022-12-19 16:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1330/1519] eta 0:03:10 lr 0.000028 time 0.9254 (1.0058) model_time 0.9252 (1.0051) loss 1.1564 (0.9169) grad_norm 8.9731 (9.0386/2.6077) mem 68106MB [2022-12-19 16:02:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1340/1519] eta 0:03:00 lr 0.000028 time 0.9309 (1.0058) model_time 0.9308 (1.0051) loss 0.9621 (0.9168) grad_norm 8.0905 (9.0533/2.6020) mem 68106MB [2022-12-19 16:02:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1350/1519] eta 0:02:49 lr 0.000028 time 0.9323 (1.0058) model_time 0.9321 (1.0051) loss 0.9670 (0.9168) grad_norm 6.9305 (9.0762/2.5916) mem 68106MB [2022-12-19 16:02:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1360/1519] eta 0:02:39 lr 0.000028 time 0.9362 (1.0057) model_time 0.9361 (1.0050) loss 0.9353 (0.9166) grad_norm 8.4696 (9.0701/2.5810) mem 68106MB [2022-12-19 16:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1370/1519] eta 0:02:29 lr 0.000028 time 0.9351 (1.0058) model_time 0.9348 (1.0050) loss 0.8563 (0.9165) grad_norm 11.9230 (9.0638/2.5876) mem 68106MB [2022-12-19 16:02:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1380/1519] eta 0:02:19 lr 0.000028 time 0.9945 (1.0058) model_time 0.9943 (1.0051) loss 0.7489 (0.9160) grad_norm 8.7538 (9.0422/2.5811) mem 68106MB [2022-12-19 16:03:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1390/1519] eta 0:02:09 lr 0.000028 time 0.9393 (1.0058) model_time 0.9391 (1.0051) loss 0.9577 (0.9161) grad_norm 7.4016 (9.0202/2.5694) mem 68106MB [2022-12-19 16:03:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1400/1519] eta 0:01:59 lr 0.000028 time 0.9321 (1.0058) model_time 0.9319 (1.0051) loss 0.9806 (0.9159) grad_norm 8.1433 (9.0002/2.5699) mem 68106MB [2022-12-19 16:03:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1410/1519] eta 0:01:49 lr 0.000028 time 0.9322 (1.0058) model_time 0.9321 (1.0051) loss 0.9213 (0.9157) grad_norm 6.0048 (8.9729/2.5682) mem 68106MB [2022-12-19 16:03:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1420/1519] eta 0:01:39 lr 0.000028 time 0.9428 (1.0058) model_time 0.9427 (1.0051) loss 0.9411 (0.9154) grad_norm 7.9431 (8.9818/2.5662) mem 68106MB [2022-12-19 16:03:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1430/1519] eta 0:01:29 lr 0.000028 time 0.9833 (1.0058) model_time 0.9831 (1.0051) loss 0.7497 (0.9149) grad_norm 8.7885 (8.9811/2.5343) mem 68106MB [2022-12-19 16:03:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1440/1519] eta 0:01:19 lr 0.000028 time 0.9312 (1.0057) model_time 0.9311 (1.0050) loss 0.8658 (0.9145) grad_norm 6.0099 (8.9903/2.5327) mem 68106MB [2022-12-19 16:04:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1450/1519] eta 0:01:09 lr 0.000028 time 0.9331 (1.0057) model_time 0.9329 (1.0050) loss 0.8659 (0.9148) grad_norm 7.9326 (9.0190/2.5364) mem 68106MB [2022-12-19 16:04:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1460/1519] eta 0:00:59 lr 0.000028 time 0.9194 (1.0058) model_time 0.9192 (1.0051) loss 1.2945 (0.9147) grad_norm 7.6307 (9.0603/2.5393) mem 68106MB [2022-12-19 16:04:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1470/1519] eta 0:00:49 lr 0.000028 time 0.9318 (1.0057) model_time 0.9316 (1.0050) loss 0.8538 (0.9148) grad_norm 7.8672 (9.0788/2.5480) mem 68106MB [2022-12-19 16:04:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1480/1519] eta 0:00:39 lr 0.000028 time 0.9338 (1.0057) model_time 0.9336 (1.0050) loss 0.7082 (0.9142) grad_norm 7.0601 (9.0843/2.5416) mem 68106MB [2022-12-19 16:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1490/1519] eta 0:00:29 lr 0.000028 time 0.9821 (1.0057) model_time 0.9819 (1.0050) loss 0.8019 (0.9142) grad_norm 8.1786 (9.0556/2.4538) mem 68106MB [2022-12-19 16:04:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1500/1519] eta 0:00:19 lr 0.000028 time 0.9323 (1.0057) model_time 0.9320 (1.0050) loss 0.7425 (0.9144) grad_norm 8.9482 (9.0630/2.4517) mem 68106MB [2022-12-19 16:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [26/100][1510/1519] eta 0:00:09 lr 0.000028 time 0.9268 (1.0057) model_time 0.9267 (1.0050) loss 1.0315 (0.9146) grad_norm 8.1885 (9.0890/2.4685) mem 68106MB [2022-12-19 16:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 26 training takes 0:25:27 [2022-12-19 16:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_26.pth saving...... [2022-12-19 16:05:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_26.pth saved !!! [2022-12-19 16:05:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.628 (0.628) Loss 0.5177 (0.5177) Acc@1 90.972 (90.972) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 16:05:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.326) Loss 0.5264 (0.5007) Acc@1 90.625 (91.477) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-19 16:05:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.312) Loss 0.4552 (0.5003) Acc@1 91.667 (91.154) Acc@5 98.958 (98.247) Mem 68106MB [2022-12-19 16:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.291 (0.307) Loss 0.5792 (0.5032) Acc@1 89.931 (91.062) Acc@5 98.264 (98.241) Mem 68106MB [2022-12-19 16:05:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.306) Loss 0.4625 (0.4958) Acc@1 92.014 (91.091) Acc@5 98.958 (98.332) Mem 68106MB [2022-12-19 16:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.305) Loss 0.5074 (0.4930) Acc@1 88.542 (91.074) Acc@5 99.653 (98.386) Mem 68106MB [2022-12-19 16:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.304) Loss 0.5209 (0.4937) Acc@1 88.542 (91.046) Acc@5 97.569 (98.366) Mem 68106MB [2022-12-19 16:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5692 (0.4960) Acc@1 90.625 (91.055) Acc@5 98.264 (98.367) Mem 68106MB [2022-12-19 16:06:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.4303 (0.4950) Acc@1 92.708 (91.079) Acc@5 98.264 (98.384) Mem 68106MB [2022-12-19 16:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:26] * Acc@1 91.073 Acc@5 98.391 [2022-12-19 16:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.1% [2022-12-19 16:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 16:06:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 16:06:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.07% [2022-12-19 16:06:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][0/1519] eta 0:33:10 lr 0.000028 time 1.3105 (1.3105) model_time 0.9126 (0.9126) loss 0.7030 (0.7030) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 16:06:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][10/1519] eta 0:26:07 lr 0.000028 time 0.9361 (1.0389) model_time 0.9360 (1.0023) loss 0.8003 (0.8671) grad_norm 8.7673 (8.3146/1.1470) mem 68106MB [2022-12-19 16:06:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][20/1519] eta 0:25:36 lr 0.000028 time 0.9184 (1.0247) model_time 0.9181 (1.0054) loss 0.7765 (0.8719) grad_norm 9.9281 (8.4194/1.1479) mem 68106MB [2022-12-19 16:07:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][30/1519] eta 0:25:20 lr 0.000028 time 0.9367 (1.0209) model_time 0.9366 (1.0077) loss 1.1846 (0.8965) grad_norm 7.0563 (8.3066/1.0630) mem 68106MB [2022-12-19 16:07:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][40/1519] eta 0:25:06 lr 0.000028 time 0.9706 (1.0185) model_time 0.9705 (1.0083) loss 1.1374 (0.8931) grad_norm 10.3234 (8.7691/1.5541) mem 68106MB [2022-12-19 16:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][50/1519] eta 0:24:49 lr 0.000028 time 0.9329 (1.0142) model_time 0.9328 (1.0060) loss 0.8798 (0.8929) grad_norm 7.1439 (8.6037/1.4847) mem 68106MB [2022-12-19 16:07:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][60/1519] eta 0:24:42 lr 0.000028 time 0.9341 (1.0158) model_time 0.9340 (1.0089) loss 0.7082 (0.8863) grad_norm 7.3725 (8.3845/1.4713) mem 68106MB [2022-12-19 16:07:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][70/1519] eta 0:24:29 lr 0.000028 time 0.9229 (1.0142) model_time 0.9228 (1.0081) loss 1.0847 (0.8831) grad_norm 9.9585 (8.7446/1.8201) mem 68106MB [2022-12-19 16:07:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][80/1519] eta 0:24:18 lr 0.000028 time 0.9268 (1.0133) model_time 0.9266 (1.0079) loss 1.3358 (0.8911) grad_norm 8.7258 (8.8206/1.8188) mem 68106MB [2022-12-19 16:08:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][90/1519] eta 0:24:06 lr 0.000028 time 0.9678 (1.0123) model_time 0.9676 (1.0075) loss 0.9247 (0.8870) grad_norm 15.2191 (8.9292/2.0482) mem 68106MB [2022-12-19 16:08:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][100/1519] eta 0:23:54 lr 0.000028 time 0.9276 (1.0106) model_time 0.9275 (1.0063) loss 0.7407 (0.8896) grad_norm 8.9874 (8.9184/1.9541) mem 68106MB [2022-12-19 16:08:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][110/1519] eta 0:23:42 lr 0.000028 time 0.9335 (1.0098) model_time 0.9334 (1.0058) loss 1.0219 (0.8865) grad_norm 6.0047 (9.0203/2.2620) mem 68106MB [2022-12-19 16:08:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][120/1519] eta 0:23:31 lr 0.000028 time 0.9331 (1.0090) model_time 0.9330 (1.0053) loss 0.8622 (0.8885) grad_norm 7.6640 (9.0483/2.2306) mem 68106MB [2022-12-19 16:08:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][130/1519] eta 0:23:20 lr 0.000028 time 0.9311 (1.0081) model_time 0.9310 (1.0047) loss 1.0247 (0.8846) grad_norm 8.8922 (8.9553/2.2012) mem 68106MB [2022-12-19 16:08:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][140/1519] eta 0:23:09 lr 0.000028 time 0.9298 (1.0075) model_time 0.9296 (1.0042) loss 1.0626 (0.8867) grad_norm 10.4818 (8.8654/2.1902) mem 68106MB [2022-12-19 16:09:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][150/1519] eta 0:22:59 lr 0.000028 time 0.9291 (1.0074) model_time 0.9290 (1.0043) loss 0.8185 (0.8875) grad_norm 10.1906 (8.8468/2.1352) mem 68106MB [2022-12-19 16:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][160/1519] eta 0:22:48 lr 0.000028 time 0.9302 (1.0069) model_time 0.9301 (1.0040) loss 0.9313 (0.8862) grad_norm 7.8911 (8.8303/2.1007) mem 68106MB [2022-12-19 16:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][170/1519] eta 0:22:37 lr 0.000028 time 0.9211 (1.0064) model_time 0.9209 (1.0037) loss 0.7052 (0.8881) grad_norm 10.0068 (8.8180/2.0789) mem 68106MB [2022-12-19 16:09:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][180/1519] eta 0:22:28 lr 0.000028 time 1.0332 (1.0069) model_time 1.0330 (1.0043) loss 1.0195 (0.8911) grad_norm 8.3486 (8.7866/2.0632) mem 68106MB [2022-12-19 16:09:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][190/1519] eta 0:22:18 lr 0.000028 time 0.9285 (1.0070) model_time 0.9283 (1.0045) loss 0.7415 (0.8925) grad_norm 9.3905 (8.7444/2.0356) mem 68106MB [2022-12-19 16:09:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][200/1519] eta 0:22:07 lr 0.000028 time 0.9244 (1.0068) model_time 0.9242 (1.0044) loss 1.1897 (0.8965) grad_norm 8.1909 (8.7556/2.0335) mem 68106MB [2022-12-19 16:10:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][210/1519] eta 0:21:58 lr 0.000028 time 0.9271 (1.0070) model_time 0.9270 (1.0047) loss 0.9741 (0.8944) grad_norm 6.4897 (8.7002/2.0277) mem 68106MB [2022-12-19 16:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][220/1519] eta 0:21:47 lr 0.000028 time 0.9280 (1.0066) model_time 0.9279 (1.0044) loss 1.2347 (0.8969) grad_norm 8.0552 (8.6728/1.9979) mem 68106MB [2022-12-19 16:10:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][230/1519] eta 0:21:37 lr 0.000028 time 0.9203 (1.0064) model_time 0.9202 (1.0043) loss 0.8672 (0.8979) grad_norm 11.3161 (8.6876/1.9825) mem 68106MB [2022-12-19 16:10:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][240/1519] eta 0:21:26 lr 0.000028 time 0.9312 (1.0060) model_time 0.9310 (1.0040) loss 0.9363 (0.8996) grad_norm 17.0359 (8.7669/2.0856) mem 68106MB [2022-12-19 16:10:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][250/1519] eta 0:21:16 lr 0.000028 time 0.9324 (1.0059) model_time 0.9321 (1.0039) loss 0.7173 (0.9013) grad_norm 11.5319 (8.8505/2.1071) mem 68106MB [2022-12-19 16:10:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][260/1519] eta 0:21:06 lr 0.000028 time 0.9215 (1.0057) model_time 0.9214 (1.0038) loss 0.9282 (0.9016) grad_norm 14.4132 (8.9023/2.1402) mem 68106MB [2022-12-19 16:11:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][270/1519] eta 0:20:55 lr 0.000028 time 0.9776 (1.0056) model_time 0.9774 (1.0037) loss 1.0244 (0.8997) grad_norm 10.9576 (8.8825/2.1244) mem 68106MB [2022-12-19 16:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][280/1519] eta 0:20:45 lr 0.000028 time 0.9328 (1.0053) model_time 0.9327 (1.0035) loss 0.9903 (0.9035) grad_norm 6.9452 (8.8109/2.1283) mem 68106MB [2022-12-19 16:11:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][290/1519] eta 0:20:35 lr 0.000028 time 0.9337 (1.0051) model_time 0.9336 (1.0033) loss 0.7468 (0.9024) grad_norm 10.1883 (8.8254/2.1663) mem 68106MB [2022-12-19 16:11:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][300/1519] eta 0:20:25 lr 0.000028 time 0.9292 (1.0051) model_time 0.9291 (1.0034) loss 0.7379 (0.9025) grad_norm 10.6586 (8.8585/2.1628) mem 68106MB [2022-12-19 16:11:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][310/1519] eta 0:20:15 lr 0.000028 time 0.9369 (1.0051) model_time 0.9368 (1.0034) loss 1.1783 (0.9013) grad_norm 8.1836 (8.8648/2.1383) mem 68106MB [2022-12-19 16:11:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][320/1519] eta 0:20:05 lr 0.000028 time 0.9310 (1.0052) model_time 0.9308 (1.0035) loss 0.7828 (0.9032) grad_norm 7.2666 (8.8341/2.1286) mem 68106MB [2022-12-19 16:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][330/1519] eta 0:19:55 lr 0.000028 time 0.9344 (1.0054) model_time 0.9342 (1.0038) loss 0.8353 (0.9003) grad_norm 7.0395 (8.7894/2.1217) mem 68106MB [2022-12-19 16:12:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][340/1519] eta 0:19:45 lr 0.000028 time 0.9812 (1.0054) model_time 0.9811 (1.0039) loss 1.2725 (0.9044) grad_norm 11.4997 (8.7850/2.1077) mem 68106MB [2022-12-19 16:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][350/1519] eta 0:19:35 lr 0.000028 time 0.9335 (1.0055) model_time 0.9332 (1.0040) loss 0.9087 (0.9057) grad_norm 8.4851 (8.7793/2.0858) mem 68106MB [2022-12-19 16:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][360/1519] eta 0:19:25 lr 0.000028 time 1.0535 (1.0060) model_time 1.0533 (1.0045) loss 0.8891 (0.9071) grad_norm 10.0669 (8.7898/2.0940) mem 68106MB [2022-12-19 16:12:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][370/1519] eta 0:19:15 lr 0.000028 time 0.9261 (1.0061) model_time 0.9259 (1.0046) loss 0.8771 (0.9053) grad_norm 9.3961 (8.7862/2.0696) mem 68106MB [2022-12-19 16:12:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][380/1519] eta 0:19:05 lr 0.000028 time 0.9100 (1.0060) model_time 0.9098 (1.0046) loss 0.7402 (0.9053) grad_norm 12.3937 (8.8021/2.0675) mem 68106MB [2022-12-19 16:13:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][390/1519] eta 0:18:55 lr 0.000028 time 0.9879 (1.0060) model_time 0.9878 (1.0046) loss 0.9804 (0.9055) grad_norm 10.4355 (8.7927/2.0529) mem 68106MB [2022-12-19 16:13:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][400/1519] eta 0:18:45 lr 0.000028 time 0.9335 (1.0058) model_time 0.9333 (1.0044) loss 1.3744 (0.9071) grad_norm 7.1701 (8.7951/2.0418) mem 68106MB [2022-12-19 16:13:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][410/1519] eta 0:18:35 lr 0.000028 time 0.9312 (1.0058) model_time 0.9311 (1.0045) loss 0.9548 (0.9065) grad_norm 7.4543 (8.7810/2.0420) mem 68106MB [2022-12-19 16:13:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][420/1519] eta 0:18:25 lr 0.000028 time 0.9456 (1.0057) model_time 0.9455 (1.0044) loss 0.7938 (0.9075) grad_norm 6.0557 (8.7818/2.0352) mem 68106MB [2022-12-19 16:13:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][430/1519] eta 0:18:15 lr 0.000028 time 0.9294 (1.0057) model_time 0.9293 (1.0043) loss 0.7025 (0.9068) grad_norm 8.4697 (8.8057/2.0323) mem 68106MB [2022-12-19 16:13:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][440/1519] eta 0:18:04 lr 0.000028 time 0.9326 (1.0055) model_time 0.9325 (1.0042) loss 1.0643 (0.9094) grad_norm 9.2378 (8.8312/2.0437) mem 68106MB [2022-12-19 16:14:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][450/1519] eta 0:17:54 lr 0.000028 time 0.9301 (1.0053) model_time 0.9299 (1.0041) loss 1.1896 (0.9113) grad_norm 7.8578 (8.8153/2.0249) mem 68106MB [2022-12-19 16:14:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][460/1519] eta 0:17:44 lr 0.000028 time 0.9296 (1.0053) model_time 0.9294 (1.0041) loss 0.7799 (0.9106) grad_norm 8.4493 (8.8109/2.0053) mem 68106MB [2022-12-19 16:14:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][470/1519] eta 0:17:34 lr 0.000028 time 0.9225 (1.0052) model_time 0.9224 (1.0039) loss 1.1170 (0.9110) grad_norm 7.0683 (8.8164/1.9956) mem 68106MB [2022-12-19 16:14:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][480/1519] eta 0:17:24 lr 0.000028 time 0.9324 (1.0050) model_time 0.9323 (1.0038) loss 0.8478 (0.9101) grad_norm 9.1437 (8.7984/1.9813) mem 68106MB [2022-12-19 16:14:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][490/1519] eta 0:17:14 lr 0.000028 time 0.9209 (1.0049) model_time 0.9207 (1.0038) loss 0.7000 (0.9100) grad_norm 7.9130 (8.7869/1.9712) mem 68106MB [2022-12-19 16:14:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][500/1519] eta 0:17:04 lr 0.000028 time 0.9204 (1.0050) model_time 0.9202 (1.0038) loss 0.7767 (0.9094) grad_norm 7.7527 (8.7967/1.9724) mem 68106MB [2022-12-19 16:15:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][510/1519] eta 0:16:54 lr 0.000028 time 0.9304 (1.0050) model_time 0.9303 (1.0038) loss 0.7242 (0.9076) grad_norm 9.2718 (8.7865/1.9667) mem 68106MB [2022-12-19 16:15:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][520/1519] eta 0:16:44 lr 0.000028 time 0.8851 (1.0054) model_time 0.8850 (1.0042) loss 0.8452 (0.9061) grad_norm 13.7425 (8.7814/1.9885) mem 68106MB [2022-12-19 16:15:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][530/1519] eta 0:16:34 lr 0.000028 time 0.9308 (1.0058) model_time 0.9306 (1.0047) loss 0.8518 (0.9066) grad_norm 6.1533 (8.7719/1.9885) mem 68106MB [2022-12-19 16:15:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][540/1519] eta 0:16:24 lr 0.000028 time 0.9285 (1.0057) model_time 0.9284 (1.0046) loss 0.7091 (0.9062) grad_norm 7.7800 (8.7607/1.9772) mem 68106MB [2022-12-19 16:15:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][550/1519] eta 0:16:14 lr 0.000028 time 0.9330 (1.0056) model_time 0.9328 (1.0045) loss 1.0515 (0.9050) grad_norm 6.8474 (8.7520/1.9792) mem 68106MB [2022-12-19 16:15:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][560/1519] eta 0:16:04 lr 0.000028 time 0.9273 (1.0055) model_time 0.9271 (1.0044) loss 0.8047 (0.9046) grad_norm 7.4042 (8.7388/1.9846) mem 68106MB [2022-12-19 16:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][570/1519] eta 0:15:54 lr 0.000028 time 0.9879 (1.0056) model_time 0.9877 (1.0045) loss 0.8216 (0.9042) grad_norm 7.0965 (8.7212/1.9741) mem 68106MB [2022-12-19 16:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][580/1519] eta 0:15:44 lr 0.000028 time 0.9237 (1.0055) model_time 0.9235 (1.0044) loss 0.8722 (0.9045) grad_norm 8.9139 (8.7248/1.9624) mem 68106MB [2022-12-19 16:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][590/1519] eta 0:15:34 lr 0.000028 time 0.9382 (1.0055) model_time 0.9381 (1.0044) loss 0.9165 (0.9036) grad_norm 7.3968 (8.7247/1.9586) mem 68106MB [2022-12-19 16:16:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][600/1519] eta 0:15:23 lr 0.000028 time 0.9339 (1.0054) model_time 0.9337 (1.0044) loss 0.8420 (0.9037) grad_norm 10.1847 (8.7228/1.9455) mem 68106MB [2022-12-19 16:16:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][610/1519] eta 0:15:13 lr 0.000028 time 0.9286 (1.0054) model_time 0.9285 (1.0044) loss 0.9137 (0.9038) grad_norm 12.6498 (8.7230/1.9617) mem 68106MB [2022-12-19 16:16:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][620/1519] eta 0:15:03 lr 0.000028 time 0.9312 (1.0053) model_time 0.9310 (1.0043) loss 0.8411 (0.9046) grad_norm 8.2034 (8.7129/1.9699) mem 68106MB [2022-12-19 16:17:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][630/1519] eta 0:14:53 lr 0.000028 time 0.9309 (1.0053) model_time 0.9306 (1.0042) loss 0.9332 (0.9038) grad_norm 8.7966 (8.7189/1.9700) mem 68106MB [2022-12-19 16:17:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][640/1519] eta 0:14:43 lr 0.000028 time 0.9448 (1.0052) model_time 0.9446 (1.0042) loss 0.8394 (0.9026) grad_norm 6.0874 (8.6941/1.9536) mem 68106MB [2022-12-19 16:17:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][650/1519] eta 0:14:33 lr 0.000028 time 0.9579 (1.0052) model_time 0.9578 (1.0042) loss 0.9336 (0.9018) grad_norm 9.2647 (8.6943/1.9609) mem 68106MB [2022-12-19 16:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][660/1519] eta 0:14:23 lr 0.000028 time 0.9404 (1.0052) model_time 0.9403 (1.0042) loss 0.8095 (0.9011) grad_norm 8.0954 (8.7528/2.0784) mem 68106MB [2022-12-19 16:17:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][670/1519] eta 0:14:13 lr 0.000028 time 0.9291 (1.0052) model_time 0.9288 (1.0042) loss 1.0589 (0.9029) grad_norm 6.8954 (8.7287/2.0594) mem 68106MB [2022-12-19 16:17:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][680/1519] eta 0:14:03 lr 0.000028 time 1.1952 (1.0057) model_time 1.1951 (1.0047) loss 0.8521 (0.9034) grad_norm 12.9893 (8.7495/2.0882) mem 68106MB [2022-12-19 16:18:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][690/1519] eta 0:13:53 lr 0.000028 time 0.9294 (1.0056) model_time 0.9292 (1.0046) loss 0.7115 (0.9024) grad_norm 6.7012 (8.7104/2.0490) mem 68106MB [2022-12-19 16:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][700/1519] eta 0:13:43 lr 0.000028 time 0.9322 (1.0056) model_time 0.9320 (1.0047) loss 1.0069 (0.9037) grad_norm 7.8136 (8.7114/2.0556) mem 68106MB [2022-12-19 16:18:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][710/1519] eta 0:13:33 lr 0.000028 time 0.9353 (1.0056) model_time 0.9351 (1.0046) loss 1.0920 (0.9036) grad_norm 7.9627 (8.6845/1.9811) mem 68106MB [2022-12-19 16:18:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][720/1519] eta 0:13:23 lr 0.000028 time 0.9352 (1.0056) model_time 0.9350 (1.0047) loss 0.9396 (0.9029) grad_norm 8.8393 (8.6686/1.9724) mem 68106MB [2022-12-19 16:18:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][730/1519] eta 0:13:13 lr 0.000028 time 0.9350 (1.0055) model_time 0.9348 (1.0046) loss 0.9310 (0.9036) grad_norm 7.5598 (8.6680/1.9685) mem 68106MB [2022-12-19 16:18:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][740/1519] eta 0:13:03 lr 0.000028 time 0.9273 (1.0056) model_time 0.9271 (1.0046) loss 0.8960 (0.9037) grad_norm 5.6872 (8.6837/1.9720) mem 68106MB [2022-12-19 16:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][750/1519] eta 0:12:53 lr 0.000028 time 0.9310 (1.0055) model_time 0.9308 (1.0046) loss 1.1244 (0.9035) grad_norm 10.0263 (8.6832/1.9717) mem 68106MB [2022-12-19 16:19:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][760/1519] eta 0:12:43 lr 0.000028 time 0.9377 (1.0054) model_time 0.9374 (1.0045) loss 1.1049 (0.9030) grad_norm 12.5208 (8.6893/1.9817) mem 68106MB [2022-12-19 16:19:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][770/1519] eta 0:12:33 lr 0.000028 time 0.9830 (1.0054) model_time 0.9829 (1.0045) loss 1.0220 (0.9037) grad_norm 13.4677 (8.6948/1.9917) mem 68106MB [2022-12-19 16:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][780/1519] eta 0:12:22 lr 0.000028 time 0.9318 (1.0053) model_time 0.9315 (1.0044) loss 0.7678 (0.9033) grad_norm 7.2209 (8.7197/2.0136) mem 68106MB [2022-12-19 16:19:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][790/1519] eta 0:12:12 lr 0.000028 time 0.9328 (1.0052) model_time 0.9326 (1.0043) loss 0.9069 (0.9034) grad_norm 8.1918 (8.7295/2.0086) mem 68106MB [2022-12-19 16:19:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][800/1519] eta 0:12:02 lr 0.000028 time 0.9386 (1.0052) model_time 0.9384 (1.0043) loss 1.0082 (0.9032) grad_norm 14.7665 (8.7670/2.0768) mem 68106MB [2022-12-19 16:20:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][810/1519] eta 0:11:52 lr 0.000028 time 0.9416 (1.0055) model_time 0.9415 (1.0046) loss 0.7070 (0.9027) grad_norm 6.8571 (8.7854/2.0667) mem 68106MB [2022-12-19 16:20:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][820/1519] eta 0:11:42 lr 0.000028 time 0.9163 (1.0055) model_time 0.9161 (1.0046) loss 0.9102 (0.9033) grad_norm 7.7928 (8.7743/2.0710) mem 68106MB [2022-12-19 16:20:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][830/1519] eta 0:11:33 lr 0.000028 time 0.9494 (1.0058) model_time 0.9492 (1.0050) loss 0.8841 (0.9035) grad_norm 7.3496 (8.7688/2.0814) mem 68106MB [2022-12-19 16:20:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][840/1519] eta 0:11:23 lr 0.000028 time 0.9347 (1.0059) model_time 0.9345 (1.0051) loss 0.9086 (0.9034) grad_norm 8.8406 (8.7411/2.0405) mem 68106MB [2022-12-19 16:20:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][850/1519] eta 0:11:12 lr 0.000028 time 0.9324 (1.0060) model_time 0.9323 (1.0051) loss 1.2643 (0.9045) grad_norm 9.6651 (8.6977/2.0221) mem 68106MB [2022-12-19 16:20:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][860/1519] eta 0:11:02 lr 0.000028 time 0.9273 (1.0059) model_time 0.9271 (1.0051) loss 1.1605 (0.9054) grad_norm 7.3538 (8.6750/2.0093) mem 68106MB [2022-12-19 16:21:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][870/1519] eta 0:10:52 lr 0.000028 time 0.9373 (1.0059) model_time 0.9372 (1.0051) loss 1.1125 (0.9067) grad_norm 6.6182 (8.6701/2.0102) mem 68106MB [2022-12-19 16:21:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][880/1519] eta 0:10:42 lr 0.000028 time 0.9707 (1.0059) model_time 0.9705 (1.0051) loss 0.9325 (0.9066) grad_norm 9.1329 (8.7120/2.0074) mem 68106MB [2022-12-19 16:21:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][890/1519] eta 0:10:32 lr 0.000028 time 0.9307 (1.0059) model_time 0.9306 (1.0051) loss 0.7163 (0.9061) grad_norm 8.5436 (8.7038/1.9695) mem 68106MB [2022-12-19 16:21:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][900/1519] eta 0:10:22 lr 0.000028 time 0.9339 (1.0058) model_time 0.9338 (1.0050) loss 0.8150 (0.9057) grad_norm 8.2683 (8.6801/1.9517) mem 68106MB [2022-12-19 16:21:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][910/1519] eta 0:10:12 lr 0.000028 time 0.9290 (1.0058) model_time 0.9288 (1.0049) loss 0.9859 (0.9063) grad_norm 12.3414 (8.6763/1.9631) mem 68106MB [2022-12-19 16:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][920/1519] eta 0:10:02 lr 0.000028 time 1.0103 (1.0058) model_time 1.0102 (1.0050) loss 0.8558 (0.9054) grad_norm 10.0122 (8.6967/1.9564) mem 68106MB [2022-12-19 16:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][930/1519] eta 0:09:52 lr 0.000028 time 0.9324 (1.0057) model_time 0.9323 (1.0049) loss 0.8967 (0.9059) grad_norm 17.3398 (8.7363/2.0142) mem 68106MB [2022-12-19 16:22:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][940/1519] eta 0:09:42 lr 0.000028 time 0.9352 (1.0056) model_time 0.9350 (1.0048) loss 0.8004 (0.9061) grad_norm 7.8124 (8.7259/2.0286) mem 68106MB [2022-12-19 16:22:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][950/1519] eta 0:09:32 lr 0.000028 time 0.9380 (1.0056) model_time 0.9378 (1.0048) loss 0.9824 (0.9064) grad_norm 7.1383 (8.7145/2.0324) mem 68106MB [2022-12-19 16:22:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][960/1519] eta 0:09:22 lr 0.000028 time 0.9323 (1.0055) model_time 0.9321 (1.0047) loss 0.8581 (0.9066) grad_norm 7.8777 (8.6870/2.0202) mem 68106MB [2022-12-19 16:22:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][970/1519] eta 0:09:12 lr 0.000028 time 0.9320 (1.0055) model_time 0.9319 (1.0047) loss 0.7878 (0.9064) grad_norm 9.6274 (8.6710/2.0274) mem 68106MB [2022-12-19 16:22:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][980/1519] eta 0:09:01 lr 0.000028 time 0.9828 (1.0055) model_time 0.9827 (1.0048) loss 0.8615 (0.9061) grad_norm 9.9356 (8.6746/2.0286) mem 68106MB [2022-12-19 16:23:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][990/1519] eta 0:08:51 lr 0.000028 time 0.9419 (1.0056) model_time 0.9418 (1.0048) loss 1.1880 (0.9061) grad_norm 11.2564 (8.6670/2.0392) mem 68106MB [2022-12-19 16:23:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1000/1519] eta 0:08:41 lr 0.000028 time 0.9376 (1.0057) model_time 0.9374 (1.0049) loss 0.7555 (0.9057) grad_norm 9.1823 (8.6687/2.0388) mem 68106MB [2022-12-19 16:23:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1010/1519] eta 0:08:31 lr 0.000028 time 0.9171 (1.0057) model_time 0.9169 (1.0049) loss 0.6748 (0.9050) grad_norm 7.1001 (8.6682/2.0253) mem 68106MB [2022-12-19 16:23:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1020/1519] eta 0:08:21 lr 0.000028 time 0.9360 (1.0057) model_time 0.9358 (1.0049) loss 0.8313 (0.9046) grad_norm 8.5494 (8.6773/2.0186) mem 68106MB [2022-12-19 16:23:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1030/1519] eta 0:08:11 lr 0.000028 time 1.0021 (1.0057) model_time 1.0019 (1.0049) loss 1.1887 (0.9053) grad_norm 7.6325 (8.6392/2.0080) mem 68106MB [2022-12-19 16:23:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1040/1519] eta 0:08:01 lr 0.000028 time 0.9326 (1.0056) model_time 0.9324 (1.0049) loss 0.8796 (0.9055) grad_norm 7.8811 (8.5986/1.9918) mem 68106MB [2022-12-19 16:24:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1050/1519] eta 0:07:51 lr 0.000028 time 0.9289 (1.0056) model_time 0.9288 (1.0048) loss 0.7299 (0.9053) grad_norm 11.3464 (8.6101/2.0003) mem 68106MB [2022-12-19 16:24:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1060/1519] eta 0:07:41 lr 0.000028 time 0.9248 (1.0055) model_time 0.9246 (1.0048) loss 0.7867 (0.9045) grad_norm 6.5416 (8.6198/2.0283) mem 68106MB [2022-12-19 16:24:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1070/1519] eta 0:07:31 lr 0.000028 time 0.9322 (1.0055) model_time 0.9320 (1.0047) loss 1.0490 (0.9040) grad_norm 9.0509 (8.6357/2.0661) mem 68106MB [2022-12-19 16:24:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1080/1519] eta 0:07:21 lr 0.000028 time 0.9275 (1.0054) model_time 0.9273 (1.0047) loss 0.7017 (0.9036) grad_norm 10.9789 (8.6459/2.0727) mem 68106MB [2022-12-19 16:24:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1090/1519] eta 0:07:11 lr 0.000028 time 0.9320 (1.0054) model_time 0.9319 (1.0047) loss 0.7465 (0.9036) grad_norm 7.8782 (8.6704/2.0898) mem 68106MB [2022-12-19 16:24:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1100/1519] eta 0:07:01 lr 0.000028 time 1.0302 (1.0055) model_time 1.0300 (1.0047) loss 0.9205 (0.9036) grad_norm 8.1315 (8.6850/2.0872) mem 68106MB [2022-12-19 16:25:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1110/1519] eta 0:06:51 lr 0.000028 time 0.9316 (1.0055) model_time 0.9314 (1.0047) loss 1.0070 (0.9042) grad_norm 10.5962 (8.6859/2.0836) mem 68106MB [2022-12-19 16:25:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1120/1519] eta 0:06:41 lr 0.000028 time 1.0127 (1.0055) model_time 1.0120 (1.0048) loss 1.1195 (0.9043) grad_norm 7.4616 (8.6845/2.0546) mem 68106MB [2022-12-19 16:25:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1130/1519] eta 0:06:31 lr 0.000028 time 0.9310 (1.0055) model_time 0.9309 (1.0047) loss 0.9172 (0.9045) grad_norm 8.3299 (8.6929/2.0545) mem 68106MB [2022-12-19 16:25:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1140/1519] eta 0:06:21 lr 0.000028 time 0.9332 (1.0055) model_time 0.9330 (1.0048) loss 1.0008 (0.9045) grad_norm 8.6215 (8.6851/2.0551) mem 68106MB [2022-12-19 16:25:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1150/1519] eta 0:06:11 lr 0.000028 time 0.9364 (1.0056) model_time 0.9362 (1.0048) loss 0.9471 (0.9048) grad_norm 10.5915 (8.6810/2.0489) mem 68106MB [2022-12-19 16:25:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1160/1519] eta 0:06:01 lr 0.000028 time 0.9319 (1.0058) model_time 0.9318 (1.0051) loss 0.8090 (0.9049) grad_norm 10.8730 (8.6950/2.0442) mem 68106MB [2022-12-19 16:26:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1170/1519] eta 0:05:51 lr 0.000028 time 0.9309 (1.0057) model_time 0.9306 (1.0050) loss 0.9891 (0.9054) grad_norm 12.0333 (8.7287/2.0504) mem 68106MB [2022-12-19 16:26:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1180/1519] eta 0:05:40 lr 0.000028 time 0.9397 (1.0058) model_time 0.9396 (1.0051) loss 1.4363 (0.9055) grad_norm 10.5582 (8.7200/2.0563) mem 68106MB [2022-12-19 16:26:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1190/1519] eta 0:05:30 lr 0.000028 time 0.9187 (1.0058) model_time 0.9185 (1.0051) loss 0.7601 (0.9055) grad_norm 10.1348 (8.7235/2.0618) mem 68106MB [2022-12-19 16:26:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1200/1519] eta 0:05:20 lr 0.000028 time 0.9300 (1.0058) model_time 0.9299 (1.0051) loss 0.7761 (0.9055) grad_norm 7.4942 (8.7301/2.0638) mem 68106MB [2022-12-19 16:26:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1210/1519] eta 0:05:10 lr 0.000028 time 0.9263 (1.0058) model_time 0.9262 (1.0051) loss 1.0591 (0.9057) grad_norm 7.9703 (8.7189/2.0483) mem 68106MB [2022-12-19 16:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1220/1519] eta 0:05:00 lr 0.000028 time 0.9285 (1.0057) model_time 0.9283 (1.0050) loss 0.7926 (0.9064) grad_norm 9.7083 (8.7416/2.0460) mem 68106MB [2022-12-19 16:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1230/1519] eta 0:04:50 lr 0.000028 time 0.9294 (1.0057) model_time 0.9290 (1.0050) loss 0.7482 (0.9058) grad_norm 9.8439 (8.7369/2.0564) mem 68106MB [2022-12-19 16:27:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1240/1519] eta 0:04:40 lr 0.000028 time 0.9214 (1.0056) model_time 0.9213 (1.0049) loss 1.0354 (0.9062) grad_norm 9.4231 (8.7345/2.0534) mem 68106MB [2022-12-19 16:27:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1250/1519] eta 0:04:30 lr 0.000028 time 0.9267 (1.0056) model_time 0.9265 (1.0049) loss 0.9390 (0.9066) grad_norm 9.4085 (8.7346/2.0585) mem 68106MB [2022-12-19 16:27:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1260/1519] eta 0:04:20 lr 0.000028 time 0.9252 (1.0055) model_time 0.9251 (1.0048) loss 0.8195 (0.9066) grad_norm 10.3173 (8.7042/1.9466) mem 68106MB [2022-12-19 16:27:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1270/1519] eta 0:04:10 lr 0.000028 time 0.9255 (1.0056) model_time 0.9254 (1.0049) loss 1.0200 (0.9067) grad_norm 7.7707 (8.6892/1.9271) mem 68106MB [2022-12-19 16:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1280/1519] eta 0:04:00 lr 0.000028 time 0.9249 (1.0055) model_time 0.9247 (1.0048) loss 1.0090 (0.9068) grad_norm 7.0947 (8.6641/1.8884) mem 68106MB [2022-12-19 16:28:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1290/1519] eta 0:03:50 lr 0.000028 time 0.9291 (1.0055) model_time 0.9290 (1.0048) loss 0.9062 (0.9071) grad_norm 7.9738 (8.6838/1.8900) mem 68106MB [2022-12-19 16:28:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1300/1519] eta 0:03:40 lr 0.000028 time 1.2308 (1.0058) model_time 1.2307 (1.0051) loss 0.7663 (0.9065) grad_norm 6.9317 (8.6456/1.9016) mem 68106MB [2022-12-19 16:28:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1310/1519] eta 0:03:30 lr 0.000028 time 0.9309 (1.0057) model_time 0.9307 (1.0051) loss 1.0971 (0.9065) grad_norm 5.9831 (8.6299/1.9098) mem 68106MB [2022-12-19 16:28:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1320/1519] eta 0:03:20 lr 0.000028 time 0.9271 (1.0059) model_time 0.9269 (1.0052) loss 1.1148 (0.9069) grad_norm 6.9792 (8.6418/1.9092) mem 68106MB [2022-12-19 16:28:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1330/1519] eta 0:03:10 lr 0.000028 time 0.9278 (1.0061) model_time 0.9277 (1.0054) loss 0.8228 (0.9069) grad_norm 9.7616 (8.6625/1.9056) mem 68106MB [2022-12-19 16:28:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1340/1519] eta 0:03:00 lr 0.000028 time 0.9273 (1.0060) model_time 0.9272 (1.0053) loss 0.7624 (0.9073) grad_norm 7.8884 (8.6499/1.8904) mem 68106MB [2022-12-19 16:29:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1350/1519] eta 0:02:50 lr 0.000028 time 0.9284 (1.0061) model_time 0.9282 (1.0054) loss 1.1652 (0.9071) grad_norm 8.3440 (8.6427/1.8956) mem 68106MB [2022-12-19 16:29:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1360/1519] eta 0:02:39 lr 0.000028 time 0.9835 (1.0060) model_time 0.9834 (1.0054) loss 0.7016 (0.9075) grad_norm 6.0272 (8.6511/1.8951) mem 68106MB [2022-12-19 16:29:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1370/1519] eta 0:02:29 lr 0.000028 time 0.9417 (1.0060) model_time 0.9416 (1.0053) loss 1.0766 (0.9078) grad_norm 8.6478 (8.6745/2.0037) mem 68106MB [2022-12-19 16:29:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1380/1519] eta 0:02:19 lr 0.000028 time 0.9322 (1.0060) model_time 0.9316 (1.0053) loss 0.7129 (0.9080) grad_norm 8.8227 (8.6512/1.9857) mem 68106MB [2022-12-19 16:29:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1390/1519] eta 0:02:09 lr 0.000028 time 0.9217 (1.0059) model_time 0.9216 (1.0053) loss 0.9316 (0.9076) grad_norm 7.4087 (8.6434/1.9896) mem 68106MB [2022-12-19 16:29:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1400/1519] eta 0:01:59 lr 0.000028 time 0.9182 (1.0059) model_time 0.9181 (1.0052) loss 0.9161 (0.9076) grad_norm 7.2054 (8.5929/1.9012) mem 68106MB [2022-12-19 16:30:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1410/1519] eta 0:01:49 lr 0.000028 time 0.9299 (1.0059) model_time 0.9298 (1.0052) loss 0.7659 (0.9077) grad_norm 10.5039 (8.6243/1.9167) mem 68106MB [2022-12-19 16:30:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1420/1519] eta 0:01:39 lr 0.000028 time 0.9453 (1.0059) model_time 0.9452 (1.0052) loss 0.8773 (0.9072) grad_norm 8.5678 (8.6412/1.9162) mem 68106MB [2022-12-19 16:30:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1430/1519] eta 0:01:29 lr 0.000028 time 0.9118 (1.0059) model_time 0.9116 (1.0052) loss 0.6990 (0.9068) grad_norm 10.2103 (8.6639/1.9089) mem 68106MB [2022-12-19 16:30:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1440/1519] eta 0:01:19 lr 0.000028 time 0.9148 (1.0058) model_time 0.9146 (1.0052) loss 0.9981 (0.9066) grad_norm 8.7469 (8.6487/1.9025) mem 68106MB [2022-12-19 16:30:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1450/1519] eta 0:01:09 lr 0.000028 time 0.9226 (1.0060) model_time 0.9224 (1.0053) loss 0.9849 (0.9075) grad_norm 6.0117 (8.6390/1.9017) mem 68106MB [2022-12-19 16:30:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1460/1519] eta 0:00:59 lr 0.000028 time 0.9211 (1.0059) model_time 0.9210 (1.0053) loss 0.9563 (0.9074) grad_norm 6.3557 (8.6383/1.8944) mem 68106MB [2022-12-19 16:31:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1470/1519] eta 0:00:49 lr 0.000028 time 0.9278 (1.0060) model_time 0.9277 (1.0053) loss 1.1516 (0.9077) grad_norm 12.6686 (8.6904/1.9386) mem 68106MB [2022-12-19 16:31:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1480/1519] eta 0:00:39 lr 0.000028 time 0.9194 (1.0060) model_time 0.9192 (1.0054) loss 0.9883 (0.9078) grad_norm 7.5259 (8.6878/1.9379) mem 68106MB [2022-12-19 16:31:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1490/1519] eta 0:00:29 lr 0.000028 time 0.9224 (1.0061) model_time 0.9223 (1.0055) loss 0.9487 (0.9073) grad_norm 6.5342 (8.6813/1.9470) mem 68106MB [2022-12-19 16:31:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1500/1519] eta 0:00:19 lr 0.000028 time 0.9283 (1.0061) model_time 0.9281 (1.0054) loss 0.7243 (0.9073) grad_norm 8.7692 (8.6841/1.9650) mem 68106MB [2022-12-19 16:31:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [27/100][1510/1519] eta 0:00:09 lr 0.000028 time 0.9165 (1.0060) model_time 0.9164 (1.0054) loss 0.9896 (0.9082) grad_norm 13.9891 (8.6932/1.9756) mem 68106MB [2022-12-19 16:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 27 training takes 0:25:28 [2022-12-19 16:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_27.pth saving...... [2022-12-19 16:32:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_27.pth saved !!! [2022-12-19 16:32:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.681 (0.681) Loss 0.5107 (0.5107) Acc@1 91.319 (91.319) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 16:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.334) Loss 0.5059 (0.4993) Acc@1 92.708 (91.635) Acc@5 98.264 (98.390) Mem 68106MB [2022-12-19 16:32:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.316) Loss 0.4602 (0.4982) Acc@1 92.361 (91.336) Acc@5 99.306 (98.380) Mem 68106MB [2022-12-19 16:32:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 0.5975 (0.5052) Acc@1 89.236 (91.107) Acc@5 97.569 (98.253) Mem 68106MB [2022-12-19 16:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.308) Loss 0.4701 (0.4982) Acc@1 90.972 (91.201) Acc@5 99.306 (98.323) Mem 68106MB [2022-12-19 16:32:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.306) Loss 0.5258 (0.4952) Acc@1 87.500 (91.122) Acc@5 99.653 (98.393) Mem 68106MB [2022-12-19 16:32:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 0.5098 (0.4959) Acc@1 90.278 (91.109) Acc@5 98.264 (98.401) Mem 68106MB [2022-12-19 16:32:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5567 (0.4978) Acc@1 91.667 (91.060) Acc@5 98.611 (98.396) Mem 68106MB [2022-12-19 16:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.298 (0.303) Loss 0.4365 (0.4965) Acc@1 92.014 (91.058) Acc@5 98.611 (98.410) Mem 68106MB [2022-12-19 16:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:27] * Acc@1 91.069 Acc@5 98.408 [2022-12-19 16:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.1% [2022-12-19 16:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.07% [2022-12-19 16:32:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][0/1519] eta 0:44:25 lr 0.000028 time 1.7550 (1.7550) model_time 1.0503 (1.0503) loss 0.7935 (0.7935) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 16:33:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][10/1519] eta 0:26:57 lr 0.000028 time 0.9342 (1.0718) model_time 0.9340 (1.0074) loss 0.7175 (0.9123) grad_norm 18.2847 (10.2616/4.0846) mem 68106MB [2022-12-19 16:33:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][20/1519] eta 0:25:54 lr 0.000028 time 0.9289 (1.0372) model_time 0.9288 (1.0033) loss 0.7597 (0.9167) grad_norm 9.6210 (10.4233/3.2812) mem 68106MB [2022-12-19 16:33:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][30/1519] eta 0:25:25 lr 0.000028 time 0.9240 (1.0247) model_time 0.9239 (1.0016) loss 0.7613 (0.9163) grad_norm 6.3605 (9.7496/3.0404) mem 68106MB [2022-12-19 16:33:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][40/1519] eta 0:25:09 lr 0.000028 time 0.9248 (1.0203) model_time 0.9247 (1.0027) loss 1.0856 (0.9483) grad_norm 9.2330 (9.5482/2.6634) mem 68106MB [2022-12-19 16:33:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][50/1519] eta 0:24:52 lr 0.000028 time 0.9294 (1.0160) model_time 0.9292 (1.0018) loss 1.0993 (0.9335) grad_norm 6.3988 (9.3339/2.6463) mem 68106MB [2022-12-19 16:33:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][60/1519] eta 0:24:40 lr 0.000028 time 0.9565 (1.0146) model_time 0.9563 (1.0027) loss 1.0273 (0.9311) grad_norm 12.8542 (9.4280/2.5711) mem 68106MB [2022-12-19 16:34:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][70/1519] eta 0:24:26 lr 0.000028 time 0.9185 (1.0123) model_time 0.9182 (1.0020) loss 1.1049 (0.9202) grad_norm 6.6416 (9.5444/2.6885) mem 68106MB [2022-12-19 16:34:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][80/1519] eta 0:24:14 lr 0.000028 time 0.9247 (1.0106) model_time 0.9231 (1.0015) loss 0.9161 (0.9123) grad_norm 7.1257 (9.5057/2.6153) mem 68106MB [2022-12-19 16:34:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][90/1519] eta 0:24:02 lr 0.000028 time 0.9374 (1.0093) model_time 0.9371 (1.0012) loss 1.1247 (0.9187) grad_norm 10.0797 (9.2587/2.6094) mem 68106MB [2022-12-19 16:34:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][100/1519] eta 0:23:51 lr 0.000028 time 0.9281 (1.0090) model_time 0.9279 (1.0016) loss 0.7192 (0.9155) grad_norm 7.4141 (9.0639/2.5465) mem 68106MB [2022-12-19 16:34:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][110/1519] eta 0:23:41 lr 0.000028 time 0.9662 (1.0086) model_time 0.9661 (1.0018) loss 0.7740 (0.9220) grad_norm 8.8458 (9.0273/2.4704) mem 68106MB [2022-12-19 16:34:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][120/1519] eta 0:23:31 lr 0.000028 time 0.9339 (1.0092) model_time 0.9338 (1.0030) loss 0.9901 (0.9228) grad_norm 6.7667 (9.1351/2.5327) mem 68106MB [2022-12-19 16:35:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][130/1519] eta 0:23:21 lr 0.000028 time 0.9322 (1.0087) model_time 0.9320 (1.0029) loss 1.0874 (0.9282) grad_norm 7.6512 (9.0228/2.4712) mem 68106MB [2022-12-19 16:35:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][140/1519] eta 0:23:10 lr 0.000028 time 0.9232 (1.0083) model_time 0.9230 (1.0029) loss 1.1036 (0.9244) grad_norm 9.8128 (9.0107/2.3997) mem 68106MB [2022-12-19 16:35:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][150/1519] eta 0:23:00 lr 0.000028 time 1.0125 (1.0085) model_time 1.0123 (1.0034) loss 0.9273 (0.9199) grad_norm 7.3063 (8.9075/2.3662) mem 68106MB [2022-12-19 16:35:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][160/1519] eta 0:22:50 lr 0.000028 time 0.9232 (1.0087) model_time 0.9230 (1.0040) loss 1.0906 (0.9203) grad_norm 5.7210 (8.8280/2.3431) mem 68106MB [2022-12-19 16:35:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][170/1519] eta 0:22:40 lr 0.000028 time 0.9389 (1.0088) model_time 0.9387 (1.0043) loss 0.7582 (0.9227) grad_norm 7.2863 (8.7204/2.3156) mem 68106MB [2022-12-19 16:35:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][180/1519] eta 0:22:30 lr 0.000028 time 0.9286 (1.0086) model_time 0.9284 (1.0043) loss 0.8840 (0.9167) grad_norm 9.4297 (8.7900/2.3323) mem 68106MB [2022-12-19 16:36:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][190/1519] eta 0:22:20 lr 0.000028 time 0.9364 (1.0087) model_time 0.9362 (1.0046) loss 0.8562 (0.9171) grad_norm 13.4631 (8.8591/2.3677) mem 68106MB [2022-12-19 16:36:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][200/1519] eta 0:22:09 lr 0.000028 time 0.9269 (1.0082) model_time 0.9268 (1.0043) loss 0.7940 (0.9174) grad_norm 7.7126 (8.8737/2.3186) mem 68106MB [2022-12-19 16:36:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][210/1519] eta 0:21:59 lr 0.000028 time 0.9273 (1.0077) model_time 0.9271 (1.0040) loss 0.7810 (0.9193) grad_norm 6.4421 (8.8325/2.3013) mem 68106MB [2022-12-19 16:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][220/1519] eta 0:21:48 lr 0.000028 time 0.9318 (1.0075) model_time 0.9316 (1.0040) loss 0.7137 (0.9203) grad_norm 13.4561 (8.8679/2.3031) mem 68106MB [2022-12-19 16:36:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][230/1519] eta 0:21:38 lr 0.000028 time 0.9337 (1.0078) model_time 0.9335 (1.0043) loss 0.8064 (0.9192) grad_norm 10.1666 (8.8397/2.2755) mem 68106MB [2022-12-19 16:36:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][240/1519] eta 0:21:29 lr 0.000028 time 0.9312 (1.0081) model_time 0.9310 (1.0048) loss 0.8299 (0.9166) grad_norm 9.8490 (8.8152/2.2522) mem 68106MB [2022-12-19 16:37:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][250/1519] eta 0:21:18 lr 0.000028 time 0.9275 (1.0079) model_time 0.9273 (1.0047) loss 0.6969 (0.9133) grad_norm 6.2802 (8.8256/2.2479) mem 68106MB [2022-12-19 16:37:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][260/1519] eta 0:21:08 lr 0.000028 time 0.9290 (1.0078) model_time 0.9288 (1.0047) loss 0.9425 (0.9139) grad_norm 8.0855 (8.8286/2.2167) mem 68106MB [2022-12-19 16:37:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][270/1519] eta 0:20:58 lr 0.000028 time 0.9282 (1.0076) model_time 0.9281 (1.0047) loss 0.9775 (0.9135) grad_norm 8.8197 (8.8393/2.2031) mem 68106MB [2022-12-19 16:37:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][280/1519] eta 0:20:49 lr 0.000028 time 0.9316 (1.0081) model_time 0.9314 (1.0052) loss 1.1082 (0.9118) grad_norm 8.6176 (8.8223/2.1829) mem 68106MB [2022-12-19 16:37:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][290/1519] eta 0:20:38 lr 0.000028 time 0.9783 (1.0080) model_time 0.9781 (1.0052) loss 0.6900 (0.9101) grad_norm 6.3920 (8.8026/2.1754) mem 68106MB [2022-12-19 16:37:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][300/1519] eta 0:20:29 lr 0.000028 time 0.9336 (1.0084) model_time 0.9334 (1.0056) loss 0.8551 (0.9117) grad_norm 12.9500 (8.8681/2.2059) mem 68106MB [2022-12-19 16:38:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][310/1519] eta 0:20:19 lr 0.000028 time 0.9346 (1.0083) model_time 0.9344 (1.0057) loss 0.8336 (0.9143) grad_norm 8.0668 (8.8283/2.1829) mem 68106MB [2022-12-19 16:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][320/1519] eta 0:20:09 lr 0.000028 time 0.9266 (1.0085) model_time 0.9265 (1.0059) loss 0.8119 (0.9138) grad_norm 8.2341 (8.8079/2.1593) mem 68106MB [2022-12-19 16:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][330/1519] eta 0:19:58 lr 0.000028 time 0.9229 (1.0083) model_time 0.9228 (1.0057) loss 0.7325 (0.9122) grad_norm 8.2754 (8.7954/2.1465) mem 68106MB [2022-12-19 16:38:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][340/1519] eta 0:19:48 lr 0.000028 time 0.9218 (1.0080) model_time 0.9216 (1.0055) loss 0.9804 (0.9131) grad_norm 7.3863 (8.7760/2.1265) mem 68106MB [2022-12-19 16:38:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][350/1519] eta 0:19:38 lr 0.000028 time 0.9215 (1.0079) model_time 0.9214 (1.0055) loss 0.9050 (0.9146) grad_norm 7.0130 (8.7675/2.1092) mem 68106MB [2022-12-19 16:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][360/1519] eta 0:19:27 lr 0.000028 time 0.9354 (1.0077) model_time 0.9353 (1.0054) loss 1.0453 (0.9126) grad_norm 7.9831 (8.7607/2.0933) mem 68106MB [2022-12-19 16:39:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][370/1519] eta 0:19:17 lr 0.000028 time 0.9263 (1.0077) model_time 0.9261 (1.0054) loss 1.1131 (0.9127) grad_norm 8.7730 (8.7611/2.0717) mem 68106MB [2022-12-19 16:39:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][380/1519] eta 0:19:07 lr 0.000028 time 0.9362 (1.0075) model_time 0.9361 (1.0053) loss 0.8493 (0.9118) grad_norm 6.5350 (8.7247/2.0598) mem 68106MB [2022-12-19 16:39:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][390/1519] eta 0:18:57 lr 0.000028 time 0.9319 (1.0073) model_time 0.9317 (1.0051) loss 0.8448 (0.9107) grad_norm 10.3340 (8.7335/2.0646) mem 68106MB [2022-12-19 16:39:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][400/1519] eta 0:18:47 lr 0.000028 time 0.9304 (1.0072) model_time 0.9303 (1.0050) loss 0.7629 (0.9110) grad_norm 7.2322 (8.7361/2.0450) mem 68106MB [2022-12-19 16:39:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][410/1519] eta 0:18:36 lr 0.000028 time 0.9243 (1.0070) model_time 0.9241 (1.0049) loss 1.1873 (0.9100) grad_norm 8.9695 (8.7150/2.0300) mem 68106MB [2022-12-19 16:39:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][420/1519] eta 0:18:26 lr 0.000028 time 0.9275 (1.0068) model_time 0.9273 (1.0048) loss 0.6922 (0.9088) grad_norm 8.3857 (8.7499/2.0502) mem 68106MB [2022-12-19 16:40:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][430/1519] eta 0:18:16 lr 0.000028 time 0.9215 (1.0069) model_time 0.9214 (1.0049) loss 0.7109 (0.9112) grad_norm 12.8509 (8.7677/2.0487) mem 68106MB [2022-12-19 16:40:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][440/1519] eta 0:18:06 lr 0.000028 time 0.9244 (1.0068) model_time 0.9243 (1.0049) loss 0.6979 (0.9117) grad_norm 8.8032 (8.7672/2.0327) mem 68106MB [2022-12-19 16:40:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][450/1519] eta 0:17:56 lr 0.000028 time 0.9286 (1.0068) model_time 0.9284 (1.0048) loss 0.8871 (0.9123) grad_norm 6.9033 (8.7685/2.0174) mem 68106MB [2022-12-19 16:40:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][460/1519] eta 0:17:46 lr 0.000028 time 0.9229 (1.0068) model_time 0.9227 (1.0049) loss 0.9148 (0.9129) grad_norm 7.2716 (8.7669/2.0072) mem 68106MB [2022-12-19 16:40:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][470/1519] eta 0:17:36 lr 0.000028 time 0.9210 (1.0067) model_time 0.9209 (1.0049) loss 0.9647 (0.9125) grad_norm 8.9983 (8.7482/1.9999) mem 68106MB [2022-12-19 16:40:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][480/1519] eta 0:17:26 lr 0.000028 time 0.9336 (1.0069) model_time 0.9334 (1.0050) loss 0.7598 (0.9121) grad_norm 8.9797 (8.7283/1.9941) mem 68106MB [2022-12-19 16:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][490/1519] eta 0:17:16 lr 0.000028 time 0.9381 (1.0069) model_time 0.9378 (1.0051) loss 0.7976 (0.9121) grad_norm 9.2420 (8.7274/1.9806) mem 68106MB [2022-12-19 16:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][500/1519] eta 0:17:06 lr 0.000028 time 0.9284 (1.0069) model_time 0.9282 (1.0052) loss 1.0109 (0.9120) grad_norm 8.0271 (8.7317/1.9663) mem 68106MB [2022-12-19 16:41:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][510/1519] eta 0:16:55 lr 0.000028 time 0.9293 (1.0069) model_time 0.9292 (1.0051) loss 0.8472 (0.9138) grad_norm 7.5651 (8.7440/1.9563) mem 68106MB [2022-12-19 16:41:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][520/1519] eta 0:16:45 lr 0.000028 time 0.9335 (1.0067) model_time 0.9333 (1.0050) loss 0.8659 (0.9139) grad_norm 9.4316 (8.7659/1.9598) mem 68106MB [2022-12-19 16:41:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][530/1519] eta 0:16:35 lr 0.000028 time 0.9293 (1.0067) model_time 0.9291 (1.0050) loss 0.9346 (0.9142) grad_norm 10.3901 (8.7967/1.9626) mem 68106MB [2022-12-19 16:41:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][540/1519] eta 0:16:25 lr 0.000028 time 0.9318 (1.0068) model_time 0.9316 (1.0051) loss 1.0288 (0.9137) grad_norm 7.6628 (8.7628/1.9632) mem 68106MB [2022-12-19 16:42:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][550/1519] eta 0:16:15 lr 0.000028 time 0.9146 (1.0068) model_time 0.9144 (1.0051) loss 1.0501 (0.9128) grad_norm 9.1995 (8.7513/1.9524) mem 68106MB [2022-12-19 16:42:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][560/1519] eta 0:16:05 lr 0.000028 time 0.9266 (1.0066) model_time 0.9262 (1.0050) loss 0.7677 (0.9137) grad_norm 8.8005 (8.7504/1.9389) mem 68106MB [2022-12-19 16:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][570/1519] eta 0:15:55 lr 0.000028 time 0.9364 (1.0066) model_time 0.9362 (1.0050) loss 0.7147 (0.9126) grad_norm 8.7383 (8.7537/1.9438) mem 68106MB [2022-12-19 16:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][580/1519] eta 0:15:45 lr 0.000028 time 0.9324 (1.0065) model_time 0.9322 (1.0049) loss 0.8599 (0.9117) grad_norm 7.0689 (8.7524/1.9430) mem 68106MB [2022-12-19 16:42:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][590/1519] eta 0:15:35 lr 0.000028 time 0.9303 (1.0065) model_time 0.9301 (1.0049) loss 1.0654 (0.9136) grad_norm 5.9430 (8.7290/1.9379) mem 68106MB [2022-12-19 16:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][600/1519] eta 0:15:24 lr 0.000028 time 0.9291 (1.0064) model_time 0.9286 (1.0049) loss 0.7929 (0.9142) grad_norm 8.1197 (8.7447/1.9333) mem 68106MB [2022-12-19 16:43:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][610/1519] eta 0:15:14 lr 0.000028 time 0.9027 (1.0065) model_time 0.9024 (1.0050) loss 0.7637 (0.9134) grad_norm 6.6744 (8.7041/1.8610) mem 68106MB [2022-12-19 16:43:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][620/1519] eta 0:15:04 lr 0.000028 time 0.9433 (1.0064) model_time 0.9431 (1.0049) loss 0.8734 (0.9127) grad_norm 10.9154 (8.6771/1.8308) mem 68106MB [2022-12-19 16:43:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][630/1519] eta 0:14:54 lr 0.000028 time 0.9366 (1.0063) model_time 0.9364 (1.0048) loss 1.2237 (0.9129) grad_norm 8.8375 (8.6687/1.8194) mem 68106MB [2022-12-19 16:43:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][640/1519] eta 0:14:44 lr 0.000028 time 0.9397 (1.0062) model_time 0.9395 (1.0047) loss 1.0563 (0.9120) grad_norm 7.0775 (8.6587/1.8238) mem 68106MB [2022-12-19 16:43:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][650/1519] eta 0:14:34 lr 0.000028 time 0.9278 (1.0060) model_time 0.9272 (1.0046) loss 0.7082 (0.9111) grad_norm 8.5529 (8.6616/1.8105) mem 68106MB [2022-12-19 16:43:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][660/1519] eta 0:14:24 lr 0.000028 time 0.9272 (1.0060) model_time 0.9271 (1.0046) loss 1.0158 (0.9118) grad_norm 14.9281 (8.6604/1.8505) mem 68106MB [2022-12-19 16:44:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][670/1519] eta 0:14:14 lr 0.000028 time 0.9356 (1.0059) model_time 0.9354 (1.0045) loss 1.1505 (0.9122) grad_norm 8.5365 (8.6250/1.8047) mem 68106MB [2022-12-19 16:44:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][680/1519] eta 0:14:04 lr 0.000028 time 0.9317 (1.0060) model_time 0.9315 (1.0046) loss 0.7474 (0.9120) grad_norm 10.0000 (8.6114/1.8082) mem 68106MB [2022-12-19 16:44:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][690/1519] eta 0:13:53 lr 0.000028 time 0.9325 (1.0059) model_time 0.9322 (1.0045) loss 0.7515 (0.9118) grad_norm 10.3002 (8.6755/1.8235) mem 68106MB [2022-12-19 16:44:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][700/1519] eta 0:13:43 lr 0.000028 time 0.9297 (1.0058) model_time 0.9295 (1.0044) loss 0.9376 (0.9105) grad_norm 10.0740 (8.7144/1.8351) mem 68106MB [2022-12-19 16:44:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][710/1519] eta 0:13:33 lr 0.000028 time 0.9280 (1.0057) model_time 0.9278 (1.0044) loss 0.7400 (0.9098) grad_norm 10.9687 (8.7212/1.8457) mem 68106MB [2022-12-19 16:44:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][720/1519] eta 0:13:23 lr 0.000028 time 0.9286 (1.0058) model_time 0.9284 (1.0044) loss 0.8748 (0.9088) grad_norm 10.0946 (8.7031/1.8103) mem 68106MB [2022-12-19 16:45:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][730/1519] eta 0:13:13 lr 0.000027 time 0.9359 (1.0057) model_time 0.9357 (1.0043) loss 0.9781 (0.9086) grad_norm 7.6761 (8.7181/1.8161) mem 68106MB [2022-12-19 16:45:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][740/1519] eta 0:13:03 lr 0.000027 time 0.9823 (1.0057) model_time 0.9821 (1.0043) loss 1.0746 (0.9091) grad_norm 8.6917 (8.7442/1.8659) mem 68106MB [2022-12-19 16:45:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][750/1519] eta 0:12:53 lr 0.000027 time 0.9361 (1.0057) model_time 0.9359 (1.0044) loss 0.6875 (0.9081) grad_norm 8.7906 (8.7487/1.8618) mem 68106MB [2022-12-19 16:45:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][760/1519] eta 0:12:43 lr 0.000027 time 0.9138 (1.0057) model_time 0.9136 (1.0044) loss 0.9770 (0.9085) grad_norm 7.8113 (8.7487/1.8550) mem 68106MB [2022-12-19 16:45:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][770/1519] eta 0:12:33 lr 0.000027 time 0.9299 (1.0058) model_time 0.9297 (1.0045) loss 0.7184 (0.9081) grad_norm 10.5505 (8.7910/1.8538) mem 68106MB [2022-12-19 16:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][780/1519] eta 0:12:23 lr 0.000027 time 0.9360 (1.0057) model_time 0.9356 (1.0044) loss 1.3462 (0.9081) grad_norm 10.2416 (8.7942/1.8361) mem 68106MB [2022-12-19 16:46:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][790/1519] eta 0:12:13 lr 0.000027 time 0.9674 (1.0057) model_time 0.9672 (1.0045) loss 0.9137 (0.9075) grad_norm 8.0022 (8.8005/1.8409) mem 68106MB [2022-12-19 16:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][800/1519] eta 0:12:03 lr 0.000027 time 0.9005 (1.0061) model_time 0.9003 (1.0048) loss 0.7650 (0.9074) grad_norm 7.3363 (8.7769/1.8456) mem 68106MB [2022-12-19 16:46:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][810/1519] eta 0:11:53 lr 0.000027 time 0.9218 (1.0061) model_time 0.9217 (1.0049) loss 0.9854 (0.9075) grad_norm 8.1059 (8.7775/1.8328) mem 68106MB [2022-12-19 16:46:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][820/1519] eta 0:11:43 lr 0.000027 time 0.9296 (1.0060) model_time 0.9294 (1.0048) loss 0.9602 (0.9075) grad_norm 8.2268 (8.7381/1.8204) mem 68106MB [2022-12-19 16:46:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][830/1519] eta 0:11:33 lr 0.000027 time 0.9348 (1.0059) model_time 0.9347 (1.0047) loss 0.7894 (0.9067) grad_norm 10.1726 (8.7666/1.8514) mem 68106MB [2022-12-19 16:46:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][840/1519] eta 0:11:23 lr 0.000027 time 0.9124 (1.0061) model_time 0.9123 (1.0048) loss 0.8514 (0.9058) grad_norm 8.3616 (8.7550/1.8488) mem 68106MB [2022-12-19 16:47:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][850/1519] eta 0:11:12 lr 0.000027 time 0.9303 (1.0060) model_time 0.9302 (1.0048) loss 0.8262 (0.9060) grad_norm 9.7285 (8.7476/1.8301) mem 68106MB [2022-12-19 16:47:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][860/1519] eta 0:11:02 lr 0.000027 time 0.9350 (1.0059) model_time 0.9349 (1.0047) loss 0.7109 (0.9053) grad_norm 8.0079 (8.7587/1.8383) mem 68106MB [2022-12-19 16:47:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][870/1519] eta 0:10:52 lr 0.000027 time 0.9320 (1.0059) model_time 0.9319 (1.0048) loss 0.8286 (0.9050) grad_norm 17.5315 (8.7779/1.9124) mem 68106MB [2022-12-19 16:47:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][880/1519] eta 0:10:42 lr 0.000027 time 0.9294 (1.0059) model_time 0.9293 (1.0047) loss 0.7334 (0.9052) grad_norm 7.4804 (8.7963/1.9270) mem 68106MB [2022-12-19 16:47:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][890/1519] eta 0:10:32 lr 0.000027 time 0.9329 (1.0058) model_time 0.9328 (1.0047) loss 0.8457 (0.9054) grad_norm 9.6749 (8.7832/1.9239) mem 68106MB [2022-12-19 16:47:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][900/1519] eta 0:10:22 lr 0.000027 time 0.9994 (1.0059) model_time 0.9992 (1.0047) loss 0.6943 (0.9047) grad_norm 8.3377 (8.7469/1.8860) mem 68106MB [2022-12-19 16:48:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][910/1519] eta 0:10:12 lr 0.000027 time 0.9386 (1.0058) model_time 0.9385 (1.0047) loss 0.9952 (0.9053) grad_norm 9.7046 (8.7681/1.8857) mem 68106MB [2022-12-19 16:48:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][920/1519] eta 0:10:02 lr 0.000027 time 1.0013 (1.0058) model_time 1.0012 (1.0047) loss 0.9978 (0.9060) grad_norm 7.9917 (8.7851/1.8881) mem 68106MB [2022-12-19 16:48:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][930/1519] eta 0:09:52 lr 0.000027 time 0.9302 (1.0059) model_time 0.9301 (1.0047) loss 0.9405 (0.9070) grad_norm 8.0028 (8.7945/1.8941) mem 68106MB [2022-12-19 16:48:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][940/1519] eta 0:09:42 lr 0.000027 time 0.9309 (1.0058) model_time 0.9307 (1.0047) loss 0.7981 (0.9061) grad_norm 7.2219 (8.7984/1.8984) mem 68106MB [2022-12-19 16:48:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][950/1519] eta 0:09:32 lr 0.000027 time 0.9360 (1.0058) model_time 0.9358 (1.0046) loss 0.9935 (0.9058) grad_norm 7.3794 (8.7852/1.9008) mem 68106MB [2022-12-19 16:48:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][960/1519] eta 0:09:22 lr 0.000027 time 0.9319 (1.0057) model_time 0.9318 (1.0046) loss 0.8887 (0.9056) grad_norm 8.4854 (8.7658/1.9037) mem 68106MB [2022-12-19 16:49:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][970/1519] eta 0:09:12 lr 0.000027 time 0.9311 (1.0056) model_time 0.9310 (1.0045) loss 1.1207 (0.9058) grad_norm 7.1290 (8.7566/1.9032) mem 68106MB [2022-12-19 16:49:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][980/1519] eta 0:09:02 lr 0.000027 time 0.9343 (1.0056) model_time 0.9342 (1.0045) loss 0.7181 (0.9060) grad_norm 8.4606 (8.7625/1.8989) mem 68106MB [2022-12-19 16:49:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][990/1519] eta 0:08:51 lr 0.000027 time 0.9183 (1.0056) model_time 0.9181 (1.0045) loss 1.0684 (0.9062) grad_norm 7.3228 (8.7592/1.8964) mem 68106MB [2022-12-19 16:49:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1000/1519] eta 0:08:41 lr 0.000027 time 0.9389 (1.0056) model_time 0.9388 (1.0045) loss 1.0335 (0.9072) grad_norm 7.1592 (8.7524/1.8996) mem 68106MB [2022-12-19 16:49:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1010/1519] eta 0:08:31 lr 0.000027 time 0.9290 (1.0055) model_time 0.9289 (1.0044) loss 0.8933 (0.9070) grad_norm 8.0570 (8.7747/1.9297) mem 68106MB [2022-12-19 16:49:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1020/1519] eta 0:08:21 lr 0.000027 time 0.9277 (1.0055) model_time 0.9275 (1.0044) loss 0.8987 (0.9064) grad_norm 16.1766 (8.8012/2.0290) mem 68106MB [2022-12-19 16:50:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1030/1519] eta 0:08:11 lr 0.000027 time 0.9354 (1.0056) model_time 0.9353 (1.0045) loss 0.9410 (0.9062) grad_norm 8.5292 (8.8017/2.0263) mem 68106MB [2022-12-19 16:50:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1040/1519] eta 0:08:01 lr 0.000027 time 0.9316 (1.0055) model_time 0.9315 (1.0045) loss 1.1822 (0.9069) grad_norm 11.9773 (8.8122/2.0346) mem 68106MB [2022-12-19 16:50:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1050/1519] eta 0:07:51 lr 0.000027 time 0.9374 (1.0054) model_time 0.9372 (1.0044) loss 0.8277 (0.9069) grad_norm 5.3995 (8.8206/2.0624) mem 68106MB [2022-12-19 16:50:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1060/1519] eta 0:07:41 lr 0.000027 time 0.9374 (1.0055) model_time 0.9372 (1.0045) loss 0.7744 (0.9062) grad_norm 7.0712 (8.7965/2.0666) mem 68106MB [2022-12-19 16:50:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1070/1519] eta 0:07:31 lr 0.000027 time 0.9310 (1.0055) model_time 0.9308 (1.0045) loss 1.1530 (0.9061) grad_norm 7.9509 (8.8072/2.0670) mem 68106MB [2022-12-19 16:50:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1080/1519] eta 0:07:21 lr 0.000027 time 0.9412 (1.0056) model_time 0.9411 (1.0046) loss 1.3905 (0.9064) grad_norm 9.8269 (8.8177/2.0615) mem 68106MB [2022-12-19 16:51:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1090/1519] eta 0:07:11 lr 0.000027 time 0.9291 (1.0056) model_time 0.9290 (1.0046) loss 0.7692 (0.9064) grad_norm 6.6767 (8.8152/2.0619) mem 68106MB [2022-12-19 16:51:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1100/1519] eta 0:07:01 lr 0.000027 time 1.2069 (1.0058) model_time 1.2068 (1.0048) loss 0.9359 (0.9064) grad_norm 7.5389 (8.7935/2.0734) mem 68106MB [2022-12-19 16:51:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1110/1519] eta 0:06:51 lr 0.000027 time 0.9267 (1.0058) model_time 0.9266 (1.0048) loss 0.6958 (0.9074) grad_norm 8.6646 (8.7770/2.0684) mem 68106MB [2022-12-19 16:51:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1120/1519] eta 0:06:41 lr 0.000027 time 0.9329 (1.0058) model_time 0.9327 (1.0048) loss 0.8465 (0.9070) grad_norm 10.0176 (8.7964/2.0945) mem 68106MB [2022-12-19 16:51:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1130/1519] eta 0:06:31 lr 0.000027 time 0.9304 (1.0058) model_time 0.9303 (1.0048) loss 1.1525 (0.9068) grad_norm 11.5140 (8.7678/2.0855) mem 68106MB [2022-12-19 16:51:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1140/1519] eta 0:06:21 lr 0.000027 time 0.9330 (1.0057) model_time 0.9329 (1.0048) loss 0.7101 (0.9063) grad_norm 9.8322 (8.8025/2.0723) mem 68106MB [2022-12-19 16:52:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1150/1519] eta 0:06:11 lr 0.000027 time 0.9305 (1.0056) model_time 0.9303 (1.0046) loss 0.7110 (0.9063) grad_norm 5.4557 (8.8065/2.0775) mem 68106MB [2022-12-19 16:52:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1160/1519] eta 0:06:01 lr 0.000027 time 0.9309 (1.0056) model_time 0.9308 (1.0046) loss 1.1056 (0.9058) grad_norm 11.8026 (8.8090/2.0971) mem 68106MB [2022-12-19 16:52:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1170/1519] eta 0:05:50 lr 0.000027 time 0.9364 (1.0055) model_time 0.9362 (1.0046) loss 1.0564 (0.9063) grad_norm 7.8606 (8.8253/2.1406) mem 68106MB [2022-12-19 16:52:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1180/1519] eta 0:05:40 lr 0.000027 time 0.9288 (1.0055) model_time 0.9287 (1.0045) loss 0.8742 (0.9060) grad_norm 9.2990 (8.8127/2.1347) mem 68106MB [2022-12-19 16:52:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1190/1519] eta 0:05:30 lr 0.000027 time 0.9307 (1.0055) model_time 0.9306 (1.0045) loss 0.8881 (0.9062) grad_norm 6.2576 (8.8093/2.1426) mem 68106MB [2022-12-19 16:52:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1200/1519] eta 0:05:20 lr 0.000027 time 0.9276 (1.0054) model_time 0.9275 (1.0044) loss 0.8024 (0.9070) grad_norm 11.8935 (8.8197/2.1543) mem 68106MB [2022-12-19 16:53:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1210/1519] eta 0:05:10 lr 0.000027 time 0.9427 (1.0054) model_time 0.9426 (1.0045) loss 0.8874 (0.9080) grad_norm 7.3068 (8.8371/2.1551) mem 68106MB [2022-12-19 16:53:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1220/1519] eta 0:05:00 lr 0.000027 time 0.9534 (1.0054) model_time 0.9532 (1.0045) loss 0.7262 (0.9075) grad_norm 10.0627 (8.8510/2.1520) mem 68106MB [2022-12-19 16:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1230/1519] eta 0:04:50 lr 0.000027 time 0.9343 (1.0054) model_time 0.9342 (1.0045) loss 1.0589 (0.9082) grad_norm 6.3413 (8.8453/2.1544) mem 68106MB [2022-12-19 16:53:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1240/1519] eta 0:04:40 lr 0.000027 time 0.9414 (1.0055) model_time 0.9412 (1.0046) loss 0.7539 (0.9080) grad_norm 8.3616 (8.8396/2.1599) mem 68106MB [2022-12-19 16:53:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1250/1519] eta 0:04:30 lr 0.000027 time 0.9396 (1.0055) model_time 0.9394 (1.0046) loss 1.0540 (0.9083) grad_norm 9.2418 (8.8421/2.1511) mem 68106MB [2022-12-19 16:53:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1260/1519] eta 0:04:20 lr 0.000027 time 0.9359 (1.0055) model_time 0.9358 (1.0045) loss 1.3405 (0.9086) grad_norm 10.2402 (8.8540/2.1361) mem 68106MB [2022-12-19 16:54:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1270/1519] eta 0:04:10 lr 0.000027 time 0.9449 (1.0055) model_time 0.9447 (1.0046) loss 0.6847 (0.9078) grad_norm 7.2162 (8.8372/2.1043) mem 68106MB [2022-12-19 16:54:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1280/1519] eta 0:04:00 lr 0.000027 time 0.9334 (1.0055) model_time 0.9333 (1.0045) loss 1.0113 (0.9073) grad_norm 6.9029 (8.8467/2.0883) mem 68106MB [2022-12-19 16:54:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1290/1519] eta 0:03:50 lr 0.000027 time 0.9300 (1.0055) model_time 0.9299 (1.0046) loss 0.7337 (0.9074) grad_norm 7.4892 (8.7811/2.0777) mem 68106MB [2022-12-19 16:54:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1300/1519] eta 0:03:40 lr 0.000027 time 0.9717 (1.0054) model_time 0.9716 (1.0045) loss 0.7690 (0.9071) grad_norm 6.9392 (8.7466/2.0826) mem 68106MB [2022-12-19 16:54:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1310/1519] eta 0:03:30 lr 0.000027 time 0.9293 (1.0054) model_time 0.9292 (1.0045) loss 0.7178 (0.9065) grad_norm 7.4684 (8.7284/2.0728) mem 68106MB [2022-12-19 16:54:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1320/1519] eta 0:03:20 lr 0.000027 time 0.9367 (1.0054) model_time 0.9366 (1.0045) loss 0.9718 (0.9068) grad_norm 9.8430 (8.7357/2.0699) mem 68106MB [2022-12-19 16:55:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1330/1519] eta 0:03:10 lr 0.000027 time 0.9390 (1.0053) model_time 0.9389 (1.0044) loss 0.8071 (0.9064) grad_norm 10.3862 (8.7252/2.0669) mem 68106MB [2022-12-19 16:55:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1340/1519] eta 0:02:59 lr 0.000027 time 0.9323 (1.0053) model_time 0.9321 (1.0044) loss 0.9318 (0.9063) grad_norm 9.2976 (8.6875/2.0247) mem 68106MB [2022-12-19 16:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1350/1519] eta 0:02:49 lr 0.000027 time 0.9330 (1.0053) model_time 0.9328 (1.0044) loss 0.8166 (0.9067) grad_norm 5.7054 (8.7123/2.0913) mem 68106MB [2022-12-19 16:55:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1360/1519] eta 0:02:39 lr 0.000027 time 0.9313 (1.0053) model_time 0.9311 (1.0044) loss 1.2047 (0.9073) grad_norm 8.0581 (8.7149/2.0943) mem 68106MB [2022-12-19 16:55:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1370/1519] eta 0:02:29 lr 0.000027 time 0.9205 (1.0053) model_time 0.9203 (1.0044) loss 0.7097 (0.9070) grad_norm 7.9003 (8.6917/2.0902) mem 68106MB [2022-12-19 16:55:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1380/1519] eta 0:02:19 lr 0.000027 time 0.9292 (1.0053) model_time 0.9291 (1.0044) loss 1.1640 (0.9074) grad_norm 6.5498 (8.6484/2.0862) mem 68106MB [2022-12-19 16:56:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1390/1519] eta 0:02:09 lr 0.000027 time 1.0375 (1.0053) model_time 1.0373 (1.0045) loss 0.7884 (0.9073) grad_norm 12.5940 (8.6498/2.1292) mem 68106MB [2022-12-19 16:56:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1400/1519] eta 0:01:59 lr 0.000027 time 0.9361 (1.0055) model_time 0.9360 (1.0046) loss 0.9515 (0.9082) grad_norm 6.8658 (8.6570/2.1262) mem 68106MB [2022-12-19 16:56:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1410/1519] eta 0:01:49 lr 0.000027 time 0.9371 (1.0055) model_time 0.9370 (1.0046) loss 0.7818 (0.9079) grad_norm 8.4678 (8.6717/2.1377) mem 68106MB [2022-12-19 16:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1420/1519] eta 0:01:39 lr 0.000027 time 0.9359 (1.0055) model_time 0.9358 (1.0047) loss 1.0421 (0.9081) grad_norm 8.5183 (8.7197/2.1566) mem 68106MB [2022-12-19 16:56:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1430/1519] eta 0:01:29 lr 0.000027 time 0.9354 (1.0056) model_time 0.9353 (1.0047) loss 0.7050 (0.9088) grad_norm 10.9834 (8.7203/2.1426) mem 68106MB [2022-12-19 16:56:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1440/1519] eta 0:01:19 lr 0.000027 time 0.9356 (1.0055) model_time 0.9355 (1.0047) loss 0.8596 (0.9087) grad_norm 9.6408 (8.7271/2.1361) mem 68106MB [2022-12-19 16:57:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1450/1519] eta 0:01:09 lr 0.000027 time 0.9293 (1.0055) model_time 0.9291 (1.0047) loss 0.9343 (0.9086) grad_norm 9.5281 (8.7666/2.1697) mem 68106MB [2022-12-19 16:57:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1460/1519] eta 0:00:59 lr 0.000027 time 0.9310 (1.0055) model_time 0.9309 (1.0046) loss 0.6913 (0.9086) grad_norm 8.5850 (8.7347/2.1607) mem 68106MB [2022-12-19 16:57:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1470/1519] eta 0:00:49 lr 0.000027 time 0.9291 (1.0055) model_time 0.9288 (1.0046) loss 0.8083 (0.9087) grad_norm 11.5462 (8.7497/2.1514) mem 68106MB [2022-12-19 16:57:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1480/1519] eta 0:00:39 lr 0.000027 time 0.9784 (1.0055) model_time 0.9783 (1.0046) loss 0.6986 (0.9087) grad_norm 8.0923 (8.7032/2.0767) mem 68106MB [2022-12-19 16:57:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1490/1519] eta 0:00:29 lr 0.000027 time 0.9337 (1.0054) model_time 0.9335 (1.0046) loss 0.9028 (0.9087) grad_norm 9.3775 (8.7423/2.0792) mem 68106MB [2022-12-19 16:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1500/1519] eta 0:00:19 lr 0.000027 time 0.9485 (1.0054) model_time 0.9483 (1.0046) loss 1.0349 (0.9093) grad_norm 9.6812 (8.7567/2.1012) mem 68106MB [2022-12-19 16:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [28/100][1510/1519] eta 0:00:09 lr 0.000027 time 0.9215 (1.0054) model_time 0.9214 (1.0045) loss 0.7518 (0.9088) grad_norm 7.1389 (8.7767/2.1254) mem 68106MB [2022-12-19 16:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 28 training takes 0:25:27 [2022-12-19 16:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_28.pth saving...... [2022-12-19 16:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_28.pth saved !!! [2022-12-19 16:58:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.691 (0.691) Loss 0.5185 (0.5185) Acc@1 91.319 (91.319) Acc@5 98.264 (98.264) Mem 68106MB [2022-12-19 16:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.332) Loss 0.5391 (0.4985) Acc@1 91.667 (91.572) Acc@5 97.569 (98.264) Mem 68106MB [2022-12-19 16:58:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.318) Loss 0.4568 (0.4931) Acc@1 92.361 (91.452) Acc@5 99.306 (98.347) Mem 68106MB [2022-12-19 16:58:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.300 (0.312) Loss 0.5887 (0.4978) Acc@1 89.583 (91.207) Acc@5 97.569 (98.342) Mem 68106MB [2022-12-19 16:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.308) Loss 0.4598 (0.4900) Acc@1 92.361 (91.336) Acc@5 98.611 (98.416) Mem 68106MB [2022-12-19 16:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.307) Loss 0.4947 (0.4871) Acc@1 88.194 (91.360) Acc@5 99.306 (98.482) Mem 68106MB [2022-12-19 16:58:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.306) Loss 0.5098 (0.4874) Acc@1 89.583 (91.337) Acc@5 98.264 (98.463) Mem 68106MB [2022-12-19 16:59:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5347 (0.4896) Acc@1 92.014 (91.236) Acc@5 98.958 (98.435) Mem 68106MB [2022-12-19 16:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4196 (0.4883) Acc@1 93.403 (91.328) Acc@5 98.958 (98.461) Mem 68106MB [2022-12-19 16:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:28] * Acc@1 91.327 Acc@5 98.465 [2022-12-19 16:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.3% [2022-12-19 16:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 16:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 16:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.33% [2022-12-19 16:59:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][0/1519] eta 0:36:10 lr 0.000027 time 1.4287 (1.4287) model_time 0.9877 (0.9877) loss 0.7677 (0.7677) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 16:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][10/1519] eta 0:26:07 lr 0.000027 time 0.9225 (1.0385) model_time 0.9224 (0.9979) loss 0.8936 (0.8565) grad_norm 7.2540 (7.9254/0.4774) mem 68106MB [2022-12-19 16:59:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][20/1519] eta 0:25:31 lr 0.000027 time 0.9357 (1.0218) model_time 0.9356 (1.0003) loss 0.8816 (0.8869) grad_norm 6.2759 (8.2803/1.7443) mem 68106MB [2022-12-19 17:00:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][30/1519] eta 0:25:19 lr 0.000027 time 0.9290 (1.0203) model_time 0.9288 (1.0056) loss 0.7760 (0.8625) grad_norm 6.9829 (8.4106/1.6053) mem 68106MB [2022-12-19 17:00:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][40/1519] eta 0:25:02 lr 0.000027 time 0.9159 (1.0160) model_time 0.9158 (1.0048) loss 0.8651 (0.8568) grad_norm 7.0173 (8.2522/1.6023) mem 68106MB [2022-12-19 17:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][50/1519] eta 0:24:48 lr 0.000027 time 0.9301 (1.0131) model_time 0.9300 (1.0040) loss 0.9478 (0.8775) grad_norm 8.1858 (8.3491/1.7575) mem 68106MB [2022-12-19 17:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][60/1519] eta 0:24:35 lr 0.000027 time 0.9375 (1.0116) model_time 0.9374 (1.0039) loss 1.0895 (0.8987) grad_norm 9.0118 (8.4285/1.7054) mem 68106MB [2022-12-19 17:00:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][70/1519] eta 0:24:24 lr 0.000027 time 0.9645 (1.0105) model_time 0.9643 (1.0038) loss 0.6948 (0.8868) grad_norm 5.8140 (8.4443/1.7074) mem 68106MB [2022-12-19 17:00:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][80/1519] eta 0:24:15 lr 0.000027 time 0.9292 (1.0115) model_time 0.9291 (1.0056) loss 0.7530 (0.8922) grad_norm 7.4486 (8.4285/1.7469) mem 68106MB [2022-12-19 17:01:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][90/1519] eta 0:24:06 lr 0.000027 time 0.9356 (1.0123) model_time 0.9353 (1.0070) loss 0.8621 (0.8876) grad_norm 6.7358 (8.4654/1.8051) mem 68106MB [2022-12-19 17:01:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][100/1519] eta 0:23:54 lr 0.000027 time 0.9253 (1.0110) model_time 0.9250 (1.0061) loss 0.8543 (0.8849) grad_norm 7.0583 (8.4927/1.8343) mem 68106MB [2022-12-19 17:01:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][110/1519] eta 0:23:42 lr 0.000027 time 0.9287 (1.0098) model_time 0.9285 (1.0053) loss 0.7602 (0.8938) grad_norm 9.2725 (8.4334/1.8114) mem 68106MB [2022-12-19 17:01:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][120/1519] eta 0:23:31 lr 0.000027 time 0.9298 (1.0090) model_time 0.9296 (1.0049) loss 0.8045 (0.8889) grad_norm 8.9182 (8.4583/1.7625) mem 68106MB [2022-12-19 17:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][130/1519] eta 0:23:21 lr 0.000027 time 0.9366 (1.0088) model_time 0.9364 (1.0050) loss 0.8934 (0.8900) grad_norm 8.8251 (8.4634/1.7253) mem 68106MB [2022-12-19 17:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][140/1519] eta 0:23:10 lr 0.000027 time 0.9385 (1.0082) model_time 0.9382 (1.0046) loss 0.9181 (0.8901) grad_norm 6.9852 (8.3998/1.6930) mem 68106MB [2022-12-19 17:02:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][150/1519] eta 0:22:59 lr 0.000027 time 0.9336 (1.0077) model_time 0.9335 (1.0043) loss 0.9826 (0.8881) grad_norm 6.7208 (8.3428/1.6725) mem 68106MB [2022-12-19 17:02:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][160/1519] eta 0:22:48 lr 0.000027 time 0.9361 (1.0073) model_time 0.9359 (1.0041) loss 0.8942 (0.8861) grad_norm 10.3020 (8.3930/1.6802) mem 68106MB [2022-12-19 17:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][170/1519] eta 0:22:39 lr 0.000027 time 0.9433 (1.0080) model_time 0.9431 (1.0049) loss 1.0816 (0.8998) grad_norm 6.6599 (8.4285/1.7741) mem 68106MB [2022-12-19 17:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][180/1519] eta 0:22:29 lr 0.000027 time 0.9352 (1.0075) model_time 0.9350 (1.0046) loss 0.7168 (0.8964) grad_norm 11.6399 (8.4848/1.8519) mem 68106MB [2022-12-19 17:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][190/1519] eta 0:22:19 lr 0.000027 time 0.9277 (1.0081) model_time 0.9276 (1.0054) loss 0.8228 (0.8950) grad_norm 10.4295 (8.4644/1.8328) mem 68106MB [2022-12-19 17:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][200/1519] eta 0:22:09 lr 0.000027 time 0.9332 (1.0077) model_time 0.9330 (1.0051) loss 0.8482 (0.8956) grad_norm 7.8368 (8.5039/1.8067) mem 68106MB [2022-12-19 17:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][210/1519] eta 0:22:00 lr 0.000027 time 0.9496 (1.0090) model_time 0.9495 (1.0064) loss 0.8143 (0.8918) grad_norm 11.3141 (8.4971/1.7934) mem 68106MB [2022-12-19 17:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][220/1519] eta 0:21:50 lr 0.000027 time 0.9266 (1.0088) model_time 0.9264 (1.0063) loss 0.7955 (0.8896) grad_norm 5.5988 (8.4562/1.7827) mem 68106MB [2022-12-19 17:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][230/1519] eta 0:21:39 lr 0.000027 time 0.9337 (1.0084) model_time 0.9336 (1.0060) loss 0.9440 (0.8889) grad_norm 6.9674 (8.5280/1.9675) mem 68106MB [2022-12-19 17:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][240/1519] eta 0:21:29 lr 0.000027 time 0.9296 (1.0080) model_time 0.9294 (1.0058) loss 0.9865 (0.8903) grad_norm 11.6425 (8.5975/1.9868) mem 68106MB [2022-12-19 17:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][250/1519] eta 0:21:18 lr 0.000027 time 0.9333 (1.0077) model_time 0.9332 (1.0055) loss 0.7361 (0.8926) grad_norm 7.2161 (8.5752/1.9736) mem 68106MB [2022-12-19 17:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][260/1519] eta 0:21:08 lr 0.000027 time 0.9318 (1.0075) model_time 0.9316 (1.0053) loss 0.9757 (0.8909) grad_norm 8.9970 (8.6280/2.0658) mem 68106MB [2022-12-19 17:04:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][270/1519] eta 0:20:58 lr 0.000027 time 0.9392 (1.0072) model_time 0.9391 (1.0052) loss 0.7190 (0.8908) grad_norm 8.1412 (8.6023/2.0394) mem 68106MB [2022-12-19 17:04:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][280/1519] eta 0:20:47 lr 0.000027 time 0.9318 (1.0071) model_time 0.9317 (1.0051) loss 0.9815 (0.8921) grad_norm 13.8441 (8.6343/2.0696) mem 68106MB [2022-12-19 17:04:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][290/1519] eta 0:20:37 lr 0.000027 time 0.9331 (1.0069) model_time 0.9330 (1.0049) loss 0.9162 (0.8929) grad_norm 9.4250 (8.6618/2.0517) mem 68106MB [2022-12-19 17:04:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][300/1519] eta 0:20:27 lr 0.000027 time 0.9408 (1.0067) model_time 0.9407 (1.0048) loss 1.0628 (0.8941) grad_norm 6.9240 (8.6760/2.0401) mem 68106MB [2022-12-19 17:04:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][310/1519] eta 0:20:17 lr 0.000027 time 0.9389 (1.0066) model_time 0.9387 (1.0048) loss 0.9220 (0.8935) grad_norm 7.8065 (8.6629/2.0347) mem 68106MB [2022-12-19 17:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][320/1519] eta 0:20:06 lr 0.000027 time 0.9360 (1.0064) model_time 0.9359 (1.0046) loss 0.7842 (0.8934) grad_norm 8.1278 (8.6565/2.0202) mem 68106MB [2022-12-19 17:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][330/1519] eta 0:19:56 lr 0.000027 time 0.9342 (1.0062) model_time 0.9341 (1.0045) loss 1.0166 (0.8936) grad_norm 6.7848 (8.6376/2.0055) mem 68106MB [2022-12-19 17:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][340/1519] eta 0:19:46 lr 0.000027 time 0.9332 (1.0063) model_time 0.9330 (1.0046) loss 1.0508 (0.8920) grad_norm 8.8062 (8.6186/1.9848) mem 68106MB [2022-12-19 17:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][350/1519] eta 0:19:36 lr 0.000027 time 0.9830 (1.0065) model_time 0.9828 (1.0048) loss 1.2154 (0.8913) grad_norm 6.7160 (8.5922/1.9695) mem 68106MB [2022-12-19 17:05:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][360/1519] eta 0:19:26 lr 0.000027 time 0.9322 (1.0063) model_time 0.9321 (1.0046) loss 0.7568 (0.8886) grad_norm 6.9238 (8.5879/1.9483) mem 68106MB [2022-12-19 17:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][370/1519] eta 0:19:16 lr 0.000027 time 0.9306 (1.0063) model_time 0.9305 (1.0048) loss 1.1707 (0.8918) grad_norm 9.0363 (8.6066/1.9584) mem 68106MB [2022-12-19 17:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][380/1519] eta 0:19:06 lr 0.000027 time 0.9291 (1.0062) model_time 0.9290 (1.0046) loss 0.7227 (0.8911) grad_norm 9.2406 (8.6483/1.9618) mem 68106MB [2022-12-19 17:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][390/1519] eta 0:18:56 lr 0.000027 time 0.9337 (1.0070) model_time 0.9336 (1.0054) loss 0.7958 (0.8909) grad_norm 10.7200 (8.6825/1.9572) mem 68106MB [2022-12-19 17:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][400/1519] eta 0:18:46 lr 0.000027 time 1.0371 (1.0070) model_time 1.0370 (1.0055) loss 0.8632 (0.8895) grad_norm 7.9504 (8.6555/1.9412) mem 68106MB [2022-12-19 17:06:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][410/1519] eta 0:18:36 lr 0.000027 time 0.9257 (1.0070) model_time 0.9256 (1.0055) loss 0.7702 (0.8899) grad_norm 7.8510 (8.7097/1.9922) mem 68106MB [2022-12-19 17:06:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][420/1519] eta 0:18:26 lr 0.000027 time 0.9314 (1.0069) model_time 0.9313 (1.0054) loss 1.0065 (0.8886) grad_norm 7.5372 (8.7440/2.0166) mem 68106MB [2022-12-19 17:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][430/1519] eta 0:18:16 lr 0.000027 time 0.9347 (1.0068) model_time 0.9346 (1.0054) loss 0.7133 (0.8870) grad_norm 5.5523 (8.7422/2.0158) mem 68106MB [2022-12-19 17:06:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][440/1519] eta 0:18:06 lr 0.000027 time 0.9461 (1.0067) model_time 0.9459 (1.0053) loss 0.8632 (0.8895) grad_norm 9.1194 (8.7639/2.0070) mem 68106MB [2022-12-19 17:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][450/1519] eta 0:17:56 lr 0.000027 time 0.9431 (1.0066) model_time 0.9430 (1.0052) loss 0.7277 (0.8918) grad_norm 8.8356 (8.7758/1.9915) mem 68106MB [2022-12-19 17:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][460/1519] eta 0:17:45 lr 0.000027 time 0.9885 (1.0066) model_time 0.9883 (1.0052) loss 1.0160 (0.8911) grad_norm 6.9956 (8.8136/2.0409) mem 68106MB [2022-12-19 17:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][470/1519] eta 0:17:35 lr 0.000027 time 0.9359 (1.0065) model_time 0.9357 (1.0052) loss 0.8279 (0.8903) grad_norm 6.4458 (8.8038/2.0272) mem 68106MB [2022-12-19 17:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][480/1519] eta 0:17:25 lr 0.000027 time 0.9300 (1.0063) model_time 0.9298 (1.0050) loss 0.7252 (0.8914) grad_norm 8.2999 (8.7843/2.0123) mem 68106MB [2022-12-19 17:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][490/1519] eta 0:17:15 lr 0.000027 time 0.9434 (1.0062) model_time 0.9433 (1.0049) loss 0.9581 (0.8949) grad_norm 9.5125 (8.7730/2.0017) mem 68106MB [2022-12-19 17:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][500/1519] eta 0:17:05 lr 0.000027 time 0.9283 (1.0067) model_time 0.9282 (1.0054) loss 0.7678 (0.8944) grad_norm 15.3430 (8.7707/2.0409) mem 68106MB [2022-12-19 17:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][510/1519] eta 0:16:55 lr 0.000027 time 0.9372 (1.0065) model_time 0.9370 (1.0053) loss 0.7069 (0.8952) grad_norm 6.7479 (8.7459/2.0341) mem 68106MB [2022-12-19 17:08:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][520/1519] eta 0:16:45 lr 0.000027 time 0.9765 (1.0070) model_time 0.9764 (1.0057) loss 0.8846 (0.8943) grad_norm 9.8036 (8.7334/2.0262) mem 68106MB [2022-12-19 17:08:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][530/1519] eta 0:16:35 lr 0.000027 time 0.9951 (1.0071) model_time 0.9949 (1.0058) loss 0.7891 (0.8922) grad_norm 8.5876 (8.7255/2.0121) mem 68106MB [2022-12-19 17:08:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][540/1519] eta 0:16:25 lr 0.000027 time 0.9384 (1.0069) model_time 0.9382 (1.0057) loss 0.9956 (0.8929) grad_norm 7.0433 (8.7512/2.0455) mem 68106MB [2022-12-19 17:08:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][550/1519] eta 0:16:15 lr 0.000027 time 0.9333 (1.0068) model_time 0.9332 (1.0056) loss 0.8455 (0.8941) grad_norm 7.6078 (8.7402/2.0442) mem 68106MB [2022-12-19 17:08:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][560/1519] eta 0:16:05 lr 0.000027 time 0.9305 (1.0067) model_time 0.9303 (1.0055) loss 0.8940 (0.8928) grad_norm 7.1503 (8.7241/2.0371) mem 68106MB [2022-12-19 17:09:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][570/1519] eta 0:15:55 lr 0.000027 time 0.9886 (1.0072) model_time 0.9885 (1.0060) loss 0.8215 (0.8931) grad_norm 14.1754 (8.7839/2.1012) mem 68106MB [2022-12-19 17:09:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][580/1519] eta 0:15:45 lr 0.000027 time 0.9332 (1.0070) model_time 0.9329 (1.0059) loss 0.6667 (0.8931) grad_norm 6.7479 (8.7621/2.0998) mem 68106MB [2022-12-19 17:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][590/1519] eta 0:15:35 lr 0.000027 time 0.9528 (1.0070) model_time 0.9526 (1.0059) loss 1.2319 (0.8950) grad_norm 7.5947 (8.7568/2.0897) mem 68106MB [2022-12-19 17:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][600/1519] eta 0:15:25 lr 0.000027 time 0.9338 (1.0069) model_time 0.9336 (1.0058) loss 0.7404 (0.8944) grad_norm 14.8768 (8.7791/2.1041) mem 68106MB [2022-12-19 17:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][610/1519] eta 0:15:15 lr 0.000027 time 0.9388 (1.0068) model_time 0.9387 (1.0057) loss 0.7774 (0.8953) grad_norm 6.7653 (8.7884/2.1171) mem 68106MB [2022-12-19 17:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][620/1519] eta 0:15:05 lr 0.000027 time 0.9402 (1.0068) model_time 0.9400 (1.0057) loss 0.8507 (0.8946) grad_norm 10.1570 (8.7983/2.1115) mem 68106MB [2022-12-19 17:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][630/1519] eta 0:14:54 lr 0.000027 time 0.9324 (1.0067) model_time 0.9322 (1.0056) loss 0.9129 (0.8963) grad_norm 6.4992 (8.7896/2.1208) mem 68106MB [2022-12-19 17:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][640/1519] eta 0:14:44 lr 0.000027 time 0.9960 (1.0067) model_time 0.9958 (1.0056) loss 0.9113 (0.8973) grad_norm 9.4849 (8.8393/2.1451) mem 68106MB [2022-12-19 17:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][650/1519] eta 0:14:34 lr 0.000027 time 0.9333 (1.0066) model_time 0.9331 (1.0055) loss 0.9597 (0.8995) grad_norm 6.4610 (8.8285/2.1353) mem 68106MB [2022-12-19 17:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][660/1519] eta 0:14:24 lr 0.000027 time 0.9410 (1.0066) model_time 0.9408 (1.0055) loss 0.7903 (0.8991) grad_norm 10.2988 (8.8171/2.1422) mem 68106MB [2022-12-19 17:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][670/1519] eta 0:14:14 lr 0.000027 time 0.9328 (1.0064) model_time 0.9326 (1.0054) loss 1.0240 (0.9004) grad_norm 13.4163 (8.8414/2.1480) mem 68106MB [2022-12-19 17:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][680/1519] eta 0:14:04 lr 0.000027 time 0.9269 (1.0065) model_time 0.9264 (1.0054) loss 0.7109 (0.9010) grad_norm 9.5861 (8.8459/2.1337) mem 68106MB [2022-12-19 17:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][690/1519] eta 0:13:54 lr 0.000027 time 0.9387 (1.0064) model_time 0.9385 (1.0054) loss 1.0746 (0.8999) grad_norm 6.6302 (8.8532/2.1331) mem 68106MB [2022-12-19 17:11:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][700/1519] eta 0:13:44 lr 0.000027 time 1.0178 (1.0069) model_time 1.0176 (1.0059) loss 0.8730 (0.8983) grad_norm 14.0272 (8.9124/2.1949) mem 68106MB [2022-12-19 17:11:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][710/1519] eta 0:13:34 lr 0.000027 time 0.9275 (1.0069) model_time 0.9272 (1.0059) loss 1.1015 (0.8972) grad_norm 8.6497 (8.9131/2.1973) mem 68106MB [2022-12-19 17:11:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][720/1519] eta 0:13:24 lr 0.000027 time 0.9366 (1.0069) model_time 0.9364 (1.0059) loss 0.8893 (0.8984) grad_norm 6.5080 (8.8998/2.2024) mem 68106MB [2022-12-19 17:11:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][730/1519] eta 0:13:14 lr 0.000027 time 0.9325 (1.0068) model_time 0.9323 (1.0058) loss 0.8526 (0.8980) grad_norm 6.9253 (8.8851/2.2034) mem 68106MB [2022-12-19 17:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][740/1519] eta 0:13:04 lr 0.000027 time 0.9356 (1.0067) model_time 0.9354 (1.0057) loss 0.8553 (0.8994) grad_norm 7.5272 (8.8950/2.2040) mem 68106MB [2022-12-19 17:12:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][750/1519] eta 0:12:54 lr 0.000027 time 0.9863 (1.0067) model_time 0.9861 (1.0057) loss 0.7067 (0.8988) grad_norm 13.3518 (8.9223/2.2192) mem 68106MB [2022-12-19 17:12:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][760/1519] eta 0:12:44 lr 0.000027 time 0.9320 (1.0066) model_time 0.9318 (1.0056) loss 0.8787 (0.8985) grad_norm 10.9645 (8.9153/2.2168) mem 68106MB [2022-12-19 17:12:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][770/1519] eta 0:12:33 lr 0.000027 time 0.9297 (1.0065) model_time 0.9295 (1.0056) loss 1.3464 (0.8994) grad_norm 9.0414 (8.9254/2.1982) mem 68106MB [2022-12-19 17:12:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][780/1519] eta 0:12:23 lr 0.000027 time 0.9313 (1.0064) model_time 0.9311 (1.0055) loss 0.8838 (0.8989) grad_norm 8.0400 (8.9190/2.1870) mem 68106MB [2022-12-19 17:12:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][790/1519] eta 0:12:13 lr 0.000027 time 0.9317 (1.0064) model_time 0.9315 (1.0054) loss 1.0138 (0.8983) grad_norm 6.8338 (8.9188/2.1885) mem 68106MB [2022-12-19 17:12:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][800/1519] eta 0:12:03 lr 0.000027 time 0.9291 (1.0063) model_time 0.9289 (1.0054) loss 0.8468 (0.8984) grad_norm 6.4120 (8.8906/2.2004) mem 68106MB [2022-12-19 17:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][810/1519] eta 0:11:53 lr 0.000027 time 0.9286 (1.0063) model_time 0.9285 (1.0053) loss 1.0049 (0.8980) grad_norm 5.8516 (8.8899/2.2021) mem 68106MB [2022-12-19 17:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][820/1519] eta 0:11:43 lr 0.000027 time 0.9373 (1.0062) model_time 0.9371 (1.0053) loss 0.8751 (0.8974) grad_norm 6.7441 (8.9028/2.1994) mem 68106MB [2022-12-19 17:13:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][830/1519] eta 0:11:33 lr 0.000027 time 0.9316 (1.0062) model_time 0.9315 (1.0053) loss 0.7366 (0.8978) grad_norm 12.0923 (8.9079/2.1540) mem 68106MB [2022-12-19 17:13:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][840/1519] eta 0:11:23 lr 0.000027 time 1.0022 (1.0063) model_time 1.0021 (1.0054) loss 0.7844 (0.8976) grad_norm 11.0138 (8.8933/2.1399) mem 68106MB [2022-12-19 17:13:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][850/1519] eta 0:11:13 lr 0.000027 time 0.9382 (1.0063) model_time 0.9381 (1.0054) loss 1.0932 (0.8989) grad_norm 8.5026 (8.9154/2.1381) mem 68106MB [2022-12-19 17:13:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][860/1519] eta 0:11:03 lr 0.000027 time 0.9336 (1.0062) model_time 0.9335 (1.0053) loss 0.9788 (0.8987) grad_norm 8.5385 (8.9130/2.1087) mem 68106MB [2022-12-19 17:14:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][870/1519] eta 0:10:53 lr 0.000027 time 0.9304 (1.0062) model_time 0.9303 (1.0053) loss 0.8216 (0.8984) grad_norm 10.2148 (8.9515/2.1268) mem 68106MB [2022-12-19 17:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][880/1519] eta 0:10:43 lr 0.000027 time 0.9839 (1.0063) model_time 0.9838 (1.0054) loss 0.7099 (0.8989) grad_norm 8.3122 (8.9372/2.1092) mem 68106MB [2022-12-19 17:14:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][890/1519] eta 0:10:32 lr 0.000027 time 0.9357 (1.0063) model_time 0.9356 (1.0054) loss 0.9212 (0.8992) grad_norm 9.4329 (8.9368/2.1161) mem 68106MB [2022-12-19 17:14:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][900/1519] eta 0:10:22 lr 0.000027 time 0.9323 (1.0064) model_time 0.9321 (1.0055) loss 0.9673 (0.8992) grad_norm 9.6883 (8.9521/2.1176) mem 68106MB [2022-12-19 17:14:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][910/1519] eta 0:10:12 lr 0.000027 time 0.9285 (1.0063) model_time 0.9284 (1.0054) loss 0.8519 (0.8992) grad_norm 11.5316 (8.9775/2.1216) mem 68106MB [2022-12-19 17:14:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][920/1519] eta 0:10:02 lr 0.000027 time 0.9385 (1.0062) model_time 0.9383 (1.0054) loss 0.8209 (0.8991) grad_norm 8.8098 (8.9826/2.1187) mem 68106MB [2022-12-19 17:15:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][930/1519] eta 0:09:52 lr 0.000027 time 0.9616 (1.0062) model_time 0.9614 (1.0053) loss 0.9190 (0.8985) grad_norm 8.8220 (9.0189/2.1329) mem 68106MB [2022-12-19 17:15:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][940/1519] eta 0:09:42 lr 0.000027 time 0.9313 (1.0061) model_time 0.9312 (1.0052) loss 0.7433 (0.8980) grad_norm 6.3686 (9.0500/2.1501) mem 68106MB [2022-12-19 17:15:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][950/1519] eta 0:09:32 lr 0.000027 time 0.9352 (1.0059) model_time 0.9351 (1.0051) loss 0.6812 (0.8977) grad_norm 9.0586 (9.0779/2.1630) mem 68106MB [2022-12-19 17:15:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][960/1519] eta 0:09:22 lr 0.000027 time 0.9311 (1.0059) model_time 0.9309 (1.0051) loss 1.0192 (0.8980) grad_norm 7.6731 (9.0804/2.1597) mem 68106MB [2022-12-19 17:15:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][970/1519] eta 0:09:12 lr 0.000027 time 0.9389 (1.0060) model_time 0.9387 (1.0051) loss 0.7834 (0.8987) grad_norm 7.9333 (9.0818/2.1561) mem 68106MB [2022-12-19 17:15:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][980/1519] eta 0:09:02 lr 0.000027 time 0.9360 (1.0060) model_time 0.9358 (1.0051) loss 0.7808 (0.8984) grad_norm 7.9951 (9.0673/2.1640) mem 68106MB [2022-12-19 17:16:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][990/1519] eta 0:08:52 lr 0.000027 time 1.0455 (1.0061) model_time 1.0454 (1.0052) loss 0.8862 (0.8980) grad_norm 8.5189 (9.0617/2.1606) mem 68106MB [2022-12-19 17:16:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1000/1519] eta 0:08:42 lr 0.000027 time 0.9366 (1.0061) model_time 0.9365 (1.0053) loss 0.8850 (0.8981) grad_norm 7.0208 (9.0705/2.1609) mem 68106MB [2022-12-19 17:16:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1010/1519] eta 0:08:32 lr 0.000027 time 0.9396 (1.0062) model_time 0.9394 (1.0053) loss 0.7934 (0.8977) grad_norm 12.4352 (9.0417/2.1377) mem 68106MB [2022-12-19 17:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1020/1519] eta 0:08:22 lr 0.000027 time 1.0003 (1.0064) model_time 1.0002 (1.0056) loss 0.8068 (0.8978) grad_norm 7.8319 (9.0133/2.1174) mem 68106MB [2022-12-19 17:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1030/1519] eta 0:08:12 lr 0.000027 time 0.9338 (1.0064) model_time 0.9336 (1.0055) loss 0.8816 (0.8985) grad_norm 9.5262 (9.0353/2.1651) mem 68106MB [2022-12-19 17:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1040/1519] eta 0:08:02 lr 0.000027 time 0.9387 (1.0063) model_time 0.9386 (1.0055) loss 1.1612 (0.8986) grad_norm 6.1413 (9.0477/2.1919) mem 68106MB [2022-12-19 17:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1050/1519] eta 0:07:51 lr 0.000027 time 0.9242 (1.0062) model_time 0.9240 (1.0054) loss 0.6687 (0.8979) grad_norm 5.8215 (9.0110/2.2061) mem 68106MB [2022-12-19 17:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1060/1519] eta 0:07:41 lr 0.000027 time 0.9366 (1.0062) model_time 0.9365 (1.0054) loss 0.9074 (0.8980) grad_norm 11.8936 (9.0033/2.1832) mem 68106MB [2022-12-19 17:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1070/1519] eta 0:07:31 lr 0.000027 time 0.9432 (1.0062) model_time 0.9431 (1.0054) loss 0.9030 (0.8990) grad_norm 16.4241 (9.0341/2.2233) mem 68106MB [2022-12-19 17:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1080/1519] eta 0:07:21 lr 0.000027 time 0.9223 (1.0062) model_time 0.9222 (1.0054) loss 1.0578 (0.8988) grad_norm 10.7690 (9.0654/2.2178) mem 68106MB [2022-12-19 17:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1090/1519] eta 0:07:11 lr 0.000027 time 0.9367 (1.0062) model_time 0.9365 (1.0054) loss 0.6811 (0.8988) grad_norm 8.6753 (9.0797/2.2129) mem 68106MB [2022-12-19 17:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1100/1519] eta 0:07:01 lr 0.000027 time 0.9242 (1.0061) model_time 0.9240 (1.0053) loss 1.1835 (0.8990) grad_norm 5.5881 (9.0931/2.2095) mem 68106MB [2022-12-19 17:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1110/1519] eta 0:06:51 lr 0.000027 time 0.9295 (1.0061) model_time 0.9294 (1.0053) loss 0.7278 (0.8987) grad_norm 10.7852 (9.0991/2.1786) mem 68106MB [2022-12-19 17:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1120/1519] eta 0:06:41 lr 0.000027 time 0.9373 (1.0060) model_time 0.9371 (1.0052) loss 0.7177 (0.8989) grad_norm 11.2594 (9.1592/2.1878) mem 68106MB [2022-12-19 17:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1130/1519] eta 0:06:31 lr 0.000027 time 0.9318 (1.0060) model_time 0.9317 (1.0052) loss 0.7011 (0.8993) grad_norm 8.7016 (9.1525/2.1862) mem 68106MB [2022-12-19 17:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1140/1519] eta 0:06:21 lr 0.000027 time 1.0056 (1.0060) model_time 1.0054 (1.0052) loss 0.9328 (0.8993) grad_norm 6.4972 (9.1109/2.1651) mem 68106MB [2022-12-19 17:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1150/1519] eta 0:06:11 lr 0.000027 time 0.9358 (1.0060) model_time 0.9357 (1.0052) loss 0.8923 (0.8995) grad_norm 9.3474 (9.1233/2.1563) mem 68106MB [2022-12-19 17:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1160/1519] eta 0:06:01 lr 0.000027 time 0.9441 (1.0060) model_time 0.9440 (1.0052) loss 0.7594 (0.8992) grad_norm 7.2701 (9.1055/2.1651) mem 68106MB [2022-12-19 17:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1170/1519] eta 0:05:51 lr 0.000027 time 1.0078 (1.0060) model_time 1.0077 (1.0053) loss 1.1345 (0.8995) grad_norm 6.4348 (9.0675/2.1284) mem 68106MB [2022-12-19 17:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1180/1519] eta 0:05:41 lr 0.000027 time 0.9278 (1.0060) model_time 0.9277 (1.0052) loss 0.7314 (0.8997) grad_norm 8.2623 (9.0364/2.1087) mem 68106MB [2022-12-19 17:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1190/1519] eta 0:05:31 lr 0.000027 time 0.9368 (1.0061) model_time 0.9367 (1.0054) loss 0.7722 (0.8997) grad_norm 10.1011 (9.0508/2.1009) mem 68106MB [2022-12-19 17:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1200/1519] eta 0:05:21 lr 0.000027 time 0.9361 (1.0063) model_time 0.9358 (1.0055) loss 0.7059 (0.8994) grad_norm 7.4616 (9.0413/2.1046) mem 68106MB [2022-12-19 17:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1210/1519] eta 0:05:10 lr 0.000027 time 0.9334 (1.0063) model_time 0.9332 (1.0056) loss 0.7123 (0.9001) grad_norm 8.9002 (9.0311/2.0899) mem 68106MB [2022-12-19 17:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1220/1519] eta 0:05:00 lr 0.000027 time 0.9291 (1.0062) model_time 0.9290 (1.0055) loss 0.9382 (0.9002) grad_norm 7.3382 (9.0274/2.0897) mem 68106MB [2022-12-19 17:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1230/1519] eta 0:04:50 lr 0.000027 time 0.9340 (1.0062) model_time 0.9338 (1.0054) loss 0.7389 (0.9009) grad_norm 9.0928 (9.0136/2.0839) mem 68106MB [2022-12-19 17:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1240/1519] eta 0:04:40 lr 0.000027 time 0.9299 (1.0061) model_time 0.9297 (1.0054) loss 1.0771 (0.9005) grad_norm 10.5594 (8.9991/2.0553) mem 68106MB [2022-12-19 17:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1250/1519] eta 0:04:30 lr 0.000027 time 0.9359 (1.0061) model_time 0.9358 (1.0054) loss 0.8089 (0.8996) grad_norm 5.3878 (8.9955/2.0701) mem 68106MB [2022-12-19 17:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1260/1519] eta 0:04:20 lr 0.000027 time 0.9392 (1.0061) model_time 0.9391 (1.0053) loss 1.3808 (0.9000) grad_norm 6.8528 (8.9945/2.0748) mem 68106MB [2022-12-19 17:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1270/1519] eta 0:04:10 lr 0.000027 time 0.9317 (1.0060) model_time 0.9316 (1.0053) loss 1.1270 (0.8999) grad_norm 10.4534 (8.9992/2.0797) mem 68106MB [2022-12-19 17:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1280/1519] eta 0:04:00 lr 0.000027 time 0.9908 (1.0061) model_time 0.9906 (1.0054) loss 1.5152 (0.9003) grad_norm 10.6567 (8.9960/2.0683) mem 68106MB [2022-12-19 17:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1290/1519] eta 0:03:50 lr 0.000027 time 0.9562 (1.0061) model_time 0.9561 (1.0053) loss 1.1721 (0.9010) grad_norm 9.6425 (8.9770/2.0602) mem 68106MB [2022-12-19 17:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1300/1519] eta 0:03:40 lr 0.000027 time 0.9299 (1.0060) model_time 0.9297 (1.0053) loss 1.0646 (0.9010) grad_norm 8.5039 (8.9401/1.9998) mem 68106MB [2022-12-19 17:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1310/1519] eta 0:03:30 lr 0.000027 time 0.9305 (1.0063) model_time 0.9304 (1.0055) loss 0.7213 (0.9016) grad_norm 8.6196 (8.9348/1.9721) mem 68106MB [2022-12-19 17:21:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1320/1519] eta 0:03:20 lr 0.000027 time 1.0922 (1.0064) model_time 1.0920 (1.0056) loss 1.1174 (0.9018) grad_norm 7.1688 (8.9274/1.9771) mem 68106MB [2022-12-19 17:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1330/1519] eta 0:03:10 lr 0.000027 time 0.9138 (1.0065) model_time 0.9137 (1.0058) loss 0.9594 (0.9014) grad_norm 8.6267 (8.9587/1.9814) mem 68106MB [2022-12-19 17:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1340/1519] eta 0:03:00 lr 0.000027 time 0.9300 (1.0065) model_time 0.9298 (1.0058) loss 0.6973 (0.9014) grad_norm 7.7464 (8.9666/1.9813) mem 68106MB [2022-12-19 17:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1350/1519] eta 0:02:50 lr 0.000027 time 0.9379 (1.0065) model_time 0.9377 (1.0058) loss 1.1717 (0.9020) grad_norm 10.5733 (9.0181/2.0196) mem 68106MB [2022-12-19 17:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1360/1519] eta 0:02:40 lr 0.000027 time 0.9401 (1.0065) model_time 0.9399 (1.0057) loss 1.1502 (0.9022) grad_norm 8.8369 (9.0151/2.0100) mem 68106MB [2022-12-19 17:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1370/1519] eta 0:02:29 lr 0.000027 time 0.9321 (1.0064) model_time 0.9319 (1.0057) loss 1.0033 (0.9018) grad_norm 8.3391 (8.9860/2.0047) mem 68106MB [2022-12-19 17:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1380/1519] eta 0:02:19 lr 0.000027 time 0.9325 (1.0064) model_time 0.9324 (1.0057) loss 0.8466 (0.9014) grad_norm 6.6046 (8.9512/2.0025) mem 68106MB [2022-12-19 17:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1390/1519] eta 0:02:09 lr 0.000027 time 0.9323 (1.0063) model_time 0.9321 (1.0056) loss 1.2450 (0.9017) grad_norm 9.1875 (8.9765/2.0168) mem 68106MB [2022-12-19 17:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1400/1519] eta 0:01:59 lr 0.000027 time 0.9285 (1.0063) model_time 0.9284 (1.0056) loss 0.7872 (0.9018) grad_norm 8.5147 (8.9966/2.0151) mem 68106MB [2022-12-19 17:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1410/1519] eta 0:01:49 lr 0.000027 time 0.9334 (1.0063) model_time 0.9333 (1.0056) loss 0.6939 (0.9011) grad_norm 12.6842 (9.0170/2.0235) mem 68106MB [2022-12-19 17:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1420/1519] eta 0:01:39 lr 0.000027 time 0.9321 (1.0062) model_time 0.9320 (1.0055) loss 1.0611 (0.9016) grad_norm 6.3407 (9.0165/2.0226) mem 68106MB [2022-12-19 17:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1430/1519] eta 0:01:29 lr 0.000027 time 0.9305 (1.0062) model_time 0.9304 (1.0055) loss 0.9930 (0.9018) grad_norm 8.2300 (8.9982/2.0059) mem 68106MB [2022-12-19 17:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1440/1519] eta 0:01:19 lr 0.000027 time 0.9345 (1.0061) model_time 0.9343 (1.0054) loss 0.9078 (0.9026) grad_norm 9.8271 (8.9703/2.0096) mem 68106MB [2022-12-19 17:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1450/1519] eta 0:01:09 lr 0.000027 time 0.9309 (1.0061) model_time 0.9307 (1.0054) loss 1.0329 (0.9021) grad_norm 9.6761 (8.9485/2.0042) mem 68106MB [2022-12-19 17:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1460/1519] eta 0:00:59 lr 0.000027 time 1.0040 (1.0061) model_time 1.0038 (1.0055) loss 1.2364 (0.9023) grad_norm 7.2784 (8.9284/1.9930) mem 68106MB [2022-12-19 17:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1470/1519] eta 0:00:49 lr 0.000027 time 0.9297 (1.0061) model_time 0.9296 (1.0054) loss 1.1128 (0.9022) grad_norm 8.8891 (8.9007/1.9753) mem 68106MB [2022-12-19 17:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1480/1519] eta 0:00:39 lr 0.000027 time 0.9342 (1.0061) model_time 0.9340 (1.0054) loss 1.1161 (0.9023) grad_norm 6.7871 (8.8849/1.9764) mem 68106MB [2022-12-19 17:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1490/1519] eta 0:00:29 lr 0.000027 time 0.9272 (1.0062) model_time 0.9271 (1.0055) loss 1.2796 (0.9023) grad_norm 8.0685 (8.8754/1.9794) mem 68106MB [2022-12-19 17:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1500/1519] eta 0:00:19 lr 0.000027 time 0.9851 (1.0062) model_time 0.9849 (1.0055) loss 0.9174 (0.9019) grad_norm 12.4680 (8.9006/2.0500) mem 68106MB [2022-12-19 17:24:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [29/100][1510/1519] eta 0:00:09 lr 0.000027 time 0.9221 (1.0063) model_time 0.9220 (1.0056) loss 0.9764 (0.9016) grad_norm 7.6101 (8.8797/2.0434) mem 68106MB [2022-12-19 17:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 29 training takes 0:25:28 [2022-12-19 17:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_29.pth saving...... [2022-12-19 17:25:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_29.pth saved !!! [2022-12-19 17:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.628 (0.628) Loss 0.5118 (0.5118) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 17:25:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.328) Loss 0.5035 (0.4881) Acc@1 92.708 (92.045) Acc@5 97.569 (98.422) Mem 68106MB [2022-12-19 17:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.313) Loss 0.4667 (0.4869) Acc@1 91.667 (91.518) Acc@5 98.611 (98.313) Mem 68106MB [2022-12-19 17:25:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.308) Loss 0.5534 (0.4911) Acc@1 91.319 (91.510) Acc@5 97.917 (98.253) Mem 68106MB [2022-12-19 17:25:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.306) Loss 0.4382 (0.4857) Acc@1 92.708 (91.540) Acc@5 99.306 (98.298) Mem 68106MB [2022-12-19 17:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.305) Loss 0.5097 (0.4840) Acc@1 89.236 (91.442) Acc@5 99.306 (98.359) Mem 68106MB [2022-12-19 17:25:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.304) Loss 0.5019 (0.4834) Acc@1 90.625 (91.462) Acc@5 97.917 (98.389) Mem 68106MB [2022-12-19 17:25:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5397 (0.4851) Acc@1 92.014 (91.393) Acc@5 98.264 (98.386) Mem 68106MB [2022-12-19 17:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4294 (0.4839) Acc@1 91.667 (91.418) Acc@5 97.569 (98.397) Mem 68106MB [2022-12-19 17:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:29] * Acc@1 91.405 Acc@5 98.404 [2022-12-19 17:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.4% [2022-12-19 17:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 17:26:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 17:26:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.40% [2022-12-19 17:26:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][0/1519] eta 0:35:25 lr 0.000027 time 1.3996 (1.3996) model_time 0.9940 (0.9940) loss 0.9319 (0.9319) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 17:26:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][10/1519] eta 0:26:07 lr 0.000027 time 0.9209 (1.0385) model_time 0.9208 (1.0013) loss 0.7508 (0.9192) grad_norm 8.3385 (8.2184/1.1801) mem 68106MB [2022-12-19 17:26:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][20/1519] eta 0:25:29 lr 0.000027 time 0.9300 (1.0202) model_time 0.9298 (1.0005) loss 0.7901 (0.9298) grad_norm 7.2959 (8.0930/1.0714) mem 68106MB [2022-12-19 17:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][30/1519] eta 0:25:08 lr 0.000027 time 0.9204 (1.0133) model_time 0.9201 (0.9998) loss 0.7588 (0.9382) grad_norm 8.4475 (8.4279/1.0627) mem 68106MB [2022-12-19 17:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][40/1519] eta 0:24:57 lr 0.000027 time 0.9318 (1.0122) model_time 0.9316 (1.0018) loss 0.8512 (0.9267) grad_norm 9.4611 (8.6473/1.3023) mem 68106MB [2022-12-19 17:27:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][50/1519] eta 0:24:43 lr 0.000027 time 0.9360 (1.0100) model_time 0.9358 (1.0017) loss 0.7294 (0.9293) grad_norm 11.4363 (8.7782/1.3683) mem 68106MB [2022-12-19 17:27:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][60/1519] eta 0:24:33 lr 0.000027 time 1.0296 (1.0101) model_time 1.0295 (1.0030) loss 1.0671 (0.9323) grad_norm 8.1804 (8.7702/1.4194) mem 68106MB [2022-12-19 17:27:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][70/1519] eta 0:24:23 lr 0.000027 time 0.9836 (1.0103) model_time 0.9834 (1.0042) loss 0.9692 (0.9214) grad_norm 9.5249 (8.6628/1.5070) mem 68106MB [2022-12-19 17:27:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][80/1519] eta 0:24:13 lr 0.000027 time 0.9226 (1.0102) model_time 0.9225 (1.0048) loss 0.9959 (0.9246) grad_norm 7.0996 (8.6849/1.5109) mem 68106MB [2022-12-19 17:27:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][90/1519] eta 0:24:02 lr 0.000027 time 0.9202 (1.0098) model_time 0.9199 (1.0049) loss 1.0147 (0.9321) grad_norm 8.3213 (8.5477/1.5003) mem 68106MB [2022-12-19 17:28:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][100/1519] eta 0:23:52 lr 0.000027 time 0.9280 (1.0094) model_time 0.9278 (1.0050) loss 0.7051 (0.9232) grad_norm 10.0252 (8.6331/1.4938) mem 68106MB [2022-12-19 17:28:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][110/1519] eta 0:23:41 lr 0.000027 time 0.9266 (1.0088) model_time 0.9265 (1.0048) loss 0.8435 (0.9131) grad_norm 6.8797 (8.6277/1.5172) mem 68106MB [2022-12-19 17:28:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][120/1519] eta 0:23:30 lr 0.000027 time 0.9259 (1.0085) model_time 0.9256 (1.0047) loss 0.9538 (0.9043) grad_norm 9.0573 (8.6250/1.5124) mem 68106MB [2022-12-19 17:28:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][130/1519] eta 0:23:20 lr 0.000027 time 0.9304 (1.0081) model_time 0.9303 (1.0046) loss 1.5264 (0.9012) grad_norm 10.0713 (8.7287/1.8406) mem 68106MB [2022-12-19 17:28:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][140/1519] eta 0:23:09 lr 0.000027 time 0.9277 (1.0076) model_time 0.9275 (1.0044) loss 1.0378 (0.9033) grad_norm 10.5599 (8.7601/1.8352) mem 68106MB [2022-12-19 17:28:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][150/1519] eta 0:22:58 lr 0.000027 time 0.9347 (1.0071) model_time 0.9346 (1.0040) loss 1.0470 (0.9024) grad_norm 7.7328 (8.7313/1.8342) mem 68106MB [2022-12-19 17:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][160/1519] eta 0:22:48 lr 0.000027 time 0.9240 (1.0066) model_time 0.9238 (1.0037) loss 1.0687 (0.9086) grad_norm 11.0221 (8.7044/1.8270) mem 68106MB [2022-12-19 17:29:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][170/1519] eta 0:22:37 lr 0.000027 time 0.9283 (1.0061) model_time 0.9282 (1.0033) loss 0.7337 (0.9035) grad_norm 9.4194 (8.7623/1.8617) mem 68106MB [2022-12-19 17:29:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][180/1519] eta 0:22:27 lr 0.000027 time 0.9242 (1.0065) model_time 0.9240 (1.0039) loss 0.6936 (0.9033) grad_norm 11.3629 (8.8841/1.9347) mem 68106MB [2022-12-19 17:29:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][190/1519] eta 0:22:17 lr 0.000027 time 0.9236 (1.0061) model_time 0.9235 (1.0036) loss 0.8703 (0.9034) grad_norm 12.0466 (8.8764/1.9309) mem 68106MB [2022-12-19 17:29:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][200/1519] eta 0:22:07 lr 0.000027 time 0.9283 (1.0063) model_time 0.9281 (1.0039) loss 0.6831 (0.9040) grad_norm 8.1728 (8.8492/1.8936) mem 68106MB [2022-12-19 17:29:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][210/1519] eta 0:21:57 lr 0.000027 time 0.9323 (1.0062) model_time 0.9321 (1.0039) loss 0.8947 (0.9046) grad_norm 5.6878 (8.8557/1.9558) mem 68106MB [2022-12-19 17:30:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][220/1519] eta 0:21:46 lr 0.000027 time 0.9234 (1.0061) model_time 0.9232 (1.0038) loss 0.7293 (0.9035) grad_norm 10.6430 (8.8389/1.9336) mem 68106MB [2022-12-19 17:30:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][230/1519] eta 0:21:36 lr 0.000027 time 0.9272 (1.0061) model_time 0.9271 (1.0040) loss 1.1544 (0.9055) grad_norm 9.5759 (8.8427/1.9507) mem 68106MB [2022-12-19 17:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][240/1519] eta 0:21:26 lr 0.000027 time 0.9261 (1.0058) model_time 0.9260 (1.0038) loss 0.7934 (0.9081) grad_norm 8.3581 (8.8353/1.9438) mem 68106MB [2022-12-19 17:30:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][250/1519] eta 0:21:16 lr 0.000027 time 0.9259 (1.0057) model_time 0.9258 (1.0037) loss 1.1985 (0.9141) grad_norm 10.7824 (8.8633/1.9243) mem 68106MB [2022-12-19 17:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][260/1519] eta 0:21:06 lr 0.000027 time 0.9255 (1.0061) model_time 0.9254 (1.0041) loss 0.7783 (0.9148) grad_norm 6.3646 (8.8316/1.9046) mem 68106MB [2022-12-19 17:30:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][270/1519] eta 0:20:56 lr 0.000027 time 0.9285 (1.0061) model_time 0.9283 (1.0042) loss 0.9266 (0.9156) grad_norm 7.1964 (8.8060/1.8903) mem 68106MB [2022-12-19 17:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][280/1519] eta 0:20:46 lr 0.000027 time 0.9263 (1.0058) model_time 0.9262 (1.0040) loss 0.8573 (0.9156) grad_norm 9.9796 (8.8639/1.9325) mem 68106MB [2022-12-19 17:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][290/1519] eta 0:20:36 lr 0.000027 time 0.9290 (1.0064) model_time 0.9288 (1.0046) loss 1.2706 (0.9160) grad_norm 10.4769 (8.8326/1.9375) mem 68106MB [2022-12-19 17:31:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][300/1519] eta 0:20:26 lr 0.000027 time 0.9314 (1.0063) model_time 0.9313 (1.0045) loss 0.8887 (0.9164) grad_norm 10.9249 (8.8563/1.9255) mem 68106MB [2022-12-19 17:31:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][310/1519] eta 0:20:16 lr 0.000027 time 0.9322 (1.0066) model_time 0.9320 (1.0049) loss 0.8404 (0.9137) grad_norm 9.3110 (8.8370/1.9142) mem 68106MB [2022-12-19 17:31:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][320/1519] eta 0:20:06 lr 0.000027 time 0.9235 (1.0063) model_time 0.9234 (1.0047) loss 1.0549 (0.9129) grad_norm 7.5844 (8.8179/1.9104) mem 68106MB [2022-12-19 17:31:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][330/1519] eta 0:19:56 lr 0.000027 time 0.9354 (1.0061) model_time 0.9353 (1.0045) loss 0.8923 (0.9156) grad_norm 9.8521 (8.7878/1.9028) mem 68106MB [2022-12-19 17:32:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][340/1519] eta 0:19:46 lr 0.000027 time 0.9288 (1.0060) model_time 0.9286 (1.0045) loss 1.1321 (0.9179) grad_norm 7.4104 (8.7788/1.8793) mem 68106MB [2022-12-19 17:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][350/1519] eta 0:19:36 lr 0.000027 time 0.9222 (1.0060) model_time 0.9219 (1.0045) loss 1.3787 (0.9205) grad_norm 6.8602 (8.7328/1.8848) mem 68106MB [2022-12-19 17:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][360/1519] eta 0:19:25 lr 0.000027 time 0.9207 (1.0059) model_time 0.9206 (1.0044) loss 0.7118 (0.9193) grad_norm 6.5862 (8.7173/1.8788) mem 68106MB [2022-12-19 17:32:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][370/1519] eta 0:19:15 lr 0.000027 time 0.9244 (1.0056) model_time 0.9243 (1.0042) loss 0.9510 (0.9173) grad_norm 8.8876 (8.7058/1.8552) mem 68106MB [2022-12-19 17:32:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][380/1519] eta 0:19:06 lr 0.000027 time 0.9229 (1.0063) model_time 0.9228 (1.0049) loss 1.0058 (0.9178) grad_norm 11.0932 (8.7184/1.8412) mem 68106MB [2022-12-19 17:32:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][390/1519] eta 0:18:56 lr 0.000027 time 0.9208 (1.0063) model_time 0.9207 (1.0048) loss 0.9378 (0.9153) grad_norm 8.7567 (8.7519/1.8407) mem 68106MB [2022-12-19 17:33:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][400/1519] eta 0:18:45 lr 0.000027 time 0.9206 (1.0060) model_time 0.9204 (1.0047) loss 0.9243 (0.9161) grad_norm 10.4014 (8.7788/1.8328) mem 68106MB [2022-12-19 17:33:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][410/1519] eta 0:18:35 lr 0.000027 time 0.9105 (1.0060) model_time 0.9104 (1.0046) loss 0.9927 (0.9147) grad_norm 8.1586 (8.7628/1.8368) mem 68106MB [2022-12-19 17:33:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][420/1519] eta 0:18:25 lr 0.000027 time 0.9006 (1.0059) model_time 0.9005 (1.0046) loss 0.7087 (0.9135) grad_norm 10.2637 (8.7576/1.8282) mem 68106MB [2022-12-19 17:33:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][430/1519] eta 0:18:15 lr 0.000027 time 0.9335 (1.0061) model_time 0.9332 (1.0048) loss 0.8431 (0.9120) grad_norm 7.4672 (8.9239/3.1581) mem 68106MB [2022-12-19 17:33:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][440/1519] eta 0:18:05 lr 0.000027 time 0.9248 (1.0060) model_time 0.9245 (1.0047) loss 0.7166 (0.9116) grad_norm 6.7464 (8.8871/3.1368) mem 68106MB [2022-12-19 17:33:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][450/1519] eta 0:17:55 lr 0.000027 time 0.9311 (1.0059) model_time 0.9310 (1.0046) loss 0.7981 (0.9101) grad_norm 8.6085 (8.8791/3.1168) mem 68106MB [2022-12-19 17:34:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][460/1519] eta 0:17:45 lr 0.000027 time 0.9324 (1.0057) model_time 0.9323 (1.0045) loss 0.7482 (0.9093) grad_norm 8.9152 (8.8821/3.0855) mem 68106MB [2022-12-19 17:34:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][470/1519] eta 0:17:35 lr 0.000027 time 0.9175 (1.0062) model_time 0.9173 (1.0049) loss 0.7325 (0.9100) grad_norm 6.9621 (8.8393/3.0665) mem 68106MB [2022-12-19 17:34:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][480/1519] eta 0:17:25 lr 0.000027 time 0.9250 (1.0061) model_time 0.9249 (1.0049) loss 0.7235 (0.9071) grad_norm 7.0862 (8.8291/3.0385) mem 68106MB [2022-12-19 17:34:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][490/1519] eta 0:17:15 lr 0.000027 time 0.9289 (1.0063) model_time 0.9287 (1.0051) loss 0.6910 (0.9061) grad_norm 8.3507 (8.8108/3.0134) mem 68106MB [2022-12-19 17:34:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][500/1519] eta 0:17:05 lr 0.000027 time 0.9212 (1.0064) model_time 0.9210 (1.0052) loss 0.8290 (0.9045) grad_norm 7.0276 (8.7878/2.9920) mem 68106MB [2022-12-19 17:34:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][510/1519] eta 0:16:55 lr 0.000027 time 1.0244 (1.0065) model_time 1.0242 (1.0053) loss 1.1861 (0.9068) grad_norm 10.5636 (8.8079/2.9717) mem 68106MB [2022-12-19 17:35:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][520/1519] eta 0:16:45 lr 0.000027 time 0.9326 (1.0065) model_time 0.9324 (1.0053) loss 0.8740 (0.9062) grad_norm 9.1234 (8.7950/2.9486) mem 68106MB [2022-12-19 17:35:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][530/1519] eta 0:16:35 lr 0.000027 time 0.9295 (1.0064) model_time 0.9293 (1.0053) loss 1.3128 (0.9074) grad_norm 9.2234 (8.7838/2.9261) mem 68106MB [2022-12-19 17:35:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][540/1519] eta 0:16:25 lr 0.000027 time 0.9385 (1.0064) model_time 0.9383 (1.0053) loss 1.0587 (0.9072) grad_norm 6.9815 (8.7716/2.9014) mem 68106MB [2022-12-19 17:35:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][550/1519] eta 0:16:15 lr 0.000027 time 0.9268 (1.0063) model_time 0.9267 (1.0052) loss 0.8125 (0.9072) grad_norm 8.7383 (8.7649/2.8787) mem 68106MB [2022-12-19 17:35:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][560/1519] eta 0:16:04 lr 0.000027 time 0.9419 (1.0062) model_time 0.9417 (1.0051) loss 0.6783 (0.9073) grad_norm 8.4138 (8.7616/2.8578) mem 68106MB [2022-12-19 17:35:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][570/1519] eta 0:15:55 lr 0.000027 time 0.9391 (1.0064) model_time 0.9390 (1.0053) loss 0.7400 (0.9081) grad_norm 8.5449 (8.7503/2.8367) mem 68106MB [2022-12-19 17:36:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][580/1519] eta 0:15:45 lr 0.000027 time 0.9260 (1.0064) model_time 0.9257 (1.0054) loss 1.1554 (0.9072) grad_norm 9.1849 (8.7492/2.8189) mem 68106MB [2022-12-19 17:36:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][590/1519] eta 0:15:34 lr 0.000027 time 0.9331 (1.0063) model_time 0.9329 (1.0053) loss 0.7401 (0.9059) grad_norm 7.3900 (8.7465/2.8032) mem 68106MB [2022-12-19 17:36:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][600/1519] eta 0:15:25 lr 0.000027 time 0.9545 (1.0065) model_time 0.9543 (1.0055) loss 0.9673 (0.9058) grad_norm 6.0605 (8.7391/2.7900) mem 68106MB [2022-12-19 17:36:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][610/1519] eta 0:15:14 lr 0.000027 time 0.9231 (1.0065) model_time 0.9230 (1.0055) loss 0.8117 (0.9059) grad_norm 8.1638 (8.7391/2.7886) mem 68106MB [2022-12-19 17:36:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][620/1519] eta 0:15:04 lr 0.000027 time 0.9153 (1.0067) model_time 0.9152 (1.0056) loss 1.0212 (0.9065) grad_norm 7.8469 (8.7457/2.7959) mem 68106MB [2022-12-19 17:36:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][630/1519] eta 0:14:54 lr 0.000027 time 0.9232 (1.0066) model_time 0.9231 (1.0055) loss 0.7122 (0.9054) grad_norm 6.4505 (8.7161/2.8020) mem 68106MB [2022-12-19 17:37:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][640/1519] eta 0:14:44 lr 0.000027 time 0.9319 (1.0064) model_time 0.9317 (1.0054) loss 0.8857 (0.9049) grad_norm 9.5515 (8.7199/2.7952) mem 68106MB [2022-12-19 17:37:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][650/1519] eta 0:14:34 lr 0.000027 time 0.9301 (1.0064) model_time 0.9299 (1.0054) loss 0.8375 (0.9058) grad_norm 6.1449 (8.6998/2.7980) mem 68106MB [2022-12-19 17:37:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][660/1519] eta 0:14:24 lr 0.000027 time 0.9911 (1.0065) model_time 0.9910 (1.0055) loss 1.0961 (0.9058) grad_norm 9.3735 (8.7087/2.7935) mem 68106MB [2022-12-19 17:37:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][670/1519] eta 0:14:14 lr 0.000027 time 0.9318 (1.0063) model_time 0.9316 (1.0054) loss 0.9042 (0.9053) grad_norm 8.1434 (8.7298/2.7952) mem 68106MB [2022-12-19 17:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][680/1519] eta 0:14:04 lr 0.000027 time 0.9291 (1.0062) model_time 0.9289 (1.0052) loss 1.0539 (0.9061) grad_norm 6.3290 (8.7254/2.8009) mem 68106MB [2022-12-19 17:37:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][690/1519] eta 0:13:54 lr 0.000027 time 0.9914 (1.0063) model_time 0.9912 (1.0054) loss 1.0002 (0.9073) grad_norm 9.6469 (8.7352/2.7998) mem 68106MB [2022-12-19 17:38:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][700/1519] eta 0:13:44 lr 0.000027 time 0.9275 (1.0062) model_time 0.9273 (1.0053) loss 0.9448 (0.9071) grad_norm 8.7043 (8.7178/2.7960) mem 68106MB [2022-12-19 17:38:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][710/1519] eta 0:13:34 lr 0.000027 time 0.9313 (1.0062) model_time 0.9311 (1.0053) loss 1.0783 (0.9075) grad_norm 6.9051 (8.7062/2.7915) mem 68106MB [2022-12-19 17:38:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][720/1519] eta 0:13:23 lr 0.000027 time 0.9280 (1.0061) model_time 0.9278 (1.0051) loss 0.7488 (0.9075) grad_norm 7.0655 (8.7061/2.7909) mem 68106MB [2022-12-19 17:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][730/1519] eta 0:13:13 lr 0.000027 time 0.9276 (1.0061) model_time 0.9274 (1.0051) loss 0.8684 (0.9070) grad_norm 7.1203 (8.6793/2.7460) mem 68106MB [2022-12-19 17:38:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][740/1519] eta 0:13:03 lr 0.000027 time 0.9268 (1.0060) model_time 0.9266 (1.0051) loss 0.7918 (0.9068) grad_norm 6.5758 (8.6536/2.7421) mem 68106MB [2022-12-19 17:38:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][750/1519] eta 0:12:53 lr 0.000027 time 0.9260 (1.0060) model_time 0.9258 (1.0051) loss 0.8023 (0.9069) grad_norm 9.2591 (8.6536/2.7413) mem 68106MB [2022-12-19 17:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][760/1519] eta 0:12:43 lr 0.000027 time 0.9240 (1.0060) model_time 0.9238 (1.0051) loss 0.7500 (0.9057) grad_norm 6.4569 (8.6458/2.7364) mem 68106MB [2022-12-19 17:39:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][770/1519] eta 0:12:33 lr 0.000027 time 0.9304 (1.0059) model_time 0.9303 (1.0050) loss 0.6914 (0.9045) grad_norm 9.0248 (8.6201/2.7231) mem 68106MB [2022-12-19 17:39:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][780/1519] eta 0:12:23 lr 0.000027 time 0.9335 (1.0059) model_time 0.9333 (1.0050) loss 0.8901 (0.9045) grad_norm 7.7941 (8.5707/2.6954) mem 68106MB [2022-12-19 17:39:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][790/1519] eta 0:12:13 lr 0.000027 time 0.9419 (1.0059) model_time 0.9418 (1.0051) loss 0.7771 (0.9040) grad_norm 8.9435 (8.5700/2.6887) mem 68106MB [2022-12-19 17:39:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][800/1519] eta 0:12:03 lr 0.000027 time 0.9296 (1.0060) model_time 0.9294 (1.0051) loss 0.8158 (0.9041) grad_norm 9.1636 (8.5816/2.6936) mem 68106MB [2022-12-19 17:39:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][810/1519] eta 0:11:53 lr 0.000027 time 0.9268 (1.0061) model_time 0.9266 (1.0052) loss 0.9255 (0.9037) grad_norm 8.9912 (8.5766/2.6836) mem 68106MB [2022-12-19 17:40:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][820/1519] eta 0:11:43 lr 0.000027 time 0.9304 (1.0060) model_time 0.9302 (1.0051) loss 1.1037 (0.9039) grad_norm 6.4635 (8.5708/2.6823) mem 68106MB [2022-12-19 17:40:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][830/1519] eta 0:11:33 lr 0.000027 time 0.9316 (1.0060) model_time 0.9314 (1.0052) loss 0.8223 (0.9036) grad_norm 6.9839 (8.5514/2.6698) mem 68106MB [2022-12-19 17:40:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][840/1519] eta 0:11:23 lr 0.000027 time 0.9372 (1.0060) model_time 0.9369 (1.0051) loss 0.9232 (0.9036) grad_norm 9.3867 (8.5513/2.6631) mem 68106MB [2022-12-19 17:40:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][850/1519] eta 0:11:13 lr 0.000027 time 1.0621 (1.0061) model_time 1.0620 (1.0052) loss 1.0405 (0.9033) grad_norm 8.2100 (8.5357/2.6597) mem 68106MB [2022-12-19 17:40:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][860/1519] eta 0:11:02 lr 0.000027 time 0.9260 (1.0060) model_time 0.9258 (1.0051) loss 0.9894 (0.9035) grad_norm 7.3415 (8.5366/2.6568) mem 68106MB [2022-12-19 17:40:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][870/1519] eta 0:10:52 lr 0.000027 time 0.9286 (1.0059) model_time 0.9284 (1.0051) loss 0.7492 (0.9038) grad_norm 9.2175 (8.5615/2.6701) mem 68106MB [2022-12-19 17:41:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][880/1519] eta 0:10:42 lr 0.000027 time 0.9602 (1.0059) model_time 0.9601 (1.0051) loss 1.3210 (0.9039) grad_norm 5.3782 (8.5001/2.6541) mem 68106MB [2022-12-19 17:41:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][890/1519] eta 0:10:32 lr 0.000027 time 0.8839 (1.0061) model_time 0.8838 (1.0053) loss 0.8137 (0.9046) grad_norm 5.3837 (8.4992/2.6527) mem 68106MB [2022-12-19 17:41:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][900/1519] eta 0:10:22 lr 0.000027 time 0.9435 (1.0061) model_time 0.9432 (1.0052) loss 1.0998 (0.9048) grad_norm 7.1551 (8.4776/2.6451) mem 68106MB [2022-12-19 17:41:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][910/1519] eta 0:10:12 lr 0.000027 time 0.9304 (1.0061) model_time 0.9303 (1.0053) loss 0.6687 (0.9050) grad_norm 8.2426 (8.5181/2.6648) mem 68106MB [2022-12-19 17:41:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][920/1519] eta 0:10:02 lr 0.000027 time 0.9280 (1.0063) model_time 0.9278 (1.0054) loss 0.8584 (0.9056) grad_norm 6.3329 (8.4986/2.6651) mem 68106MB [2022-12-19 17:41:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][930/1519] eta 0:09:52 lr 0.000027 time 0.9395 (1.0062) model_time 0.9393 (1.0054) loss 0.8705 (0.9058) grad_norm 8.5145 (8.5259/2.6668) mem 68106MB [2022-12-19 17:42:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][940/1519] eta 0:09:42 lr 0.000027 time 0.9317 (1.0062) model_time 0.9315 (1.0053) loss 0.9315 (0.9057) grad_norm 7.4353 (8.5298/2.6721) mem 68106MB [2022-12-19 17:42:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][950/1519] eta 0:09:32 lr 0.000027 time 0.9320 (1.0061) model_time 0.9319 (1.0053) loss 0.7489 (0.9057) grad_norm 11.1577 (8.5474/2.6707) mem 68106MB [2022-12-19 17:42:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][960/1519] eta 0:09:22 lr 0.000027 time 0.9273 (1.0060) model_time 0.9271 (1.0052) loss 0.8016 (0.9047) grad_norm 6.8467 (8.5549/2.6755) mem 68106MB [2022-12-19 17:42:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][970/1519] eta 0:09:12 lr 0.000027 time 0.9413 (1.0060) model_time 0.9411 (1.0052) loss 0.8101 (0.9045) grad_norm 8.1061 (8.5688/2.6830) mem 68106MB [2022-12-19 17:42:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][980/1519] eta 0:09:02 lr 0.000027 time 0.9288 (1.0060) model_time 0.9286 (1.0052) loss 0.9275 (0.9046) grad_norm 7.9931 (8.5615/2.6807) mem 68106MB [2022-12-19 17:42:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][990/1519] eta 0:08:52 lr 0.000027 time 0.9320 (1.0059) model_time 0.9318 (1.0051) loss 1.3843 (0.9047) grad_norm 9.4634 (8.5838/2.7523) mem 68106MB [2022-12-19 17:43:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1000/1519] eta 0:08:42 lr 0.000027 time 0.9343 (1.0059) model_time 0.9341 (1.0051) loss 0.8297 (0.9046) grad_norm 7.4196 (8.5826/2.7568) mem 68106MB [2022-12-19 17:43:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1010/1519] eta 0:08:32 lr 0.000027 time 0.9292 (1.0059) model_time 0.9291 (1.0052) loss 0.9119 (0.9050) grad_norm 10.0158 (8.5873/2.7517) mem 68106MB [2022-12-19 17:43:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1020/1519] eta 0:08:21 lr 0.000027 time 0.9419 (1.0059) model_time 0.9418 (1.0051) loss 0.7365 (0.9043) grad_norm 5.6157 (8.5929/2.7640) mem 68106MB [2022-12-19 17:43:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1030/1519] eta 0:08:11 lr 0.000027 time 0.9984 (1.0059) model_time 0.9982 (1.0052) loss 0.9271 (0.9043) grad_norm 6.5861 (8.4688/1.6721) mem 68106MB [2022-12-19 17:43:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1040/1519] eta 0:08:01 lr 0.000027 time 0.9271 (1.0059) model_time 0.9269 (1.0051) loss 1.0349 (0.9048) grad_norm 9.0153 (8.5453/1.9699) mem 68106MB [2022-12-19 17:43:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1050/1519] eta 0:07:51 lr 0.000027 time 0.9277 (1.0058) model_time 0.9276 (1.0050) loss 0.9710 (0.9055) grad_norm 8.9912 (8.5726/1.9858) mem 68106MB [2022-12-19 17:44:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1060/1519] eta 0:07:41 lr 0.000027 time 0.9707 (1.0059) model_time 0.9701 (1.0051) loss 1.5310 (0.9067) grad_norm 10.9430 (8.5714/1.9921) mem 68106MB [2022-12-19 17:44:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1070/1519] eta 0:07:31 lr 0.000027 time 0.9293 (1.0058) model_time 0.9290 (1.0050) loss 0.8722 (0.9061) grad_norm 14.1712 (8.6111/2.0076) mem 68106MB [2022-12-19 17:44:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1080/1519] eta 0:07:21 lr 0.000027 time 0.9338 (1.0057) model_time 0.9337 (1.0050) loss 0.7247 (0.9056) grad_norm 6.9623 (8.6306/2.0306) mem 68106MB [2022-12-19 17:44:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1090/1519] eta 0:07:11 lr 0.000027 time 0.9264 (1.0059) model_time 0.9262 (1.0051) loss 0.6941 (0.9062) grad_norm 8.0890 (8.6206/2.0338) mem 68106MB [2022-12-19 17:44:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1100/1519] eta 0:07:01 lr 0.000027 time 0.9339 (1.0063) model_time 0.9338 (1.0055) loss 1.0181 (0.9059) grad_norm 11.0044 (8.6414/2.0403) mem 68106MB [2022-12-19 17:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1110/1519] eta 0:06:51 lr 0.000027 time 1.0122 (1.0063) model_time 1.0121 (1.0055) loss 1.2326 (0.9061) grad_norm 9.9383 (8.6126/2.0363) mem 68106MB [2022-12-19 17:45:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1120/1519] eta 0:06:41 lr 0.000027 time 0.9206 (1.0063) model_time 0.9202 (1.0056) loss 1.1066 (0.9075) grad_norm 9.7458 (8.6232/2.0362) mem 68106MB [2022-12-19 17:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1130/1519] eta 0:06:31 lr 0.000027 time 0.9300 (1.0063) model_time 0.9298 (1.0055) loss 0.7783 (0.9072) grad_norm 9.5955 (8.6258/2.0341) mem 68106MB [2022-12-19 17:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1140/1519] eta 0:06:21 lr 0.000027 time 0.9324 (1.0062) model_time 0.9322 (1.0055) loss 0.8262 (0.9069) grad_norm 6.5180 (8.6393/2.0399) mem 68106MB [2022-12-19 17:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1150/1519] eta 0:06:11 lr 0.000027 time 0.9212 (1.0062) model_time 0.9210 (1.0054) loss 1.1080 (0.9070) grad_norm 8.8319 (8.6591/2.0628) mem 68106MB [2022-12-19 17:45:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1160/1519] eta 0:06:01 lr 0.000027 time 0.9224 (1.0062) model_time 0.9223 (1.0055) loss 0.7887 (0.9071) grad_norm 11.6502 (8.6842/2.0781) mem 68106MB [2022-12-19 17:45:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1170/1519] eta 0:05:51 lr 0.000027 time 0.9273 (1.0062) model_time 0.9271 (1.0055) loss 0.7316 (0.9069) grad_norm 8.2885 (8.7007/2.0788) mem 68106MB [2022-12-19 17:46:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1180/1519] eta 0:05:41 lr 0.000027 time 0.9297 (1.0062) model_time 0.9295 (1.0055) loss 1.5462 (0.9074) grad_norm 15.3629 (8.7648/2.1545) mem 68106MB [2022-12-19 17:46:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1190/1519] eta 0:05:31 lr 0.000027 time 0.9686 (1.0064) model_time 0.9683 (1.0056) loss 0.9108 (0.9073) grad_norm 7.2974 (8.7765/2.1496) mem 68106MB [2022-12-19 17:46:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1200/1519] eta 0:05:21 lr 0.000027 time 0.9328 (1.0063) model_time 0.9327 (1.0056) loss 0.9065 (0.9078) grad_norm 10.7683 (8.8433/2.2556) mem 68106MB [2022-12-19 17:46:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1210/1519] eta 0:05:10 lr 0.000027 time 0.9302 (1.0063) model_time 0.9300 (1.0056) loss 1.3135 (0.9081) grad_norm 6.5295 (8.8557/2.2582) mem 68106MB [2022-12-19 17:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1220/1519] eta 0:05:00 lr 0.000027 time 0.9292 (1.0063) model_time 0.9291 (1.0056) loss 0.9822 (0.9089) grad_norm 8.8320 (8.8532/2.2510) mem 68106MB [2022-12-19 17:46:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1230/1519] eta 0:04:50 lr 0.000027 time 0.9595 (1.0063) model_time 0.9593 (1.0056) loss 0.7821 (0.9088) grad_norm 8.7400 (8.9018/2.2841) mem 68106MB [2022-12-19 17:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1240/1519] eta 0:04:40 lr 0.000027 time 0.9200 (1.0062) model_time 0.9198 (1.0055) loss 0.7904 (0.9087) grad_norm 8.6827 (8.8822/2.2838) mem 68106MB [2022-12-19 17:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1250/1519] eta 0:04:30 lr 0.000027 time 0.9261 (1.0062) model_time 0.9260 (1.0055) loss 0.7003 (0.9085) grad_norm 6.4831 (8.9039/2.2839) mem 68106MB [2022-12-19 17:47:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1260/1519] eta 0:04:20 lr 0.000027 time 0.9260 (1.0062) model_time 0.9257 (1.0055) loss 1.0458 (0.9085) grad_norm 8.7649 (8.8970/2.2826) mem 68106MB [2022-12-19 17:47:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1270/1519] eta 0:04:10 lr 0.000027 time 0.9193 (1.0061) model_time 0.9191 (1.0054) loss 0.8402 (0.9087) grad_norm 7.2596 (8.8746/2.2738) mem 68106MB [2022-12-19 17:47:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1280/1519] eta 0:04:00 lr 0.000027 time 0.9257 (1.0060) model_time 0.9255 (1.0053) loss 0.7197 (0.9086) grad_norm 7.5948 (8.8854/2.2658) mem 68106MB [2022-12-19 17:47:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1290/1519] eta 0:03:50 lr 0.000026 time 0.9897 (1.0060) model_time 0.9896 (1.0053) loss 0.8455 (0.9094) grad_norm 11.1799 (8.8972/2.2853) mem 68106MB [2022-12-19 17:48:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1300/1519] eta 0:03:40 lr 0.000026 time 0.9189 (1.0060) model_time 0.9187 (1.0053) loss 0.9157 (0.9090) grad_norm 8.2650 (8.8875/2.2945) mem 68106MB [2022-12-19 17:48:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1310/1519] eta 0:03:30 lr 0.000026 time 0.9376 (1.0060) model_time 0.9374 (1.0053) loss 0.7468 (0.9092) grad_norm 7.7320 (8.8994/2.2925) mem 68106MB [2022-12-19 17:48:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1320/1519] eta 0:03:20 lr 0.000026 time 0.9277 (1.0061) model_time 0.9276 (1.0054) loss 0.8743 (0.9089) grad_norm 7.5940 (8.9008/2.2913) mem 68106MB [2022-12-19 17:48:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1330/1519] eta 0:03:10 lr 0.000026 time 0.9186 (1.0060) model_time 0.9184 (1.0053) loss 0.7334 (0.9084) grad_norm 9.4548 (8.8996/2.2940) mem 68106MB [2022-12-19 17:48:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1340/1519] eta 0:03:00 lr 0.000026 time 0.9275 (1.0059) model_time 0.9273 (1.0052) loss 0.6845 (0.9081) grad_norm 10.1403 (8.9116/2.2932) mem 68106MB [2022-12-19 17:48:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1350/1519] eta 0:02:50 lr 0.000026 time 0.9310 (1.0059) model_time 0.9309 (1.0052) loss 0.7887 (0.9081) grad_norm 6.3030 (8.9337/2.3358) mem 68106MB [2022-12-19 17:49:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1360/1519] eta 0:02:39 lr 0.000026 time 0.9441 (1.0059) model_time 0.9439 (1.0052) loss 0.7001 (0.9079) grad_norm 10.3990 (8.9611/2.3350) mem 68106MB [2022-12-19 17:49:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1370/1519] eta 0:02:29 lr 0.000026 time 0.9804 (1.0059) model_time 0.9801 (1.0052) loss 0.6789 (0.9073) grad_norm 12.2632 (8.9886/2.3425) mem 68106MB [2022-12-19 17:49:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1380/1519] eta 0:02:19 lr 0.000026 time 0.9317 (1.0059) model_time 0.9316 (1.0052) loss 1.1284 (0.9072) grad_norm 7.8109 (9.0058/2.3435) mem 68106MB [2022-12-19 17:49:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1390/1519] eta 0:02:09 lr 0.000026 time 0.9298 (1.0061) model_time 0.9297 (1.0054) loss 0.9006 (0.9078) grad_norm 7.1921 (8.9983/2.3434) mem 68106MB [2022-12-19 17:49:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1400/1519] eta 0:01:59 lr 0.000026 time 0.9784 (1.0061) model_time 0.9783 (1.0054) loss 0.8486 (0.9069) grad_norm 8.3194 (8.9939/2.3399) mem 68106MB [2022-12-19 17:49:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1410/1519] eta 0:01:49 lr 0.000026 time 0.9655 (1.0061) model_time 0.9654 (1.0055) loss 0.6938 (0.9069) grad_norm 7.2011 (8.9823/2.3255) mem 68106MB [2022-12-19 17:50:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1420/1519] eta 0:01:39 lr 0.000026 time 0.9234 (1.0061) model_time 0.9233 (1.0054) loss 0.8624 (0.9061) grad_norm 6.4241 (8.9853/2.3302) mem 68106MB [2022-12-19 17:50:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1430/1519] eta 0:01:29 lr 0.000026 time 0.9233 (1.0061) model_time 0.9232 (1.0055) loss 0.6855 (0.9058) grad_norm 9.5288 (9.0206/2.3296) mem 68106MB [2022-12-19 17:50:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1440/1519] eta 0:01:19 lr 0.000026 time 0.9281 (1.0061) model_time 0.9280 (1.0055) loss 0.7848 (0.9062) grad_norm 11.5991 (9.0402/2.3354) mem 68106MB [2022-12-19 17:50:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1450/1519] eta 0:01:09 lr 0.000026 time 0.9240 (1.0061) model_time 0.9239 (1.0054) loss 1.3136 (0.9062) grad_norm 7.2563 (9.0765/2.4072) mem 68106MB [2022-12-19 17:50:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1460/1519] eta 0:00:59 lr 0.000026 time 0.9324 (1.0060) model_time 0.9323 (1.0054) loss 0.9279 (0.9066) grad_norm 8.8355 (9.1298/2.4609) mem 68106MB [2022-12-19 17:50:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1470/1519] eta 0:00:49 lr 0.000026 time 0.9205 (1.0060) model_time 0.9204 (1.0054) loss 0.8826 (0.9064) grad_norm 7.0882 (9.0989/2.4532) mem 68106MB [2022-12-19 17:51:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1480/1519] eta 0:00:39 lr 0.000026 time 0.9255 (1.0060) model_time 0.9253 (1.0053) loss 1.2855 (0.9069) grad_norm 8.8337 (9.1221/2.4398) mem 68106MB [2022-12-19 17:51:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1490/1519] eta 0:00:29 lr 0.000026 time 0.9218 (1.0059) model_time 0.9217 (1.0053) loss 0.9050 (0.9070) grad_norm 12.1049 (9.1460/2.4360) mem 68106MB [2022-12-19 17:51:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1500/1519] eta 0:00:19 lr 0.000026 time 0.9373 (1.0060) model_time 0.9371 (1.0053) loss 1.0917 (0.9067) grad_norm 8.5315 (9.1526/2.4338) mem 68106MB [2022-12-19 17:51:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [30/100][1510/1519] eta 0:00:09 lr 0.000026 time 0.9199 (1.0060) model_time 0.9198 (1.0053) loss 1.1068 (0.9068) grad_norm 7.7213 (9.1251/2.4219) mem 68106MB [2022-12-19 17:51:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 30 training takes 0:25:28 [2022-12-19 17:51:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_30.pth saving...... [2022-12-19 17:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_30.pth saved !!! [2022-12-19 17:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.699 (0.699) Loss 0.5110 (0.5110) Acc@1 89.931 (89.931) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 17:52:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.334) Loss 0.5204 (0.4869) Acc@1 90.972 (91.635) Acc@5 98.264 (98.359) Mem 68106MB [2022-12-19 17:52:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.316) Loss 0.4410 (0.4879) Acc@1 92.708 (91.435) Acc@5 98.958 (98.247) Mem 68106MB [2022-12-19 17:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.310) Loss 0.5933 (0.4904) Acc@1 88.889 (91.409) Acc@5 97.569 (98.230) Mem 68106MB [2022-12-19 17:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.308) Loss 0.4376 (0.4830) Acc@1 91.667 (91.472) Acc@5 98.958 (98.315) Mem 68106MB [2022-12-19 17:52:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.306) Loss 0.4867 (0.4806) Acc@1 88.889 (91.374) Acc@5 99.306 (98.346) Mem 68106MB [2022-12-19 17:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 0.5118 (0.4800) Acc@1 89.931 (91.433) Acc@5 97.917 (98.355) Mem 68106MB [2022-12-19 17:52:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5394 (0.4826) Acc@1 91.319 (91.398) Acc@5 98.611 (98.332) Mem 68106MB [2022-12-19 17:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4221 (0.4808) Acc@1 92.014 (91.414) Acc@5 98.611 (98.371) Mem 68106MB [2022-12-19 17:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:30] * Acc@1 91.405 Acc@5 98.375 [2022-12-19 17:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.4% [2022-12-19 17:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.40% [2022-12-19 17:52:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][0/1519] eta 0:48:06 lr 0.000026 time 1.9003 (1.9003) model_time 1.1831 (1.1831) loss 1.1204 (1.1204) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 17:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][10/1519] eta 0:27:15 lr 0.000026 time 0.9253 (1.0838) model_time 0.9251 (1.0183) loss 0.8457 (0.9808) grad_norm 11.7023 (9.5704/1.2531) mem 68106MB [2022-12-19 17:53:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][20/1519] eta 0:26:09 lr 0.000026 time 0.9256 (1.0469) model_time 0.9255 (1.0125) loss 1.0198 (0.9347) grad_norm 10.5699 (8.9958/1.5100) mem 68106MB [2022-12-19 17:53:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][30/1519] eta 0:25:34 lr 0.000026 time 0.9191 (1.0307) model_time 0.9190 (1.0073) loss 0.7468 (0.8931) grad_norm 7.2549 (8.4419/1.5854) mem 68106MB [2022-12-19 17:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][40/1519] eta 0:25:12 lr 0.000026 time 0.9343 (1.0228) model_time 0.9341 (1.0049) loss 0.8826 (0.8855) grad_norm 7.2498 (8.4258/1.4495) mem 68106MB [2022-12-19 17:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][50/1519] eta 0:24:55 lr 0.000026 time 0.9303 (1.0179) model_time 0.9300 (1.0035) loss 0.8334 (0.8706) grad_norm 8.1920 (8.2579/1.4030) mem 68106MB [2022-12-19 17:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][60/1519] eta 0:24:40 lr 0.000026 time 0.9235 (1.0148) model_time 0.9233 (1.0027) loss 1.0295 (0.8651) grad_norm 8.0136 (8.1553/1.3661) mem 68106MB [2022-12-19 17:53:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][70/1519] eta 0:24:27 lr 0.000026 time 0.9247 (1.0129) model_time 0.9246 (1.0025) loss 0.9002 (0.8830) grad_norm 7.9607 (8.0952/1.3220) mem 68106MB [2022-12-19 17:54:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][80/1519] eta 0:24:16 lr 0.000026 time 0.9243 (1.0119) model_time 0.9241 (1.0027) loss 0.9492 (0.8872) grad_norm 10.3651 (8.1009/1.3072) mem 68106MB [2022-12-19 17:54:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][90/1519] eta 0:24:05 lr 0.000026 time 1.0233 (1.0119) model_time 1.0232 (1.0036) loss 0.8675 (0.8919) grad_norm 8.0390 (8.1357/1.2805) mem 68106MB [2022-12-19 17:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][100/1519] eta 0:23:54 lr 0.000026 time 0.9200 (1.0111) model_time 0.9198 (1.0036) loss 1.0088 (0.8938) grad_norm 9.5470 (8.1462/1.3358) mem 68106MB [2022-12-19 17:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][110/1519] eta 0:23:44 lr 0.000026 time 0.9625 (1.0110) model_time 0.9623 (1.0042) loss 0.9074 (0.9039) grad_norm 7.4239 (8.2219/1.5194) mem 68106MB [2022-12-19 17:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][120/1519] eta 0:23:33 lr 0.000026 time 0.9241 (1.0105) model_time 0.9239 (1.0042) loss 0.8923 (0.8986) grad_norm 7.7825 (8.2757/1.5540) mem 68106MB [2022-12-19 17:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][130/1519] eta 0:23:22 lr 0.000026 time 0.9205 (1.0095) model_time 0.9203 (1.0037) loss 0.8955 (0.8924) grad_norm 8.8998 (8.3327/1.5391) mem 68106MB [2022-12-19 17:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][140/1519] eta 0:23:11 lr 0.000026 time 0.9228 (1.0090) model_time 0.9226 (1.0035) loss 1.0310 (0.8974) grad_norm 11.2524 (8.3066/1.5728) mem 68106MB [2022-12-19 17:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][150/1519] eta 0:23:01 lr 0.000026 time 1.0001 (1.0093) model_time 0.9999 (1.0043) loss 0.9216 (0.8998) grad_norm 8.6288 (8.3229/1.5253) mem 68106MB [2022-12-19 17:55:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][160/1519] eta 0:22:50 lr 0.000026 time 0.9238 (1.0087) model_time 0.9236 (1.0039) loss 0.9129 (0.9021) grad_norm 6.7384 (8.3266/1.6244) mem 68106MB [2022-12-19 17:55:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][170/1519] eta 0:22:41 lr 0.000026 time 0.9333 (1.0091) model_time 0.9332 (1.0046) loss 1.1143 (0.9023) grad_norm 7.1692 (8.2761/1.6072) mem 68106MB [2022-12-19 17:55:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][180/1519] eta 0:22:30 lr 0.000026 time 0.9261 (1.0085) model_time 0.9260 (1.0042) loss 0.7387 (0.9015) grad_norm 7.3234 (8.2798/1.6395) mem 68106MB [2022-12-19 17:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][190/1519] eta 0:22:19 lr 0.000026 time 0.9238 (1.0080) model_time 0.9237 (1.0039) loss 0.9589 (0.8967) grad_norm 8.3268 (8.2470/1.6173) mem 68106MB [2022-12-19 17:56:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][200/1519] eta 0:22:09 lr 0.000026 time 0.9331 (1.0078) model_time 0.9329 (1.0039) loss 0.7598 (0.8940) grad_norm 8.8575 (8.2528/1.6177) mem 68106MB [2022-12-19 17:56:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][210/1519] eta 0:22:00 lr 0.000026 time 0.9204 (1.0086) model_time 0.9203 (1.0048) loss 1.1769 (0.8937) grad_norm 8.2392 (8.2688/1.5913) mem 68106MB [2022-12-19 17:56:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][220/1519] eta 0:21:49 lr 0.000026 time 0.9342 (1.0084) model_time 0.9341 (1.0048) loss 1.0689 (0.8953) grad_norm 9.9419 (8.3386/1.8944) mem 68106MB [2022-12-19 17:56:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][230/1519] eta 0:21:39 lr 0.000026 time 0.9436 (1.0081) model_time 0.9435 (1.0046) loss 1.1160 (0.8956) grad_norm 6.2108 (8.3460/1.8864) mem 68106MB [2022-12-19 17:56:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][240/1519] eta 0:21:29 lr 0.000026 time 0.9270 (1.0080) model_time 0.9269 (1.0047) loss 0.9116 (0.8930) grad_norm 9.2079 (8.4326/1.9459) mem 68106MB [2022-12-19 17:56:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][250/1519] eta 0:21:18 lr 0.000026 time 0.9297 (1.0078) model_time 0.9296 (1.0046) loss 0.6741 (0.8915) grad_norm 6.7241 (8.4111/1.9395) mem 68106MB [2022-12-19 17:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][260/1519] eta 0:21:08 lr 0.000026 time 0.9418 (1.0076) model_time 0.9417 (1.0045) loss 1.0862 (0.8930) grad_norm 8.2676 (8.4394/1.9157) mem 68106MB [2022-12-19 17:57:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][270/1519] eta 0:20:58 lr 0.000026 time 0.9237 (1.0073) model_time 0.9236 (1.0043) loss 0.7195 (0.8897) grad_norm 8.8848 (8.3950/1.9155) mem 68106MB [2022-12-19 17:57:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][280/1519] eta 0:20:47 lr 0.000026 time 0.9338 (1.0071) model_time 0.9336 (1.0042) loss 0.8699 (0.8896) grad_norm 10.1363 (8.4444/1.9264) mem 68106MB [2022-12-19 17:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][290/1519] eta 0:20:37 lr 0.000026 time 0.9363 (1.0073) model_time 0.9361 (1.0045) loss 1.2494 (0.8932) grad_norm 11.3065 (8.4341/1.9198) mem 68106MB [2022-12-19 17:57:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][300/1519] eta 0:20:28 lr 0.000026 time 0.9839 (1.0074) model_time 0.9837 (1.0047) loss 0.9132 (0.8949) grad_norm 7.5206 (8.4323/1.8966) mem 68106MB [2022-12-19 17:57:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][310/1519] eta 0:20:17 lr 0.000026 time 0.9194 (1.0072) model_time 0.9193 (1.0045) loss 0.9722 (0.8964) grad_norm 8.1577 (8.4502/1.8926) mem 68106MB [2022-12-19 17:58:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][320/1519] eta 0:20:07 lr 0.000026 time 0.9664 (1.0072) model_time 0.9662 (1.0046) loss 1.0852 (0.8967) grad_norm 6.6569 (8.4596/1.8914) mem 68106MB [2022-12-19 17:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][330/1519] eta 0:19:57 lr 0.000026 time 0.9636 (1.0075) model_time 0.9635 (1.0050) loss 0.8078 (0.8948) grad_norm 7.5375 (8.4640/1.8811) mem 68106MB [2022-12-19 17:58:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][340/1519] eta 0:19:47 lr 0.000026 time 0.9228 (1.0073) model_time 0.9227 (1.0049) loss 0.9349 (0.8983) grad_norm 6.8416 (8.4660/1.8705) mem 68106MB [2022-12-19 17:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][350/1519] eta 0:19:37 lr 0.000026 time 0.9225 (1.0072) model_time 0.9224 (1.0049) loss 0.8220 (0.8974) grad_norm 7.6375 (8.4776/1.8851) mem 68106MB [2022-12-19 17:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][360/1519] eta 0:19:30 lr 0.000026 time 0.9177 (1.0098) model_time 0.9176 (1.0048) loss 0.8184 (0.8968) grad_norm 6.7750 (8.4885/1.9089) mem 68106MB [2022-12-19 17:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][370/1519] eta 0:19:19 lr 0.000026 time 0.9243 (1.0095) model_time 0.9240 (1.0046) loss 0.7911 (0.8964) grad_norm 12.0454 (8.5216/1.9315) mem 68106MB [2022-12-19 17:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][380/1519] eta 0:19:09 lr 0.000026 time 0.9281 (1.0095) model_time 0.9280 (1.0047) loss 0.7289 (0.8971) grad_norm 7.7278 (8.5201/1.9072) mem 68106MB [2022-12-19 17:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][390/1519] eta 0:18:59 lr 0.000026 time 0.9257 (1.0096) model_time 0.9256 (1.0049) loss 0.8542 (0.8983) grad_norm 8.7732 (8.5180/1.8872) mem 68106MB [2022-12-19 17:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][400/1519] eta 0:18:49 lr 0.000026 time 0.9224 (1.0095) model_time 0.9222 (1.0050) loss 1.1445 (0.9003) grad_norm 9.8645 (8.5316/1.8804) mem 68106MB [2022-12-19 17:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][410/1519] eta 0:18:39 lr 0.000026 time 0.9432 (1.0096) model_time 0.9430 (1.0052) loss 0.9124 (0.8997) grad_norm 10.1438 (8.5315/1.8638) mem 68106MB [2022-12-19 17:59:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][420/1519] eta 0:18:29 lr 0.000026 time 0.9264 (1.0093) model_time 0.9263 (1.0050) loss 0.7122 (0.8989) grad_norm 6.3724 (8.5210/1.8626) mem 68106MB [2022-12-19 17:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][430/1519] eta 0:18:19 lr 0.000026 time 0.9412 (1.0092) model_time 0.9411 (1.0050) loss 0.8642 (0.8978) grad_norm 9.0658 (8.5176/1.8459) mem 68106MB [2022-12-19 18:00:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][440/1519] eta 0:18:08 lr 0.000026 time 0.9271 (1.0093) model_time 0.9269 (1.0051) loss 0.8951 (0.8962) grad_norm 7.8691 (8.5823/1.9022) mem 68106MB [2022-12-19 18:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][450/1519] eta 0:17:58 lr 0.000026 time 0.9229 (1.0090) model_time 0.9227 (1.0050) loss 0.8026 (0.8955) grad_norm 10.3829 (8.6219/1.9039) mem 68106MB [2022-12-19 18:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][460/1519] eta 0:17:48 lr 0.000026 time 0.9219 (1.0090) model_time 0.9218 (1.0050) loss 0.7177 (0.8956) grad_norm 9.6756 (8.6474/1.9113) mem 68106MB [2022-12-19 18:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][470/1519] eta 0:17:39 lr 0.000026 time 0.9335 (1.0096) model_time 0.9334 (1.0057) loss 1.0480 (0.8969) grad_norm 7.3474 (8.6695/1.9220) mem 68106MB [2022-12-19 18:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][480/1519] eta 0:17:29 lr 0.000026 time 1.0260 (1.0098) model_time 1.0259 (1.0060) loss 0.8120 (0.8964) grad_norm 9.2742 (8.6477/1.9136) mem 68106MB [2022-12-19 18:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][490/1519] eta 0:17:18 lr 0.000026 time 0.9282 (1.0096) model_time 0.9280 (1.0058) loss 0.8177 (0.8947) grad_norm 6.0519 (8.6313/1.9237) mem 68106MB [2022-12-19 18:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][500/1519] eta 0:17:08 lr 0.000026 time 0.9824 (1.0095) model_time 0.9821 (1.0058) loss 1.1596 (0.8947) grad_norm 9.7339 (8.6426/1.9186) mem 68106MB [2022-12-19 18:01:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][510/1519] eta 0:16:58 lr 0.000026 time 0.9774 (1.0094) model_time 0.9772 (1.0057) loss 0.9222 (0.8938) grad_norm 7.8271 (8.6569/1.9377) mem 68106MB [2022-12-19 18:01:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][520/1519] eta 0:16:48 lr 0.000026 time 0.9290 (1.0091) model_time 0.9279 (1.0055) loss 0.8041 (0.8957) grad_norm 9.6375 (8.6772/1.9327) mem 68106MB [2022-12-19 18:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][530/1519] eta 0:16:37 lr 0.000026 time 0.9776 (1.0090) model_time 0.9774 (1.0055) loss 0.7805 (0.8969) grad_norm 6.9177 (8.6813/1.9320) mem 68106MB [2022-12-19 18:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][540/1519] eta 0:16:27 lr 0.000026 time 0.9319 (1.0088) model_time 0.9316 (1.0053) loss 0.8744 (0.8955) grad_norm 8.3281 (8.6546/1.9258) mem 68106MB [2022-12-19 18:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][550/1519] eta 0:16:17 lr 0.000026 time 0.9207 (1.0086) model_time 0.9206 (1.0052) loss 0.7598 (0.8950) grad_norm 6.5338 (8.6181/1.9295) mem 68106MB [2022-12-19 18:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][560/1519] eta 0:16:07 lr 0.000026 time 0.9329 (1.0085) model_time 0.9326 (1.0051) loss 1.1171 (0.8937) grad_norm 7.4191 (8.6009/1.9219) mem 68106MB [2022-12-19 18:02:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][570/1519] eta 0:15:56 lr 0.000026 time 0.9224 (1.0083) model_time 0.9222 (1.0050) loss 0.8883 (0.8923) grad_norm 11.6200 (8.6101/1.9227) mem 68106MB [2022-12-19 18:02:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][580/1519] eta 0:15:47 lr 0.000026 time 0.9268 (1.0086) model_time 0.9267 (1.0054) loss 1.0142 (0.8927) grad_norm 6.4118 (8.6108/1.9156) mem 68106MB [2022-12-19 18:02:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][590/1519] eta 0:15:37 lr 0.000026 time 0.9284 (1.0089) model_time 0.9282 (1.0057) loss 1.2180 (0.8920) grad_norm 10.3817 (8.6110/1.9038) mem 68106MB [2022-12-19 18:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][600/1519] eta 0:15:27 lr 0.000026 time 0.9219 (1.0087) model_time 0.9217 (1.0056) loss 0.8678 (0.8926) grad_norm 9.1813 (8.6062/1.8970) mem 68106MB [2022-12-19 18:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][610/1519] eta 0:15:17 lr 0.000026 time 0.9241 (1.0088) model_time 0.9240 (1.0057) loss 1.0307 (0.8919) grad_norm 9.3254 (8.6064/1.9028) mem 68106MB [2022-12-19 18:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][620/1519] eta 0:15:06 lr 0.000026 time 0.9286 (1.0087) model_time 0.9284 (1.0056) loss 0.9600 (0.8921) grad_norm 7.1665 (8.6096/1.8982) mem 68106MB [2022-12-19 18:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][630/1519] eta 0:14:56 lr 0.000026 time 0.9160 (1.0086) model_time 0.9159 (1.0056) loss 0.7875 (0.8907) grad_norm 8.3565 (8.6406/1.9026) mem 68106MB [2022-12-19 18:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][640/1519] eta 0:14:46 lr 0.000026 time 0.9307 (1.0085) model_time 0.9305 (1.0055) loss 0.8234 (0.8905) grad_norm 6.2037 (8.6295/1.9125) mem 68106MB [2022-12-19 18:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][650/1519] eta 0:14:36 lr 0.000026 time 0.9284 (1.0085) model_time 0.9272 (1.0055) loss 0.9762 (0.8915) grad_norm 7.3129 (8.6657/1.9730) mem 68106MB [2022-12-19 18:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][660/1519] eta 0:14:26 lr 0.000026 time 0.9028 (1.0085) model_time 0.9027 (1.0056) loss 0.8255 (0.8924) grad_norm 6.9958 (8.6948/1.9861) mem 68106MB [2022-12-19 18:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][670/1519] eta 0:14:16 lr 0.000026 time 0.9298 (1.0083) model_time 0.9296 (1.0055) loss 0.7980 (0.8934) grad_norm 8.6141 (8.6963/1.9853) mem 68106MB [2022-12-19 18:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][680/1519] eta 0:14:05 lr 0.000026 time 0.9609 (1.0082) model_time 0.9607 (1.0054) loss 0.6654 (0.8923) grad_norm 8.3812 (8.7081/1.9801) mem 68106MB [2022-12-19 18:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][690/1519] eta 0:13:55 lr 0.000026 time 0.9196 (1.0081) model_time 0.9194 (1.0053) loss 1.0877 (0.8927) grad_norm 9.5061 (8.7180/1.9914) mem 68106MB [2022-12-19 18:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][700/1519] eta 0:13:45 lr 0.000026 time 0.9203 (1.0082) model_time 0.9202 (1.0054) loss 0.9721 (0.8931) grad_norm 6.1632 (8.7227/2.0004) mem 68106MB [2022-12-19 18:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][710/1519] eta 0:13:35 lr 0.000026 time 0.9857 (1.0082) model_time 0.9855 (1.0055) loss 0.7669 (0.8923) grad_norm 8.6032 (8.7190/1.9780) mem 68106MB [2022-12-19 18:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][720/1519] eta 0:13:25 lr 0.000026 time 0.9204 (1.0080) model_time 0.9202 (1.0053) loss 0.9220 (0.8921) grad_norm 9.7998 (8.7506/2.0348) mem 68106MB [2022-12-19 18:04:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][730/1519] eta 0:13:15 lr 0.000026 time 0.9273 (1.0079) model_time 0.9271 (1.0052) loss 0.8817 (0.8918) grad_norm 9.8371 (8.7374/2.0340) mem 68106MB [2022-12-19 18:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][740/1519] eta 0:13:05 lr 0.000026 time 0.9331 (1.0078) model_time 0.9330 (1.0052) loss 0.7195 (0.8922) grad_norm 14.9918 (8.7781/2.0626) mem 68106MB [2022-12-19 18:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][750/1519] eta 0:12:55 lr 0.000026 time 0.9272 (1.0079) model_time 0.9271 (1.0053) loss 0.7772 (0.8915) grad_norm 10.1143 (8.7730/2.0696) mem 68106MB [2022-12-19 18:05:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][760/1519] eta 0:12:44 lr 0.000026 time 0.9253 (1.0078) model_time 0.9252 (1.0052) loss 0.9377 (0.8916) grad_norm 8.6814 (8.7767/2.0514) mem 68106MB [2022-12-19 18:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][770/1519] eta 0:12:34 lr 0.000026 time 0.9185 (1.0078) model_time 0.9184 (1.0052) loss 0.8474 (0.8917) grad_norm 5.3464 (8.7932/2.0655) mem 68106MB [2022-12-19 18:05:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][780/1519] eta 0:12:24 lr 0.000026 time 0.9239 (1.0078) model_time 0.9238 (1.0053) loss 0.8330 (0.8912) grad_norm 7.8873 (8.7952/2.0540) mem 68106MB [2022-12-19 18:05:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][790/1519] eta 0:12:14 lr 0.000026 time 0.9867 (1.0079) model_time 0.9865 (1.0054) loss 0.7642 (0.8903) grad_norm 10.6201 (8.8000/2.0635) mem 68106MB [2022-12-19 18:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][800/1519] eta 0:12:04 lr 0.000026 time 0.9252 (1.0081) model_time 0.9250 (1.0057) loss 0.9671 (0.8890) grad_norm 12.1231 (8.8201/2.0750) mem 68106MB [2022-12-19 18:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][810/1519] eta 0:11:54 lr 0.000026 time 0.9304 (1.0080) model_time 0.9303 (1.0056) loss 0.7608 (0.8888) grad_norm 8.4695 (8.8185/2.0763) mem 68106MB [2022-12-19 18:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][820/1519] eta 0:11:44 lr 0.000026 time 0.9234 (1.0080) model_time 0.9233 (1.0056) loss 0.7750 (0.8890) grad_norm 19.8234 (8.8409/2.0821) mem 68106MB [2022-12-19 18:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][830/1519] eta 0:11:34 lr 0.000026 time 0.9336 (1.0080) model_time 0.9335 (1.0056) loss 0.7363 (0.8882) grad_norm 9.9011 (8.8460/2.0731) mem 68106MB [2022-12-19 18:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][840/1519] eta 0:11:24 lr 0.000026 time 0.9301 (1.0079) model_time 0.9300 (1.0055) loss 1.0381 (0.8878) grad_norm 9.0296 (8.8290/2.0745) mem 68106MB [2022-12-19 18:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][850/1519] eta 0:11:14 lr 0.000026 time 0.9173 (1.0077) model_time 0.9172 (1.0054) loss 1.0773 (0.8873) grad_norm 8.5801 (8.8414/2.0663) mem 68106MB [2022-12-19 18:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][860/1519] eta 0:11:04 lr 0.000026 time 0.9199 (1.0076) model_time 0.9198 (1.0053) loss 1.0161 (0.8869) grad_norm 8.9164 (8.8282/2.0763) mem 68106MB [2022-12-19 18:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][870/1519] eta 0:10:53 lr 0.000026 time 0.9248 (1.0075) model_time 0.9247 (1.0053) loss 0.6968 (0.8869) grad_norm 8.2812 (8.8482/2.0656) mem 68106MB [2022-12-19 18:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][880/1519] eta 0:10:43 lr 0.000026 time 0.9217 (1.0074) model_time 0.9214 (1.0052) loss 0.7534 (0.8866) grad_norm 8.4215 (8.8425/2.0557) mem 68106MB [2022-12-19 18:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][890/1519] eta 0:10:33 lr 0.000026 time 0.9032 (1.0074) model_time 0.9031 (1.0052) loss 0.9583 (0.8861) grad_norm 10.7637 (8.8559/2.0520) mem 68106MB [2022-12-19 18:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][900/1519] eta 0:10:23 lr 0.000026 time 0.9316 (1.0074) model_time 0.9315 (1.0052) loss 0.7043 (0.8852) grad_norm 7.7954 (8.8953/2.1021) mem 68106MB [2022-12-19 18:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][910/1519] eta 0:10:13 lr 0.000026 time 0.9353 (1.0074) model_time 0.9352 (1.0052) loss 0.8281 (0.8854) grad_norm 9.2890 (8.8984/2.1132) mem 68106MB [2022-12-19 18:08:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][920/1519] eta 0:10:03 lr 0.000026 time 0.9198 (1.0075) model_time 0.9197 (1.0054) loss 0.7003 (0.8856) grad_norm 8.1008 (8.8850/2.1075) mem 68106MB [2022-12-19 18:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][930/1519] eta 0:09:53 lr 0.000026 time 0.9232 (1.0074) model_time 0.9231 (1.0053) loss 1.1044 (0.8853) grad_norm 7.0110 (8.8732/2.1111) mem 68106MB [2022-12-19 18:08:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][940/1519] eta 0:09:43 lr 0.000026 time 0.9346 (1.0074) model_time 0.9345 (1.0052) loss 1.0445 (0.8859) grad_norm 10.1208 (8.9022/2.1176) mem 68106MB [2022-12-19 18:08:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][950/1519] eta 0:09:33 lr 0.000026 time 0.9224 (1.0075) model_time 0.9223 (1.0054) loss 0.9638 (0.8865) grad_norm 8.8126 (8.8934/2.1069) mem 68106MB [2022-12-19 18:08:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][960/1519] eta 0:09:23 lr 0.000026 time 0.9269 (1.0074) model_time 0.9267 (1.0053) loss 0.9783 (0.8870) grad_norm 7.5496 (8.9008/2.0937) mem 68106MB [2022-12-19 18:08:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][970/1519] eta 0:09:13 lr 0.000026 time 0.9975 (1.0074) model_time 0.9974 (1.0054) loss 1.0556 (0.8879) grad_norm 8.4368 (8.8947/2.0913) mem 68106MB [2022-12-19 18:09:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][980/1519] eta 0:09:02 lr 0.000026 time 0.9257 (1.0073) model_time 0.9256 (1.0053) loss 0.8326 (0.8887) grad_norm 9.6333 (8.8989/2.1104) mem 68106MB [2022-12-19 18:09:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][990/1519] eta 0:08:52 lr 0.000026 time 0.9284 (1.0073) model_time 0.9283 (1.0052) loss 1.0310 (0.8881) grad_norm 5.7165 (8.9058/2.1227) mem 68106MB [2022-12-19 18:09:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1000/1519] eta 0:08:42 lr 0.000026 time 0.9236 (1.0072) model_time 0.9235 (1.0052) loss 1.2959 (0.8892) grad_norm 6.7911 (8.8961/2.1189) mem 68106MB [2022-12-19 18:09:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1010/1519] eta 0:08:32 lr 0.000026 time 0.9714 (1.0072) model_time 0.9712 (1.0052) loss 0.9201 (0.8893) grad_norm 6.6438 (8.8901/2.1356) mem 68106MB [2022-12-19 18:09:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1020/1519] eta 0:08:22 lr 0.000026 time 0.9235 (1.0072) model_time 0.9233 (1.0052) loss 0.7948 (0.8890) grad_norm 7.7145 (8.8959/2.1296) mem 68106MB [2022-12-19 18:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1030/1519] eta 0:08:12 lr 0.000026 time 0.9261 (1.0071) model_time 0.9259 (1.0052) loss 0.9058 (0.8892) grad_norm 8.8682 (8.9091/2.1289) mem 68106MB [2022-12-19 18:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1040/1519] eta 0:08:02 lr 0.000026 time 0.9307 (1.0071) model_time 0.9306 (1.0051) loss 0.8352 (0.8892) grad_norm 5.9300 (8.8510/2.1113) mem 68106MB [2022-12-19 18:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1050/1519] eta 0:07:52 lr 0.000026 time 0.9039 (1.0070) model_time 0.9037 (1.0051) loss 1.0750 (0.8903) grad_norm 8.1924 (8.8566/2.1283) mem 68106MB [2022-12-19 18:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1060/1519] eta 0:07:42 lr 0.000026 time 0.9298 (1.0070) model_time 0.9297 (1.0050) loss 0.8278 (0.8901) grad_norm 12.0303 (8.8473/2.1237) mem 68106MB [2022-12-19 18:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1070/1519] eta 0:07:32 lr 0.000026 time 0.9240 (1.0069) model_time 0.9239 (1.0050) loss 0.7104 (0.8896) grad_norm 8.3472 (8.8377/2.1058) mem 68106MB [2022-12-19 18:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1080/1519] eta 0:07:21 lr 0.000026 time 0.9240 (1.0068) model_time 0.9239 (1.0049) loss 1.0496 (0.8906) grad_norm 7.3796 (8.8395/2.1011) mem 68106MB [2022-12-19 18:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1090/1519] eta 0:07:11 lr 0.000026 time 0.9751 (1.0069) model_time 0.9749 (1.0050) loss 0.6745 (0.8909) grad_norm 8.9769 (8.8391/2.0872) mem 68106MB [2022-12-19 18:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1100/1519] eta 0:07:01 lr 0.000026 time 0.9704 (1.0069) model_time 0.9703 (1.0050) loss 0.8801 (0.8910) grad_norm 8.6693 (8.8196/2.0925) mem 68106MB [2022-12-19 18:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1110/1519] eta 0:06:51 lr 0.000026 time 0.9233 (1.0068) model_time 0.9232 (1.0050) loss 0.7390 (0.8904) grad_norm 6.5409 (8.7868/2.0729) mem 68106MB [2022-12-19 18:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1120/1519] eta 0:06:41 lr 0.000026 time 0.9304 (1.0068) model_time 0.9303 (1.0050) loss 0.7606 (0.8911) grad_norm 7.4571 (8.7530/2.0679) mem 68106MB [2022-12-19 18:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1130/1519] eta 0:06:31 lr 0.000026 time 0.9292 (1.0068) model_time 0.9291 (1.0050) loss 1.0084 (0.8915) grad_norm 8.9478 (8.7626/2.0619) mem 68106MB [2022-12-19 18:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1140/1519] eta 0:06:21 lr 0.000026 time 0.9268 (1.0068) model_time 0.9266 (1.0050) loss 0.7466 (0.8924) grad_norm 8.6118 (8.8050/2.0784) mem 68106MB [2022-12-19 18:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1150/1519] eta 0:06:11 lr 0.000026 time 0.9205 (1.0068) model_time 0.9204 (1.0050) loss 1.0457 (0.8926) grad_norm 10.7840 (8.8521/2.0640) mem 68106MB [2022-12-19 18:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1160/1519] eta 0:06:01 lr 0.000026 time 0.9235 (1.0067) model_time 0.9234 (1.0049) loss 1.3307 (0.8930) grad_norm 9.1016 (8.8634/2.0616) mem 68106MB [2022-12-19 18:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1170/1519] eta 0:05:51 lr 0.000026 time 0.9258 (1.0067) model_time 0.9257 (1.0049) loss 0.7426 (0.8928) grad_norm 7.6464 (8.8404/2.0544) mem 68106MB [2022-12-19 18:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1180/1519] eta 0:05:41 lr 0.000026 time 0.9201 (1.0066) model_time 0.9200 (1.0049) loss 0.9353 (0.8925) grad_norm 8.5074 (8.8473/2.0710) mem 68106MB [2022-12-19 18:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1190/1519] eta 0:05:31 lr 0.000026 time 0.9827 (1.0066) model_time 0.9826 (1.0049) loss 0.8409 (0.8931) grad_norm 7.7530 (8.8448/2.0795) mem 68106MB [2022-12-19 18:12:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1200/1519] eta 0:05:21 lr 0.000026 time 0.9283 (1.0065) model_time 0.9282 (1.0048) loss 0.7207 (0.8935) grad_norm 7.5500 (8.8725/2.1013) mem 68106MB [2022-12-19 18:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1210/1519] eta 0:05:11 lr 0.000026 time 0.9219 (1.0066) model_time 0.9218 (1.0049) loss 1.2009 (0.8939) grad_norm 8.7876 (8.8598/2.0936) mem 68106MB [2022-12-19 18:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1220/1519] eta 0:05:00 lr 0.000026 time 0.9391 (1.0066) model_time 0.9389 (1.0048) loss 0.7051 (0.8933) grad_norm 10.6004 (8.8731/2.0940) mem 68106MB [2022-12-19 18:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1230/1519] eta 0:04:50 lr 0.000026 time 0.9283 (1.0068) model_time 0.9282 (1.0051) loss 0.7436 (0.8931) grad_norm 9.5088 (8.8677/2.0894) mem 68106MB [2022-12-19 18:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1240/1519] eta 0:04:40 lr 0.000026 time 0.9339 (1.0069) model_time 0.9337 (1.0052) loss 1.3117 (0.8935) grad_norm 6.6733 (8.8487/2.0978) mem 68106MB [2022-12-19 18:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1250/1519] eta 0:04:30 lr 0.000026 time 0.9230 (1.0068) model_time 0.9228 (1.0051) loss 0.7612 (0.8934) grad_norm 7.8553 (8.8520/2.0536) mem 68106MB [2022-12-19 18:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1260/1519] eta 0:04:20 lr 0.000026 time 0.9280 (1.0068) model_time 0.9278 (1.0051) loss 0.9307 (0.8934) grad_norm 9.2139 (8.8959/2.0922) mem 68106MB [2022-12-19 18:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1270/1519] eta 0:04:10 lr 0.000026 time 0.9217 (1.0068) model_time 0.9216 (1.0052) loss 0.7472 (0.8927) grad_norm 11.0186 (8.9171/2.0971) mem 68106MB [2022-12-19 18:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1280/1519] eta 0:04:00 lr 0.000026 time 0.9263 (1.0068) model_time 0.9261 (1.0051) loss 0.9555 (0.8925) grad_norm 8.6046 (8.9182/2.1001) mem 68106MB [2022-12-19 18:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1290/1519] eta 0:03:50 lr 0.000026 time 0.9241 (1.0068) model_time 0.9239 (1.0052) loss 0.7210 (0.8923) grad_norm 7.6465 (8.9049/2.0943) mem 68106MB [2022-12-19 18:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1300/1519] eta 0:03:40 lr 0.000026 time 0.9315 (1.0068) model_time 0.9314 (1.0052) loss 0.9731 (0.8922) grad_norm 7.4437 (8.9235/2.0883) mem 68106MB [2022-12-19 18:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1310/1519] eta 0:03:30 lr 0.000026 time 0.9387 (1.0068) model_time 0.9386 (1.0051) loss 1.3120 (0.8931) grad_norm 9.0683 (8.9131/2.0884) mem 68106MB [2022-12-19 18:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1320/1519] eta 0:03:20 lr 0.000026 time 0.9252 (1.0067) model_time 0.9250 (1.0051) loss 0.8027 (0.8934) grad_norm 8.6590 (8.8867/2.0249) mem 68106MB [2022-12-19 18:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1330/1519] eta 0:03:10 lr 0.000026 time 0.9342 (1.0067) model_time 0.9341 (1.0051) loss 0.7457 (0.8928) grad_norm 8.1899 (8.8907/2.0270) mem 68106MB [2022-12-19 18:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1340/1519] eta 0:03:00 lr 0.000026 time 0.9291 (1.0067) model_time 0.9290 (1.0051) loss 1.0231 (0.8927) grad_norm 7.9183 (8.8475/1.9901) mem 68106MB [2022-12-19 18:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1350/1519] eta 0:02:50 lr 0.000026 time 0.9319 (1.0066) model_time 0.9318 (1.0050) loss 0.9468 (0.8930) grad_norm 8.6878 (8.8602/1.9863) mem 68106MB [2022-12-19 18:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1360/1519] eta 0:02:40 lr 0.000026 time 0.9276 (1.0066) model_time 0.9270 (1.0050) loss 0.7432 (0.8931) grad_norm 8.3873 (8.8573/1.9858) mem 68106MB [2022-12-19 18:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1370/1519] eta 0:02:29 lr 0.000026 time 0.9329 (1.0065) model_time 0.9327 (1.0049) loss 0.9771 (0.8932) grad_norm 8.4472 (8.8651/1.9782) mem 68106MB [2022-12-19 18:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1380/1519] eta 0:02:19 lr 0.000026 time 0.9278 (1.0064) model_time 0.9277 (1.0049) loss 0.8871 (0.8926) grad_norm 8.8213 (8.8742/1.9835) mem 68106MB [2022-12-19 18:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1390/1519] eta 0:02:09 lr 0.000026 time 0.9312 (1.0064) model_time 0.9311 (1.0048) loss 0.8343 (0.8925) grad_norm 7.5505 (8.8740/1.9772) mem 68106MB [2022-12-19 18:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1400/1519] eta 0:01:59 lr 0.000026 time 0.9317 (1.0063) model_time 0.9316 (1.0048) loss 0.7812 (0.8923) grad_norm 6.7805 (8.8575/1.9585) mem 68106MB [2022-12-19 18:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1410/1519] eta 0:01:49 lr 0.000026 time 0.9316 (1.0064) model_time 0.9315 (1.0049) loss 0.7089 (0.8922) grad_norm 12.0058 (8.8941/1.9781) mem 68106MB [2022-12-19 18:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1420/1519] eta 0:01:39 lr 0.000026 time 0.9301 (1.0064) model_time 0.9300 (1.0049) loss 0.7341 (0.8928) grad_norm 9.1101 (8.8903/1.9645) mem 68106MB [2022-12-19 18:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1430/1519] eta 0:01:29 lr 0.000026 time 0.9291 (1.0064) model_time 0.9290 (1.0048) loss 1.0752 (0.8930) grad_norm 8.6471 (8.8943/1.9638) mem 68106MB [2022-12-19 18:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1440/1519] eta 0:01:19 lr 0.000026 time 0.8919 (1.0065) model_time 0.8917 (1.0050) loss 1.0319 (0.8933) grad_norm 11.5192 (8.8762/1.9465) mem 68106MB [2022-12-19 18:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1450/1519] eta 0:01:09 lr 0.000026 time 0.9305 (1.0065) model_time 0.9303 (1.0050) loss 0.7082 (0.8934) grad_norm 7.9316 (8.8830/1.9710) mem 68106MB [2022-12-19 18:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1460/1519] eta 0:00:59 lr 0.000026 time 0.9233 (1.0064) model_time 0.9232 (1.0049) loss 1.1197 (0.8936) grad_norm 11.3905 (8.9071/1.9654) mem 68106MB [2022-12-19 18:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1470/1519] eta 0:00:49 lr 0.000026 time 0.9238 (1.0065) model_time 0.9237 (1.0050) loss 0.7450 (0.8935) grad_norm 10.2367 (8.9143/1.9704) mem 68106MB [2022-12-19 18:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1480/1519] eta 0:00:39 lr 0.000026 time 0.9268 (1.0065) model_time 0.9266 (1.0050) loss 0.7699 (0.8941) grad_norm 10.1111 (8.9299/2.0113) mem 68106MB [2022-12-19 18:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1490/1519] eta 0:00:29 lr 0.000026 time 0.9316 (1.0064) model_time 0.9313 (1.0049) loss 0.9744 (0.8941) grad_norm 9.5114 (8.9378/2.0128) mem 68106MB [2022-12-19 18:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1500/1519] eta 0:00:19 lr 0.000026 time 0.9359 (1.0064) model_time 0.9357 (1.0049) loss 0.6790 (0.8947) grad_norm 8.8618 (8.8976/1.9650) mem 68106MB [2022-12-19 18:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [31/100][1510/1519] eta 0:00:09 lr 0.000026 time 0.9254 (1.0064) model_time 0.9252 (1.0049) loss 0.7414 (0.8950) grad_norm 8.5342 (8.8871/1.9422) mem 68106MB [2022-12-19 18:18:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 31 training takes 0:25:28 [2022-12-19 18:18:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_31.pth saving...... [2022-12-19 18:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_31.pth saved !!! [2022-12-19 18:18:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.694 (0.694) Loss 0.5250 (0.5250) Acc@1 90.972 (90.972) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 18:18:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.335) Loss 0.5149 (0.4862) Acc@1 93.056 (91.888) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-19 18:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.317) Loss 0.4762 (0.4900) Acc@1 90.625 (91.485) Acc@5 98.958 (98.313) Mem 68106MB [2022-12-19 18:18:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.305 (0.312) Loss 0.5980 (0.4942) Acc@1 89.583 (91.353) Acc@5 97.569 (98.331) Mem 68106MB [2022-12-19 18:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.309) Loss 0.4433 (0.4867) Acc@1 91.319 (91.396) Acc@5 98.958 (98.374) Mem 68106MB [2022-12-19 18:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.294 (0.307) Loss 0.5148 (0.4848) Acc@1 88.889 (91.353) Acc@5 98.958 (98.386) Mem 68106MB [2022-12-19 18:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.305 (0.306) Loss 0.4916 (0.4847) Acc@1 89.236 (91.337) Acc@5 98.611 (98.423) Mem 68106MB [2022-12-19 18:18:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.305) Loss 0.5335 (0.4856) Acc@1 90.625 (91.285) Acc@5 98.264 (98.406) Mem 68106MB [2022-12-19 18:18:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.299 (0.304) Loss 0.4162 (0.4835) Acc@1 92.361 (91.324) Acc@5 98.611 (98.435) Mem 68106MB [2022-12-19 18:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:31] * Acc@1 91.278 Acc@5 98.441 [2022-12-19 18:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.3% [2022-12-19 18:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.40% [2022-12-19 18:19:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][0/1519] eta 0:47:53 lr 0.000026 time 1.8916 (1.8916) model_time 1.0407 (1.0407) loss 0.7949 (0.7949) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 18:19:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][10/1519] eta 0:27:25 lr 0.000026 time 0.9298 (1.0903) model_time 0.9297 (1.0126) loss 0.7950 (0.8885) grad_norm 8.4433 (7.9246/0.9609) mem 68106MB [2022-12-19 18:19:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][20/1519] eta 0:26:15 lr 0.000026 time 0.9243 (1.0512) model_time 0.9242 (1.0104) loss 0.8800 (0.8461) grad_norm 9.4959 (8.4733/1.9342) mem 68106MB [2022-12-19 18:19:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][30/1519] eta 0:25:42 lr 0.000026 time 0.9727 (1.0360) model_time 0.9724 (1.0083) loss 0.9420 (0.8562) grad_norm 9.3591 (8.5521/1.7217) mem 68106MB [2022-12-19 18:19:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][40/1519] eta 0:25:19 lr 0.000026 time 0.9219 (1.0271) model_time 0.9218 (1.0060) loss 0.7525 (0.8615) grad_norm 7.4008 (8.5955/1.7477) mem 68106MB [2022-12-19 18:19:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][50/1519] eta 0:25:00 lr 0.000026 time 0.9289 (1.0215) model_time 0.9288 (1.0044) loss 0.9505 (0.8598) grad_norm 8.1134 (8.6062/1.8334) mem 68106MB [2022-12-19 18:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][60/1519] eta 0:24:51 lr 0.000026 time 0.8880 (1.0223) model_time 0.8878 (1.0080) loss 1.0131 (0.8604) grad_norm 9.4949 (8.8700/1.8855) mem 68106MB [2022-12-19 18:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][70/1519] eta 0:24:38 lr 0.000026 time 0.9269 (1.0206) model_time 0.9268 (1.0083) loss 0.6992 (0.8619) grad_norm 11.0635 (8.8578/1.8491) mem 68106MB [2022-12-19 18:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][80/1519] eta 0:24:27 lr 0.000026 time 0.9234 (1.0196) model_time 0.9232 (1.0087) loss 0.8497 (0.8749) grad_norm 7.1915 (8.8332/1.8230) mem 68106MB [2022-12-19 18:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][90/1519] eta 0:24:13 lr 0.000026 time 0.9305 (1.0175) model_time 0.9299 (1.0078) loss 0.8922 (0.8756) grad_norm 8.7410 (8.7592/1.7704) mem 68106MB [2022-12-19 18:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][100/1519] eta 0:24:01 lr 0.000026 time 0.9181 (1.0157) model_time 0.9179 (1.0069) loss 1.0507 (0.8774) grad_norm 9.7014 (9.0273/2.4503) mem 68106MB [2022-12-19 18:20:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][110/1519] eta 0:23:49 lr 0.000026 time 0.9310 (1.0143) model_time 0.9308 (1.0063) loss 1.1065 (0.8808) grad_norm 7.6864 (9.0310/2.3487) mem 68106MB [2022-12-19 18:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][120/1519] eta 0:23:40 lr 0.000026 time 0.9311 (1.0152) model_time 0.9309 (1.0078) loss 1.2840 (0.8761) grad_norm 10.3557 (8.9940/2.3115) mem 68106MB [2022-12-19 18:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][130/1519] eta 0:23:29 lr 0.000026 time 0.9300 (1.0145) model_time 0.9296 (1.0077) loss 0.9060 (0.8766) grad_norm 5.8478 (8.9173/2.2756) mem 68106MB [2022-12-19 18:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][140/1519] eta 0:23:17 lr 0.000026 time 0.9053 (1.0137) model_time 0.9051 (1.0073) loss 0.8434 (0.8817) grad_norm 7.7183 (8.9813/2.2693) mem 68106MB [2022-12-19 18:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][150/1519] eta 0:23:06 lr 0.000026 time 0.9350 (1.0128) model_time 0.9348 (1.0068) loss 1.0174 (0.8857) grad_norm 7.3744 (8.9842/2.2882) mem 68106MB [2022-12-19 18:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][160/1519] eta 0:22:55 lr 0.000026 time 0.9336 (1.0120) model_time 0.9334 (1.0063) loss 0.8350 (0.8885) grad_norm 8.4413 (8.9683/2.2339) mem 68106MB [2022-12-19 18:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][170/1519] eta 0:22:44 lr 0.000026 time 0.9256 (1.0112) model_time 0.9255 (1.0059) loss 1.0553 (0.8833) grad_norm 7.4993 (8.8945/2.2214) mem 68106MB [2022-12-19 18:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][180/1519] eta 0:22:33 lr 0.000026 time 0.9290 (1.0111) model_time 0.9289 (1.0060) loss 0.7482 (0.8864) grad_norm 8.5558 (8.8588/2.1756) mem 68106MB [2022-12-19 18:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][190/1519] eta 0:22:23 lr 0.000026 time 0.9294 (1.0106) model_time 0.9292 (1.0058) loss 0.7170 (0.8880) grad_norm 7.6703 (8.7968/2.1418) mem 68106MB [2022-12-19 18:22:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][200/1519] eta 0:22:12 lr 0.000026 time 1.0199 (1.0105) model_time 1.0197 (1.0059) loss 1.2520 (0.8938) grad_norm 6.8589 (8.7846/2.1322) mem 68106MB [2022-12-19 18:22:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][210/1519] eta 0:22:01 lr 0.000026 time 0.9354 (1.0099) model_time 0.9353 (1.0055) loss 0.6913 (0.8894) grad_norm 8.2509 (8.8037/2.0958) mem 68106MB [2022-12-19 18:22:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][220/1519] eta 0:21:51 lr 0.000026 time 0.9274 (1.0093) model_time 0.9272 (1.0051) loss 0.8015 (0.8893) grad_norm 6.7539 (8.7584/2.1098) mem 68106MB [2022-12-19 18:22:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][230/1519] eta 0:21:40 lr 0.000026 time 0.9263 (1.0093) model_time 0.9262 (1.0052) loss 0.9183 (0.8894) grad_norm 6.5590 (8.7339/2.1527) mem 68106MB [2022-12-19 18:23:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][240/1519] eta 0:21:31 lr 0.000026 time 0.9941 (1.0095) model_time 0.9940 (1.0056) loss 0.6957 (0.8879) grad_norm 6.5416 (8.7298/2.1648) mem 68106MB [2022-12-19 18:23:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][250/1519] eta 0:21:20 lr 0.000026 time 0.9297 (1.0091) model_time 0.9295 (1.0053) loss 1.1688 (0.8887) grad_norm 6.9723 (8.7203/2.1315) mem 68106MB [2022-12-19 18:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][260/1519] eta 0:21:10 lr 0.000026 time 0.9181 (1.0091) model_time 0.9179 (1.0054) loss 0.9249 (0.8865) grad_norm 8.9847 (8.7053/2.1011) mem 68106MB [2022-12-19 18:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][270/1519] eta 0:20:59 lr 0.000026 time 0.9334 (1.0087) model_time 0.9333 (1.0051) loss 1.1194 (0.8854) grad_norm 10.3422 (8.7190/2.0903) mem 68106MB [2022-12-19 18:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][280/1519] eta 0:20:49 lr 0.000026 time 0.9298 (1.0083) model_time 0.9297 (1.0048) loss 0.8010 (0.8844) grad_norm 6.9495 (8.7444/2.1387) mem 68106MB [2022-12-19 18:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][290/1519] eta 0:20:38 lr 0.000026 time 0.9545 (1.0081) model_time 0.9542 (1.0048) loss 1.1279 (0.8833) grad_norm 8.1301 (8.7300/2.1063) mem 68106MB [2022-12-19 18:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][300/1519] eta 0:20:28 lr 0.000026 time 0.9289 (1.0079) model_time 0.9286 (1.0047) loss 0.6706 (0.8836) grad_norm 8.8061 (8.7634/2.1471) mem 68106MB [2022-12-19 18:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][310/1519] eta 0:20:18 lr 0.000026 time 0.9300 (1.0079) model_time 0.9298 (1.0047) loss 0.6891 (0.8827) grad_norm 9.2726 (8.7325/2.1417) mem 68106MB [2022-12-19 18:24:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][320/1519] eta 0:20:08 lr 0.000026 time 0.9218 (1.0081) model_time 0.9217 (1.0051) loss 0.7568 (0.8823) grad_norm 7.7973 (8.7048/2.1161) mem 68106MB [2022-12-19 18:24:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][330/1519] eta 0:19:58 lr 0.000026 time 0.9361 (1.0080) model_time 0.9360 (1.0050) loss 1.0600 (0.8839) grad_norm 8.5207 (8.7210/2.0895) mem 68106MB [2022-12-19 18:24:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][340/1519] eta 0:19:48 lr 0.000026 time 0.9316 (1.0078) model_time 0.9313 (1.0049) loss 0.9371 (0.8852) grad_norm 8.4351 (8.6858/2.0729) mem 68106MB [2022-12-19 18:24:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][350/1519] eta 0:19:38 lr 0.000026 time 0.9343 (1.0079) model_time 0.9341 (1.0050) loss 0.7614 (0.8842) grad_norm 9.3583 (8.7123/2.0743) mem 68106MB [2022-12-19 18:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][360/1519] eta 0:19:27 lr 0.000026 time 0.9282 (1.0077) model_time 0.9280 (1.0050) loss 0.9545 (0.8833) grad_norm 7.4844 (8.6787/2.0591) mem 68106MB [2022-12-19 18:25:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][370/1519] eta 0:19:17 lr 0.000026 time 0.9257 (1.0075) model_time 0.9255 (1.0048) loss 0.8472 (0.8830) grad_norm 6.8432 (8.6465/2.0443) mem 68106MB [2022-12-19 18:25:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][380/1519] eta 0:19:07 lr 0.000026 time 0.9346 (1.0076) model_time 0.9344 (1.0049) loss 0.9662 (0.8825) grad_norm 6.8458 (8.6363/2.0263) mem 68106MB [2022-12-19 18:25:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][390/1519] eta 0:18:57 lr 0.000026 time 0.9304 (1.0075) model_time 0.9302 (1.0049) loss 0.8216 (0.8845) grad_norm 6.7310 (8.6373/2.0316) mem 68106MB [2022-12-19 18:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][400/1519] eta 0:18:47 lr 0.000026 time 0.9211 (1.0072) model_time 0.9209 (1.0047) loss 0.9306 (0.8861) grad_norm 8.5663 (8.6326/2.0144) mem 68106MB [2022-12-19 18:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][410/1519] eta 0:18:36 lr 0.000026 time 0.9293 (1.0071) model_time 0.9291 (1.0046) loss 0.7876 (0.8861) grad_norm 6.7098 (8.6335/2.0121) mem 68106MB [2022-12-19 18:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][420/1519] eta 0:18:26 lr 0.000026 time 0.9893 (1.0071) model_time 0.9892 (1.0047) loss 0.7704 (0.8850) grad_norm 14.0407 (8.6606/2.0301) mem 68106MB [2022-12-19 18:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][430/1519] eta 0:18:16 lr 0.000026 time 0.9256 (1.0071) model_time 0.9255 (1.0047) loss 0.6934 (0.8860) grad_norm 12.9005 (8.6942/2.0392) mem 68106MB [2022-12-19 18:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][440/1519] eta 0:18:06 lr 0.000026 time 0.9624 (1.0070) model_time 0.9622 (1.0046) loss 0.8105 (0.8859) grad_norm 7.5359 (8.6992/2.0311) mem 68106MB [2022-12-19 18:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][450/1519] eta 0:17:56 lr 0.000026 time 0.9230 (1.0068) model_time 0.9228 (1.0045) loss 0.8248 (0.8875) grad_norm 6.4803 (8.6787/2.0230) mem 68106MB [2022-12-19 18:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][460/1519] eta 0:17:46 lr 0.000026 time 0.9260 (1.0068) model_time 0.9258 (1.0046) loss 0.6816 (0.8856) grad_norm 7.0945 (8.6548/2.0120) mem 68106MB [2022-12-19 18:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][470/1519] eta 0:17:35 lr 0.000026 time 0.9239 (1.0066) model_time 0.9238 (1.0044) loss 0.7718 (0.8857) grad_norm 6.6856 (8.6451/2.0056) mem 68106MB [2022-12-19 18:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][480/1519] eta 0:17:25 lr 0.000026 time 0.9430 (1.0067) model_time 0.9429 (1.0045) loss 0.8692 (0.8870) grad_norm 12.2563 (8.6653/2.0014) mem 68106MB [2022-12-19 18:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][490/1519] eta 0:17:15 lr 0.000026 time 0.9285 (1.0066) model_time 0.9284 (1.0045) loss 0.8943 (0.8877) grad_norm 7.8576 (8.6624/1.9883) mem 68106MB [2022-12-19 18:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][500/1519] eta 0:17:05 lr 0.000026 time 0.9580 (1.0065) model_time 0.9578 (1.0044) loss 1.0469 (0.8897) grad_norm 11.0673 (8.6720/1.9839) mem 68106MB [2022-12-19 18:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][510/1519] eta 0:16:55 lr 0.000026 time 0.9312 (1.0064) model_time 0.9311 (1.0043) loss 0.6816 (0.8888) grad_norm 6.3680 (8.6459/1.9756) mem 68106MB [2022-12-19 18:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][520/1519] eta 0:16:45 lr 0.000026 time 0.9323 (1.0064) model_time 0.9322 (1.0044) loss 0.8566 (0.8893) grad_norm 6.9582 (8.6250/1.9666) mem 68106MB [2022-12-19 18:27:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][530/1519] eta 0:16:35 lr 0.000026 time 0.9291 (1.0068) model_time 0.9290 (1.0048) loss 0.9910 (0.8912) grad_norm 7.7025 (8.5945/1.9638) mem 68106MB [2022-12-19 18:28:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][540/1519] eta 0:16:25 lr 0.000026 time 0.9885 (1.0067) model_time 0.9884 (1.0047) loss 0.9420 (0.8915) grad_norm 6.6938 (8.5710/1.9631) mem 68106MB [2022-12-19 18:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][550/1519] eta 0:16:15 lr 0.000026 time 0.9221 (1.0067) model_time 0.9219 (1.0048) loss 0.9834 (0.8917) grad_norm 8.9229 (8.5707/1.9501) mem 68106MB [2022-12-19 18:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][560/1519] eta 0:16:05 lr 0.000026 time 0.9279 (1.0067) model_time 0.9278 (1.0048) loss 0.6874 (0.8912) grad_norm 6.1863 (8.5538/1.9431) mem 68106MB [2022-12-19 18:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][570/1519] eta 0:15:55 lr 0.000026 time 0.9596 (1.0069) model_time 0.9594 (1.0050) loss 0.8028 (0.8901) grad_norm 7.3052 (8.5440/1.9291) mem 68106MB [2022-12-19 18:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][580/1519] eta 0:15:45 lr 0.000026 time 0.9264 (1.0068) model_time 0.9262 (1.0050) loss 0.7303 (0.8897) grad_norm 6.4141 (8.5341/1.9195) mem 68106MB [2022-12-19 18:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][590/1519] eta 0:15:35 lr 0.000026 time 0.9460 (1.0068) model_time 0.9458 (1.0049) loss 0.9813 (0.8916) grad_norm 7.5164 (8.5344/1.9158) mem 68106MB [2022-12-19 18:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][600/1519] eta 0:15:25 lr 0.000026 time 0.9235 (1.0067) model_time 0.9234 (1.0049) loss 0.9433 (0.8919) grad_norm 9.2338 (8.5483/1.9138) mem 68106MB [2022-12-19 18:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][610/1519] eta 0:15:15 lr 0.000026 time 0.9210 (1.0068) model_time 0.9208 (1.0050) loss 0.7566 (0.8905) grad_norm 8.0083 (8.5442/1.9160) mem 68106MB [2022-12-19 18:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][620/1519] eta 0:15:05 lr 0.000026 time 0.9591 (1.0067) model_time 0.9589 (1.0049) loss 0.7603 (0.8912) grad_norm 11.2044 (8.5642/1.9166) mem 68106MB [2022-12-19 18:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][630/1519] eta 0:14:54 lr 0.000026 time 0.9302 (1.0065) model_time 0.9300 (1.0048) loss 0.7928 (0.8900) grad_norm 7.7980 (8.5567/1.9346) mem 68106MB [2022-12-19 18:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][640/1519] eta 0:14:44 lr 0.000026 time 0.9657 (1.0067) model_time 0.9656 (1.0050) loss 0.9013 (0.8909) grad_norm 8.4087 (8.5725/1.9347) mem 68106MB [2022-12-19 18:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][650/1519] eta 0:14:34 lr 0.000026 time 0.9195 (1.0065) model_time 0.9193 (1.0048) loss 0.9467 (0.8901) grad_norm 7.1593 (8.5644/1.9239) mem 68106MB [2022-12-19 18:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][660/1519] eta 0:14:24 lr 0.000026 time 0.9310 (1.0064) model_time 0.9309 (1.0047) loss 0.8627 (0.8898) grad_norm 8.7655 (8.5337/1.9031) mem 68106MB [2022-12-19 18:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][670/1519] eta 0:14:14 lr 0.000026 time 0.9199 (1.0062) model_time 0.9198 (1.0046) loss 0.7475 (0.8895) grad_norm 8.6802 (8.5370/1.8981) mem 68106MB [2022-12-19 18:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][680/1519] eta 0:14:04 lr 0.000026 time 0.9231 (1.0061) model_time 0.9230 (1.0045) loss 0.6969 (0.8888) grad_norm 11.3954 (8.5464/1.8975) mem 68106MB [2022-12-19 18:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][690/1519] eta 0:13:54 lr 0.000026 time 0.9272 (1.0062) model_time 0.9271 (1.0046) loss 0.9205 (0.8888) grad_norm 7.5028 (8.5440/1.8958) mem 68106MB [2022-12-19 18:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][700/1519] eta 0:13:44 lr 0.000026 time 0.9830 (1.0064) model_time 0.9829 (1.0048) loss 0.7452 (0.8883) grad_norm 6.8675 (8.4816/1.7451) mem 68106MB [2022-12-19 18:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][710/1519] eta 0:13:34 lr 0.000026 time 0.9231 (1.0063) model_time 0.9229 (1.0047) loss 0.9338 (0.8888) grad_norm 8.0239 (8.4674/1.7425) mem 68106MB [2022-12-19 18:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][720/1519] eta 0:13:24 lr 0.000026 time 0.9960 (1.0063) model_time 0.9959 (1.0048) loss 0.8262 (0.8891) grad_norm 7.4941 (8.4556/1.7433) mem 68106MB [2022-12-19 18:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][730/1519] eta 0:13:13 lr 0.000026 time 0.9254 (1.0062) model_time 0.9253 (1.0047) loss 1.1234 (0.8894) grad_norm 7.2497 (8.4513/1.7390) mem 68106MB [2022-12-19 18:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][740/1519] eta 0:13:03 lr 0.000026 time 0.9222 (1.0063) model_time 0.9221 (1.0048) loss 1.5198 (0.8914) grad_norm 6.4167 (8.4272/1.7225) mem 68106MB [2022-12-19 18:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][750/1519] eta 0:12:53 lr 0.000026 time 0.9686 (1.0063) model_time 0.9685 (1.0048) loss 0.8327 (0.8915) grad_norm 8.4256 (8.4202/1.7051) mem 68106MB [2022-12-19 18:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][760/1519] eta 0:12:43 lr 0.000026 time 0.9228 (1.0063) model_time 0.9227 (1.0048) loss 0.8702 (0.8901) grad_norm 6.9097 (8.4051/1.7098) mem 68106MB [2022-12-19 18:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][770/1519] eta 0:12:33 lr 0.000026 time 0.9214 (1.0065) model_time 0.9212 (1.0050) loss 0.7485 (0.8907) grad_norm 6.8199 (8.4029/1.6991) mem 68106MB [2022-12-19 18:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][780/1519] eta 0:12:23 lr 0.000026 time 0.9280 (1.0064) model_time 0.9279 (1.0050) loss 0.7420 (0.8900) grad_norm 8.5332 (8.4451/1.7712) mem 68106MB [2022-12-19 18:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][790/1519] eta 0:12:13 lr 0.000026 time 0.9037 (1.0067) model_time 0.9035 (1.0053) loss 0.8407 (0.8906) grad_norm 9.6454 (8.4493/1.7742) mem 68106MB [2022-12-19 18:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][800/1519] eta 0:12:03 lr 0.000026 time 0.9194 (1.0067) model_time 0.9192 (1.0053) loss 1.0005 (0.8908) grad_norm 8.7428 (8.4614/1.7784) mem 68106MB [2022-12-19 18:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][810/1519] eta 0:11:53 lr 0.000026 time 0.9243 (1.0066) model_time 0.9241 (1.0052) loss 1.4833 (0.8909) grad_norm 8.6439 (8.4509/1.7792) mem 68106MB [2022-12-19 18:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][820/1519] eta 0:11:43 lr 0.000026 time 0.9294 (1.0066) model_time 0.9293 (1.0052) loss 1.0172 (0.8908) grad_norm 7.8280 (8.4481/1.7585) mem 68106MB [2022-12-19 18:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][830/1519] eta 0:11:33 lr 0.000026 time 0.9302 (1.0065) model_time 0.9300 (1.0051) loss 0.8902 (0.8911) grad_norm 9.8469 (8.4659/1.7380) mem 68106MB [2022-12-19 18:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][840/1519] eta 0:11:23 lr 0.000026 time 0.9253 (1.0065) model_time 0.9252 (1.0051) loss 0.9957 (0.8902) grad_norm 7.2027 (8.4496/1.7232) mem 68106MB [2022-12-19 18:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][850/1519] eta 0:11:13 lr 0.000026 time 0.9236 (1.0064) model_time 0.9235 (1.0051) loss 1.1569 (0.8907) grad_norm 8.2176 (8.4452/1.7295) mem 68106MB [2022-12-19 18:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][860/1519] eta 0:11:03 lr 0.000026 time 1.0196 (1.0065) model_time 1.0195 (1.0051) loss 1.0887 (0.8896) grad_norm 7.1645 (8.4525/1.7292) mem 68106MB [2022-12-19 18:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][870/1519] eta 0:10:53 lr 0.000026 time 0.9290 (1.0065) model_time 0.9289 (1.0052) loss 0.7374 (0.8885) grad_norm 7.2575 (8.4328/1.7196) mem 68106MB [2022-12-19 18:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][880/1519] eta 0:10:43 lr 0.000026 time 0.9288 (1.0065) model_time 0.9287 (1.0051) loss 1.1678 (0.8884) grad_norm 6.7168 (8.4058/1.6761) mem 68106MB [2022-12-19 18:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][890/1519] eta 0:10:33 lr 0.000026 time 0.9254 (1.0065) model_time 0.9252 (1.0051) loss 0.8936 (0.8884) grad_norm 7.5928 (8.4000/1.6905) mem 68106MB [2022-12-19 18:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][900/1519] eta 0:10:22 lr 0.000026 time 0.9769 (1.0064) model_time 0.9768 (1.0051) loss 0.9699 (0.8888) grad_norm 11.3448 (8.3793/1.6537) mem 68106MB [2022-12-19 18:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][910/1519] eta 0:10:12 lr 0.000026 time 0.9189 (1.0063) model_time 0.9187 (1.0050) loss 0.8711 (0.8896) grad_norm 8.7380 (8.3894/1.6485) mem 68106MB [2022-12-19 18:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][920/1519] eta 0:10:02 lr 0.000026 time 0.9392 (1.0063) model_time 0.9391 (1.0050) loss 0.7246 (0.8898) grad_norm 7.6507 (8.4116/1.6605) mem 68106MB [2022-12-19 18:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][930/1519] eta 0:09:52 lr 0.000026 time 0.9278 (1.0062) model_time 0.9277 (1.0049) loss 0.7962 (0.8893) grad_norm 7.6113 (8.3946/1.6584) mem 68106MB [2022-12-19 18:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][940/1519] eta 0:09:42 lr 0.000026 time 0.9269 (1.0063) model_time 0.9267 (1.0050) loss 0.9390 (0.8898) grad_norm 7.6425 (8.4419/1.6942) mem 68106MB [2022-12-19 18:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][950/1519] eta 0:09:32 lr 0.000026 time 0.9169 (1.0063) model_time 0.9168 (1.0050) loss 0.9151 (0.8905) grad_norm 9.6166 (8.4254/1.6758) mem 68106MB [2022-12-19 18:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][960/1519] eta 0:09:22 lr 0.000026 time 0.9285 (1.0063) model_time 0.9283 (1.0051) loss 0.9016 (0.8904) grad_norm 8.5438 (8.4312/1.6733) mem 68106MB [2022-12-19 18:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][970/1519] eta 0:09:12 lr 0.000026 time 0.9243 (1.0062) model_time 0.9241 (1.0050) loss 1.5692 (0.8924) grad_norm 11.1708 (8.4856/1.7039) mem 68106MB [2022-12-19 18:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][980/1519] eta 0:09:02 lr 0.000026 time 0.9279 (1.0062) model_time 0.9278 (1.0050) loss 0.8523 (0.8920) grad_norm 9.8022 (8.4833/1.7055) mem 68106MB [2022-12-19 18:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][990/1519] eta 0:08:52 lr 0.000026 time 0.9307 (1.0062) model_time 0.9305 (1.0049) loss 0.8254 (0.8930) grad_norm 7.2189 (8.4723/1.7089) mem 68106MB [2022-12-19 18:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1000/1519] eta 0:08:42 lr 0.000026 time 0.9213 (1.0061) model_time 0.9212 (1.0049) loss 0.7167 (0.8927) grad_norm 16.6221 (8.5120/1.7711) mem 68106MB [2022-12-19 18:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1010/1519] eta 0:08:32 lr 0.000026 time 0.9236 (1.0061) model_time 0.9235 (1.0049) loss 0.8805 (0.8938) grad_norm 8.4817 (8.5297/1.7844) mem 68106MB [2022-12-19 18:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1020/1519] eta 0:08:22 lr 0.000026 time 0.9328 (1.0062) model_time 0.9327 (1.0050) loss 1.0849 (0.8939) grad_norm 6.8507 (8.5076/1.7649) mem 68106MB [2022-12-19 18:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1030/1519] eta 0:08:11 lr 0.000026 time 0.9335 (1.0061) model_time 0.9333 (1.0049) loss 1.2099 (0.8941) grad_norm 7.2103 (8.4833/1.7395) mem 68106MB [2022-12-19 18:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1040/1519] eta 0:08:01 lr 0.000026 time 0.9235 (1.0061) model_time 0.9233 (1.0049) loss 0.8163 (0.8946) grad_norm 7.3194 (8.4795/1.7392) mem 68106MB [2022-12-19 18:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1050/1519] eta 0:07:51 lr 0.000026 time 0.9304 (1.0061) model_time 0.9302 (1.0049) loss 0.7880 (0.8939) grad_norm 7.8241 (8.4892/1.7390) mem 68106MB [2022-12-19 18:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1060/1519] eta 0:07:41 lr 0.000026 time 0.9265 (1.0060) model_time 0.9263 (1.0049) loss 1.0973 (0.8943) grad_norm 10.9220 (8.5036/1.7419) mem 68106MB [2022-12-19 18:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1070/1519] eta 0:07:31 lr 0.000026 time 0.9227 (1.0061) model_time 0.9225 (1.0049) loss 0.7599 (0.8937) grad_norm 7.8486 (8.4921/1.7356) mem 68106MB [2022-12-19 18:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1080/1519] eta 0:07:21 lr 0.000026 time 0.9251 (1.0060) model_time 0.9250 (1.0049) loss 0.7988 (0.8931) grad_norm 8.2423 (8.4720/1.7213) mem 68106MB [2022-12-19 18:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1090/1519] eta 0:07:11 lr 0.000026 time 0.9303 (1.0059) model_time 0.9301 (1.0048) loss 0.7000 (0.8935) grad_norm 7.8254 (8.4765/1.7304) mem 68106MB [2022-12-19 18:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1100/1519] eta 0:07:01 lr 0.000026 time 0.9365 (1.0060) model_time 0.9363 (1.0048) loss 1.2789 (0.8927) grad_norm 10.7172 (8.4779/1.7232) mem 68106MB [2022-12-19 18:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1110/1519] eta 0:06:51 lr 0.000026 time 0.9305 (1.0060) model_time 0.9303 (1.0049) loss 0.6882 (0.8926) grad_norm 8.5360 (8.5041/1.7169) mem 68106MB [2022-12-19 18:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1120/1519] eta 0:06:41 lr 0.000026 time 0.9214 (1.0060) model_time 0.9213 (1.0049) loss 0.7614 (0.8928) grad_norm 5.3671 (8.5056/1.7273) mem 68106MB [2022-12-19 18:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1130/1519] eta 0:06:31 lr 0.000026 time 0.9312 (1.0061) model_time 0.9310 (1.0049) loss 0.7636 (0.8928) grad_norm 6.4527 (8.5226/1.7245) mem 68106MB [2022-12-19 18:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1140/1519] eta 0:06:21 lr 0.000026 time 0.9207 (1.0063) model_time 0.9205 (1.0052) loss 1.1075 (0.8925) grad_norm 7.8113 (8.5345/1.7160) mem 68106MB [2022-12-19 18:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1150/1519] eta 0:06:11 lr 0.000026 time 0.9211 (1.0063) model_time 0.9210 (1.0052) loss 1.0278 (0.8919) grad_norm 10.8781 (8.5733/1.7470) mem 68106MB [2022-12-19 18:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1160/1519] eta 0:06:01 lr 0.000026 time 0.9235 (1.0062) model_time 0.9234 (1.0051) loss 0.6899 (0.8915) grad_norm 8.8479 (8.5827/1.7429) mem 68106MB [2022-12-19 18:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1170/1519] eta 0:05:51 lr 0.000026 time 0.9278 (1.0062) model_time 0.9277 (1.0051) loss 0.7234 (0.8914) grad_norm 7.3452 (8.5866/1.7495) mem 68106MB [2022-12-19 18:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1180/1519] eta 0:05:41 lr 0.000026 time 0.9617 (1.0063) model_time 0.9616 (1.0052) loss 0.8956 (0.8918) grad_norm 7.0408 (8.6219/1.7859) mem 68106MB [2022-12-19 18:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1190/1519] eta 0:05:31 lr 0.000026 time 0.9254 (1.0062) model_time 0.9253 (1.0051) loss 0.9082 (0.8919) grad_norm 7.4194 (8.6242/1.7848) mem 68106MB [2022-12-19 18:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1200/1519] eta 0:05:20 lr 0.000026 time 0.9220 (1.0063) model_time 0.9219 (1.0052) loss 0.6961 (0.8915) grad_norm 6.9396 (8.6077/1.8195) mem 68106MB [2022-12-19 18:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1210/1519] eta 0:05:10 lr 0.000026 time 0.9222 (1.0062) model_time 0.9220 (1.0051) loss 0.9831 (0.8917) grad_norm 7.3712 (8.6112/1.8154) mem 68106MB [2022-12-19 18:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1220/1519] eta 0:05:00 lr 0.000026 time 0.9329 (1.0061) model_time 0.9326 (1.0051) loss 1.2116 (0.8913) grad_norm 5.5921 (8.5640/1.8057) mem 68106MB [2022-12-19 18:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1230/1519] eta 0:04:50 lr 0.000026 time 0.9081 (1.0061) model_time 0.9079 (1.0051) loss 1.1133 (0.8921) grad_norm 8.8045 (8.5855/1.7945) mem 68106MB [2022-12-19 18:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1240/1519] eta 0:04:40 lr 0.000026 time 0.9287 (1.0061) model_time 0.9286 (1.0050) loss 0.6681 (0.8918) grad_norm 11.8411 (8.5975/1.8044) mem 68106MB [2022-12-19 18:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1250/1519] eta 0:04:30 lr 0.000026 time 0.9313 (1.0062) model_time 0.9312 (1.0052) loss 0.8283 (0.8914) grad_norm 13.4293 (8.6216/1.8241) mem 68106MB [2022-12-19 18:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1260/1519] eta 0:04:20 lr 0.000026 time 0.9259 (1.0064) model_time 0.9258 (1.0053) loss 0.6973 (0.8920) grad_norm 5.4084 (8.6019/1.8379) mem 68106MB [2022-12-19 18:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1270/1519] eta 0:04:10 lr 0.000026 time 0.9223 (1.0063) model_time 0.9222 (1.0053) loss 0.8264 (0.8913) grad_norm 8.6762 (8.5972/1.8373) mem 68106MB [2022-12-19 18:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1280/1519] eta 0:04:00 lr 0.000026 time 0.9309 (1.0062) model_time 0.9307 (1.0052) loss 0.6862 (0.8913) grad_norm 9.8238 (8.5964/1.8319) mem 68106MB [2022-12-19 18:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1290/1519] eta 0:03:50 lr 0.000026 time 0.9112 (1.0062) model_time 0.9110 (1.0052) loss 0.7209 (0.8913) grad_norm 5.5864 (8.5992/1.8454) mem 68106MB [2022-12-19 18:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1300/1519] eta 0:03:40 lr 0.000026 time 0.9206 (1.0062) model_time 0.9204 (1.0051) loss 1.1420 (0.8914) grad_norm 9.9005 (8.6104/1.8476) mem 68106MB [2022-12-19 18:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1310/1519] eta 0:03:30 lr 0.000026 time 0.9219 (1.0061) model_time 0.9218 (1.0051) loss 0.7861 (0.8908) grad_norm 8.3750 (8.6202/1.8736) mem 68106MB [2022-12-19 18:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1320/1519] eta 0:03:20 lr 0.000026 time 0.9235 (1.0061) model_time 0.9233 (1.0051) loss 0.8130 (0.8912) grad_norm 10.0417 (8.6310/1.8721) mem 68106MB [2022-12-19 18:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1330/1519] eta 0:03:10 lr 0.000026 time 0.9222 (1.0061) model_time 0.9221 (1.0051) loss 1.0209 (0.8911) grad_norm 13.0224 (8.6611/1.8904) mem 68106MB [2022-12-19 18:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1340/1519] eta 0:03:00 lr 0.000026 time 0.9314 (1.0061) model_time 0.9312 (1.0051) loss 0.8178 (0.8915) grad_norm 10.2637 (8.7134/1.9813) mem 68106MB [2022-12-19 18:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1350/1519] eta 0:02:50 lr 0.000026 time 0.9226 (1.0061) model_time 0.9224 (1.0051) loss 0.8940 (0.8911) grad_norm 11.1917 (8.7117/1.9787) mem 68106MB [2022-12-19 18:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1360/1519] eta 0:02:39 lr 0.000026 time 0.9279 (1.0062) model_time 0.9277 (1.0052) loss 0.9727 (0.8909) grad_norm 8.6873 (8.7277/1.9849) mem 68106MB [2022-12-19 18:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1370/1519] eta 0:02:29 lr 0.000026 time 0.9242 (1.0061) model_time 0.9241 (1.0052) loss 0.8511 (0.8912) grad_norm 9.6573 (8.7543/1.9847) mem 68106MB [2022-12-19 18:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1380/1519] eta 0:02:19 lr 0.000026 time 0.9194 (1.0062) model_time 0.9193 (1.0052) loss 0.8549 (0.8910) grad_norm 19.0272 (8.7423/2.0158) mem 68106MB [2022-12-19 18:42:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1390/1519] eta 0:02:09 lr 0.000026 time 0.9203 (1.0061) model_time 0.9202 (1.0052) loss 1.1285 (0.8911) grad_norm 7.9109 (8.7438/2.0101) mem 68106MB [2022-12-19 18:42:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1400/1519] eta 0:01:59 lr 0.000026 time 0.9212 (1.0061) model_time 0.9210 (1.0051) loss 0.7936 (0.8907) grad_norm 7.0580 (8.7802/2.0797) mem 68106MB [2022-12-19 18:42:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1410/1519] eta 0:01:49 lr 0.000026 time 0.9314 (1.0062) model_time 0.9312 (1.0052) loss 0.6829 (0.8902) grad_norm 9.0872 (8.7832/2.0778) mem 68106MB [2022-12-19 18:42:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1420/1519] eta 0:01:39 lr 0.000026 time 0.9218 (1.0062) model_time 0.9217 (1.0052) loss 0.7485 (0.8901) grad_norm 9.8428 (8.7989/2.0838) mem 68106MB [2022-12-19 18:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1430/1519] eta 0:01:29 lr 0.000026 time 0.9282 (1.0061) model_time 0.9280 (1.0052) loss 0.6971 (0.8900) grad_norm 7.0626 (8.7762/2.0749) mem 68106MB [2022-12-19 18:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1440/1519] eta 0:01:19 lr 0.000026 time 0.9214 (1.0062) model_time 0.9212 (1.0052) loss 0.9013 (0.8903) grad_norm 10.1022 (8.7951/2.0655) mem 68106MB [2022-12-19 18:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1450/1519] eta 0:01:09 lr 0.000026 time 0.9221 (1.0061) model_time 0.9220 (1.0052) loss 0.7437 (0.8903) grad_norm 6.2467 (8.7960/2.0686) mem 68106MB [2022-12-19 18:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1460/1519] eta 0:00:59 lr 0.000026 time 0.9734 (1.0061) model_time 0.9733 (1.0052) loss 0.8058 (0.8899) grad_norm 6.5902 (8.7877/2.0832) mem 68106MB [2022-12-19 18:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1470/1519] eta 0:00:49 lr 0.000026 time 0.9214 (1.0061) model_time 0.9212 (1.0052) loss 0.9308 (0.8901) grad_norm 7.1109 (8.8068/2.0807) mem 68106MB [2022-12-19 18:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1480/1519] eta 0:00:39 lr 0.000026 time 0.9225 (1.0060) model_time 0.9223 (1.0051) loss 1.1756 (0.8900) grad_norm 6.4391 (8.8530/2.1153) mem 68106MB [2022-12-19 18:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1490/1519] eta 0:00:29 lr 0.000026 time 0.8901 (1.0060) model_time 0.8900 (1.0051) loss 0.9784 (0.8904) grad_norm 8.9759 (8.8830/2.1728) mem 68106MB [2022-12-19 18:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1500/1519] eta 0:00:19 lr 0.000026 time 0.9208 (1.0060) model_time 0.9207 (1.0051) loss 0.9197 (0.8905) grad_norm 5.8961 (8.8848/2.1706) mem 68106MB [2022-12-19 18:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [32/100][1510/1519] eta 0:00:09 lr 0.000026 time 0.9212 (1.0060) model_time 0.9212 (1.0051) loss 0.9931 (0.8903) grad_norm 7.8562 (8.8666/2.1737) mem 68106MB [2022-12-19 18:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 32 training takes 0:25:28 [2022-12-19 18:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_32.pth saving...... [2022-12-19 18:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_32.pth saved !!! [2022-12-19 18:44:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.701 (0.701) Loss 0.5143 (0.5143) Acc@1 90.972 (90.972) Acc@5 98.264 (98.264) Mem 68106MB [2022-12-19 18:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.333) Loss 0.5218 (0.4928) Acc@1 92.708 (91.982) Acc@5 97.222 (98.327) Mem 68106MB [2022-12-19 18:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.316) Loss 0.4387 (0.4912) Acc@1 93.056 (91.832) Acc@5 98.611 (98.264) Mem 68106MB [2022-12-19 18:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.307 (0.311) Loss 0.6003 (0.4962) Acc@1 89.236 (91.543) Acc@5 97.222 (98.253) Mem 68106MB [2022-12-19 18:45:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.308) Loss 0.4686 (0.4878) Acc@1 92.014 (91.633) Acc@5 98.611 (98.315) Mem 68106MB [2022-12-19 18:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.306) Loss 0.5047 (0.4859) Acc@1 88.542 (91.537) Acc@5 99.653 (98.386) Mem 68106MB [2022-12-19 18:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.295 (0.305) Loss 0.4827 (0.4843) Acc@1 90.625 (91.593) Acc@5 98.611 (98.418) Mem 68106MB [2022-12-19 18:45:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.300 (0.305) Loss 0.5356 (0.4856) Acc@1 92.014 (91.584) Acc@5 98.611 (98.396) Mem 68106MB [2022-12-19 18:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.304) Loss 0.4131 (0.4841) Acc@1 93.056 (91.641) Acc@5 98.611 (98.414) Mem 68106MB [2022-12-19 18:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:32] * Acc@1 91.634 Acc@5 98.432 [2022-12-19 18:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.6% [2022-12-19 18:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 18:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 18:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.63% [2022-12-19 18:45:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][0/1519] eta 0:35:41 lr 0.000026 time 1.4100 (1.4100) model_time 0.9767 (0.9767) loss 1.2235 (1.2235) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 18:45:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][10/1519] eta 0:26:04 lr 0.000026 time 0.9265 (1.0368) model_time 0.9264 (0.9971) loss 0.9210 (0.9231) grad_norm 7.9265 (8.9498/1.0236) mem 68106MB [2022-12-19 18:46:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][20/1519] eta 0:25:28 lr 0.000026 time 0.9447 (1.0196) model_time 0.9446 (0.9987) loss 1.2580 (0.9621) grad_norm 6.9499 (8.5082/1.6710) mem 68106MB [2022-12-19 18:46:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][30/1519] eta 0:25:10 lr 0.000026 time 0.9242 (1.0142) model_time 0.9241 (0.9999) loss 0.7385 (0.9395) grad_norm 7.2079 (9.1147/2.5749) mem 68106MB [2022-12-19 18:46:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][40/1519] eta 0:25:03 lr 0.000026 time 0.9168 (1.0163) model_time 0.9167 (1.0054) loss 0.7374 (0.9308) grad_norm 8.7571 (8.9919/2.3085) mem 68106MB [2022-12-19 18:46:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][50/1519] eta 0:24:50 lr 0.000026 time 1.0006 (1.0146) model_time 1.0005 (1.0058) loss 0.8425 (0.9296) grad_norm 11.1691 (9.1513/2.2461) mem 68106MB [2022-12-19 18:46:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][60/1519] eta 0:24:37 lr 0.000026 time 0.9265 (1.0126) model_time 0.9264 (1.0051) loss 0.7952 (0.9114) grad_norm 9.0847 (9.2285/2.0943) mem 68106MB [2022-12-19 18:46:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][70/1519] eta 0:24:27 lr 0.000026 time 0.9199 (1.0126) model_time 0.9198 (1.0061) loss 0.9094 (0.9100) grad_norm 8.8618 (9.1649/2.0301) mem 68106MB [2022-12-19 18:47:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][80/1519] eta 0:24:14 lr 0.000026 time 0.9245 (1.0107) model_time 0.9244 (1.0050) loss 1.1610 (0.9163) grad_norm 8.8721 (9.3240/2.2075) mem 68106MB [2022-12-19 18:47:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][90/1519] eta 0:24:04 lr 0.000026 time 0.9269 (1.0111) model_time 0.9268 (1.0060) loss 1.0263 (0.9138) grad_norm 7.7513 (9.2282/2.1382) mem 68106MB [2022-12-19 18:47:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][100/1519] eta 0:23:53 lr 0.000026 time 0.9305 (1.0106) model_time 0.9303 (1.0059) loss 0.9955 (0.9222) grad_norm 6.9498 (9.1126/2.0691) mem 68106MB [2022-12-19 18:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][110/1519] eta 0:23:42 lr 0.000026 time 0.9322 (1.0098) model_time 0.9321 (1.0056) loss 1.1105 (0.9251) grad_norm 7.6782 (9.0598/2.0433) mem 68106MB [2022-12-19 18:47:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][120/1519] eta 0:23:32 lr 0.000026 time 0.9253 (1.0097) model_time 0.9251 (1.0058) loss 0.9325 (0.9187) grad_norm 7.7829 (8.9713/2.0039) mem 68106MB [2022-12-19 18:47:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][130/1519] eta 0:23:22 lr 0.000026 time 0.9078 (1.0095) model_time 0.9077 (1.0058) loss 0.6707 (0.9086) grad_norm 6.1631 (8.7988/2.0591) mem 68106MB [2022-12-19 18:48:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][140/1519] eta 0:23:11 lr 0.000026 time 0.9384 (1.0088) model_time 0.9382 (1.0054) loss 0.7852 (0.9132) grad_norm 8.1564 (8.7332/2.0608) mem 68106MB [2022-12-19 18:48:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][150/1519] eta 0:23:01 lr 0.000026 time 0.9365 (1.0091) model_time 0.9363 (1.0059) loss 0.8346 (0.9097) grad_norm 11.4552 (8.7047/2.0518) mem 68106MB [2022-12-19 18:48:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][160/1519] eta 0:22:50 lr 0.000026 time 0.9367 (1.0085) model_time 0.9365 (1.0055) loss 0.9347 (0.9063) grad_norm 6.7977 (8.7283/2.0564) mem 68106MB [2022-12-19 18:48:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][170/1519] eta 0:22:40 lr 0.000025 time 0.9285 (1.0088) model_time 0.9284 (1.0059) loss 1.0025 (0.9050) grad_norm 10.2993 (8.7108/2.0118) mem 68106MB [2022-12-19 18:48:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][180/1519] eta 0:22:29 lr 0.000025 time 0.9304 (1.0082) model_time 0.9303 (1.0054) loss 0.7648 (0.9024) grad_norm 7.5199 (8.6780/1.9994) mem 68106MB [2022-12-19 18:48:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][190/1519] eta 0:22:19 lr 0.000025 time 0.9223 (1.0077) model_time 0.9221 (1.0051) loss 0.7196 (0.8977) grad_norm 10.1031 (8.6828/1.9765) mem 68106MB [2022-12-19 18:49:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][200/1519] eta 0:22:08 lr 0.000025 time 0.9306 (1.0076) model_time 0.9305 (1.0051) loss 0.8144 (0.8980) grad_norm 7.8439 (8.6337/1.9619) mem 68106MB [2022-12-19 18:49:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][210/1519] eta 0:21:59 lr 0.000025 time 0.9219 (1.0077) model_time 0.9218 (1.0053) loss 0.8765 (0.9016) grad_norm 9.2081 (8.6378/1.9559) mem 68106MB [2022-12-19 18:49:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][220/1519] eta 0:21:50 lr 0.000025 time 0.9243 (1.0087) model_time 0.9241 (1.0064) loss 0.7796 (0.9003) grad_norm 9.0995 (8.6553/1.9713) mem 68106MB [2022-12-19 18:49:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][230/1519] eta 0:21:40 lr 0.000025 time 1.0278 (1.0087) model_time 1.0276 (1.0065) loss 0.8095 (0.9036) grad_norm 10.3640 (8.6766/1.9562) mem 68106MB [2022-12-19 18:49:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][240/1519] eta 0:21:29 lr 0.000025 time 0.9200 (1.0085) model_time 0.9199 (1.0064) loss 0.9619 (0.9049) grad_norm 11.0504 (8.6491/1.9546) mem 68106MB [2022-12-19 18:49:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][250/1519] eta 0:21:19 lr 0.000025 time 0.9262 (1.0080) model_time 0.9261 (1.0060) loss 0.8980 (0.9001) grad_norm 11.7951 (8.6454/1.9625) mem 68106MB [2022-12-19 18:50:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][260/1519] eta 0:21:08 lr 0.000025 time 0.9218 (1.0077) model_time 0.9217 (1.0057) loss 0.8097 (0.8970) grad_norm 10.2359 (8.6805/1.9555) mem 68106MB [2022-12-19 18:50:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][270/1519] eta 0:20:58 lr 0.000025 time 0.9309 (1.0079) model_time 0.9307 (1.0060) loss 0.8947 (0.8956) grad_norm 6.8750 (8.6239/1.9424) mem 68106MB [2022-12-19 18:50:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][280/1519] eta 0:20:48 lr 0.000025 time 0.9277 (1.0077) model_time 0.9276 (1.0059) loss 1.2805 (0.8968) grad_norm 8.3345 (8.6386/1.9316) mem 68106MB [2022-12-19 18:50:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][290/1519] eta 0:20:38 lr 0.000025 time 0.9294 (1.0074) model_time 0.9293 (1.0056) loss 0.7414 (0.8980) grad_norm 9.5653 (8.6448/1.9115) mem 68106MB [2022-12-19 18:50:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][300/1519] eta 0:20:27 lr 0.000025 time 0.9229 (1.0074) model_time 0.9228 (1.0056) loss 0.9482 (0.8979) grad_norm 7.0393 (8.6110/1.8965) mem 68106MB [2022-12-19 18:50:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][310/1519] eta 0:20:17 lr 0.000025 time 0.9262 (1.0071) model_time 0.9261 (1.0054) loss 0.7177 (0.8942) grad_norm 8.6832 (8.6234/1.8826) mem 68106MB [2022-12-19 18:51:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][320/1519] eta 0:20:07 lr 0.000025 time 0.9272 (1.0068) model_time 0.9271 (1.0051) loss 0.9444 (0.8941) grad_norm 7.5759 (8.6297/1.8595) mem 68106MB [2022-12-19 18:51:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][330/1519] eta 0:19:56 lr 0.000025 time 0.9160 (1.0065) model_time 0.9158 (1.0048) loss 0.8726 (0.8939) grad_norm 14.6977 (8.6547/1.8933) mem 68106MB [2022-12-19 18:51:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][340/1519] eta 0:19:46 lr 0.000025 time 0.9222 (1.0063) model_time 0.9220 (1.0047) loss 0.9964 (0.8940) grad_norm 6.6845 (8.6236/1.8970) mem 68106MB [2022-12-19 18:51:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][350/1519] eta 0:19:36 lr 0.000025 time 0.9246 (1.0068) model_time 0.9244 (1.0053) loss 0.8607 (0.8932) grad_norm 10.2056 (8.6242/1.8812) mem 68106MB [2022-12-19 18:51:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][360/1519] eta 0:19:26 lr 0.000025 time 0.9326 (1.0067) model_time 0.9325 (1.0052) loss 0.7570 (0.8930) grad_norm 11.7676 (8.6376/1.8722) mem 68106MB [2022-12-19 18:51:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][370/1519] eta 0:19:16 lr 0.000025 time 0.9300 (1.0065) model_time 0.9281 (1.0050) loss 0.8809 (0.8942) grad_norm 5.3890 (8.6219/1.8767) mem 68106MB [2022-12-19 18:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][380/1519] eta 0:19:06 lr 0.000025 time 0.9224 (1.0066) model_time 0.9223 (1.0052) loss 0.9353 (0.8943) grad_norm 11.3115 (8.6506/1.8683) mem 68106MB [2022-12-19 18:52:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][390/1519] eta 0:18:56 lr 0.000025 time 0.9217 (1.0064) model_time 0.9216 (1.0050) loss 0.9651 (0.8928) grad_norm 8.4441 (8.6824/1.9060) mem 68106MB [2022-12-19 18:52:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][400/1519] eta 0:18:46 lr 0.000025 time 0.9279 (1.0066) model_time 0.9278 (1.0052) loss 1.1924 (0.8908) grad_norm 7.7357 (8.6625/1.8882) mem 68106MB [2022-12-19 18:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][410/1519] eta 0:18:36 lr 0.000025 time 1.0241 (1.0069) model_time 1.0239 (1.0055) loss 0.9880 (0.8901) grad_norm 11.6423 (8.6997/1.9257) mem 68106MB [2022-12-19 18:52:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][420/1519] eta 0:18:26 lr 0.000025 time 0.9201 (1.0068) model_time 0.9200 (1.0055) loss 0.7364 (0.8890) grad_norm 9.7367 (8.6854/1.9144) mem 68106MB [2022-12-19 18:52:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][430/1519] eta 0:18:16 lr 0.000025 time 0.9107 (1.0068) model_time 0.9106 (1.0055) loss 0.9871 (0.8883) grad_norm 10.3971 (8.6954/1.8967) mem 68106MB [2022-12-19 18:53:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][440/1519] eta 0:18:06 lr 0.000025 time 0.9325 (1.0066) model_time 0.9323 (1.0053) loss 0.8144 (0.8877) grad_norm 9.0404 (8.6684/1.8933) mem 68106MB [2022-12-19 18:53:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][450/1519] eta 0:17:56 lr 0.000025 time 0.9513 (1.0067) model_time 0.9511 (1.0054) loss 0.9983 (0.8888) grad_norm 9.3530 (8.6794/1.8772) mem 68106MB [2022-12-19 18:53:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][460/1519] eta 0:17:46 lr 0.000025 time 0.9248 (1.0068) model_time 0.9247 (1.0055) loss 0.8776 (0.8895) grad_norm 14.4027 (8.7264/1.9027) mem 68106MB [2022-12-19 18:53:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][470/1519] eta 0:17:36 lr 0.000025 time 0.9283 (1.0067) model_time 0.9282 (1.0055) loss 1.0344 (0.8896) grad_norm 7.5238 (8.7292/1.8857) mem 68106MB [2022-12-19 18:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][480/1519] eta 0:17:26 lr 0.000025 time 1.0092 (1.0068) model_time 1.0090 (1.0056) loss 1.0302 (0.8896) grad_norm 6.3601 (8.7089/1.8755) mem 68106MB [2022-12-19 18:53:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][490/1519] eta 0:17:15 lr 0.000025 time 0.9257 (1.0066) model_time 0.9256 (1.0054) loss 0.9439 (0.8906) grad_norm 10.0712 (8.7099/1.8700) mem 68106MB [2022-12-19 18:54:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][500/1519] eta 0:17:05 lr 0.000025 time 0.9280 (1.0065) model_time 0.9279 (1.0053) loss 0.6998 (0.8905) grad_norm 6.5151 (8.7170/1.8733) mem 68106MB [2022-12-19 18:54:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][510/1519] eta 0:16:55 lr 0.000025 time 0.9278 (1.0065) model_time 0.9277 (1.0053) loss 1.1342 (0.8910) grad_norm 6.6928 (8.7077/1.8603) mem 68106MB [2022-12-19 18:54:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][520/1519] eta 0:16:45 lr 0.000025 time 0.9307 (1.0065) model_time 0.9305 (1.0053) loss 0.8965 (0.8909) grad_norm 9.3824 (8.6979/1.8477) mem 68106MB [2022-12-19 18:54:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][530/1519] eta 0:16:35 lr 0.000025 time 0.9237 (1.0065) model_time 0.9236 (1.0054) loss 0.8558 (0.8902) grad_norm 8.2473 (8.7193/1.9141) mem 68106MB [2022-12-19 18:54:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][540/1519] eta 0:16:25 lr 0.000025 time 0.9318 (1.0065) model_time 0.9317 (1.0054) loss 0.8198 (0.8896) grad_norm 6.0437 (8.7026/1.9068) mem 68106MB [2022-12-19 18:54:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][550/1519] eta 0:16:15 lr 0.000025 time 0.9268 (1.0065) model_time 0.9265 (1.0054) loss 0.8379 (0.8881) grad_norm 6.9030 (8.7197/1.9267) mem 68106MB [2022-12-19 18:55:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][560/1519] eta 0:16:05 lr 0.000025 time 0.9252 (1.0064) model_time 0.9251 (1.0053) loss 0.7969 (0.8867) grad_norm 12.3119 (8.7238/1.9415) mem 68106MB [2022-12-19 18:55:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][570/1519] eta 0:15:54 lr 0.000025 time 0.9251 (1.0063) model_time 0.9250 (1.0052) loss 0.9455 (0.8856) grad_norm 8.5149 (8.7226/1.9277) mem 68106MB [2022-12-19 18:55:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][580/1519] eta 0:15:44 lr 0.000025 time 0.9855 (1.0064) model_time 0.9852 (1.0053) loss 1.0231 (0.8862) grad_norm 8.2256 (8.7177/1.9161) mem 68106MB [2022-12-19 18:55:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][590/1519] eta 0:15:34 lr 0.000025 time 0.9223 (1.0063) model_time 0.9221 (1.0052) loss 0.9659 (0.8865) grad_norm 7.9301 (8.7173/1.9046) mem 68106MB [2022-12-19 18:55:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][600/1519] eta 0:15:24 lr 0.000025 time 0.9318 (1.0063) model_time 0.9316 (1.0052) loss 0.7575 (0.8861) grad_norm 7.2690 (8.6986/1.8979) mem 68106MB [2022-12-19 18:55:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][610/1519] eta 0:15:14 lr 0.000025 time 0.9319 (1.0061) model_time 0.9318 (1.0051) loss 0.8329 (0.8854) grad_norm 9.1050 (8.6861/1.8954) mem 68106MB [2022-12-19 18:56:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][620/1519] eta 0:15:04 lr 0.000025 time 0.9286 (1.0060) model_time 0.9285 (1.0050) loss 0.7206 (0.8849) grad_norm 8.5394 (8.6997/1.8771) mem 68106MB [2022-12-19 18:56:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][630/1519] eta 0:14:54 lr 0.000025 time 0.9311 (1.0059) model_time 0.9309 (1.0049) loss 1.0059 (0.8856) grad_norm 8.2249 (8.6494/1.8216) mem 68106MB [2022-12-19 18:56:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][640/1519] eta 0:14:44 lr 0.000025 time 0.9201 (1.0060) model_time 0.9200 (1.0049) loss 1.0101 (0.8847) grad_norm 7.7041 (8.6321/1.8257) mem 68106MB [2022-12-19 18:56:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][650/1519] eta 0:14:34 lr 0.000025 time 0.9261 (1.0058) model_time 0.9258 (1.0048) loss 0.8472 (0.8860) grad_norm 7.0408 (8.6084/1.8161) mem 68106MB [2022-12-19 18:56:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][660/1519] eta 0:14:24 lr 0.000025 time 0.9948 (1.0059) model_time 0.9946 (1.0049) loss 0.7948 (0.8864) grad_norm 6.8217 (8.5928/1.8309) mem 68106MB [2022-12-19 18:56:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][670/1519] eta 0:14:13 lr 0.000025 time 0.9307 (1.0058) model_time 0.9302 (1.0048) loss 0.7845 (0.8855) grad_norm 9.9424 (8.5989/1.8241) mem 68106MB [2022-12-19 18:57:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][680/1519] eta 0:14:03 lr 0.000025 time 0.9339 (1.0057) model_time 0.9336 (1.0047) loss 0.6974 (0.8858) grad_norm 7.0767 (8.5567/1.7751) mem 68106MB [2022-12-19 18:57:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][690/1519] eta 0:13:53 lr 0.000025 time 0.9250 (1.0057) model_time 0.9249 (1.0047) loss 0.6864 (0.8851) grad_norm 9.5878 (8.5487/1.7743) mem 68106MB [2022-12-19 18:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][700/1519] eta 0:13:43 lr 0.000025 time 0.9296 (1.0056) model_time 0.9294 (1.0046) loss 0.7921 (0.8858) grad_norm 5.8336 (8.5488/1.7942) mem 68106MB [2022-12-19 18:57:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][710/1519] eta 0:13:33 lr 0.000025 time 1.1932 (1.0059) model_time 1.1931 (1.0049) loss 0.9211 (0.8853) grad_norm 9.8200 (8.5672/1.8102) mem 68106MB [2022-12-19 18:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][720/1519] eta 0:13:23 lr 0.000025 time 0.9231 (1.0059) model_time 0.9230 (1.0049) loss 0.8147 (0.8860) grad_norm 6.9016 (8.5746/1.8081) mem 68106MB [2022-12-19 18:57:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][730/1519] eta 0:13:13 lr 0.000025 time 0.9277 (1.0059) model_time 0.9276 (1.0050) loss 1.0384 (0.8868) grad_norm 13.4339 (8.6077/1.8084) mem 68106MB [2022-12-19 18:58:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][740/1519] eta 0:13:03 lr 0.000025 time 0.9263 (1.0059) model_time 0.9262 (1.0050) loss 0.6990 (0.8856) grad_norm 9.5450 (8.6065/1.7984) mem 68106MB [2022-12-19 18:58:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][750/1519] eta 0:12:53 lr 0.000025 time 0.9344 (1.0059) model_time 0.9343 (1.0050) loss 1.0089 (0.8856) grad_norm 5.9958 (8.6156/1.7937) mem 68106MB [2022-12-19 18:58:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][760/1519] eta 0:12:43 lr 0.000025 time 0.9352 (1.0059) model_time 0.9350 (1.0050) loss 1.1621 (0.8849) grad_norm 10.1896 (8.6038/1.7905) mem 68106MB [2022-12-19 18:58:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][770/1519] eta 0:12:33 lr 0.000025 time 0.9218 (1.0058) model_time 0.9214 (1.0049) loss 0.7541 (0.8854) grad_norm 7.1626 (8.6136/1.8323) mem 68106MB [2022-12-19 18:58:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][780/1519] eta 0:12:23 lr 0.000025 time 0.9281 (1.0057) model_time 0.9279 (1.0049) loss 0.7516 (0.8852) grad_norm 8.8644 (8.6313/1.8233) mem 68106MB [2022-12-19 18:58:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][790/1519] eta 0:12:13 lr 0.000025 time 0.9383 (1.0057) model_time 0.9381 (1.0048) loss 0.7135 (0.8848) grad_norm 7.3135 (8.6328/1.8175) mem 68106MB [2022-12-19 18:59:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][800/1519] eta 0:12:03 lr 0.000025 time 0.9412 (1.0059) model_time 0.9411 (1.0050) loss 0.8450 (0.8845) grad_norm 6.9236 (8.6408/1.8183) mem 68106MB [2022-12-19 18:59:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][810/1519] eta 0:11:53 lr 0.000025 time 0.9307 (1.0058) model_time 0.9306 (1.0050) loss 1.1149 (0.8841) grad_norm 6.8325 (8.6453/1.8124) mem 68106MB [2022-12-19 18:59:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][820/1519] eta 0:11:43 lr 0.000025 time 0.8900 (1.0059) model_time 0.8899 (1.0050) loss 0.8068 (0.8837) grad_norm 8.4150 (8.6286/1.7909) mem 68106MB [2022-12-19 18:59:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][830/1519] eta 0:11:33 lr 0.000025 time 0.9195 (1.0059) model_time 0.9193 (1.0050) loss 0.7512 (0.8829) grad_norm 8.6956 (8.6115/1.7942) mem 68106MB [2022-12-19 18:59:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][840/1519] eta 0:11:23 lr 0.000025 time 0.9239 (1.0062) model_time 0.9237 (1.0053) loss 0.8264 (0.8837) grad_norm 6.2618 (8.6121/1.7884) mem 68106MB [2022-12-19 19:00:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][850/1519] eta 0:11:13 lr 0.000025 time 0.9314 (1.0061) model_time 0.9312 (1.0053) loss 0.7774 (0.8838) grad_norm 8.9138 (8.6227/1.7765) mem 68106MB [2022-12-19 19:00:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][860/1519] eta 0:11:03 lr 0.000025 time 0.9296 (1.0061) model_time 0.9292 (1.0053) loss 0.7637 (0.8845) grad_norm 5.5947 (8.6088/1.7739) mem 68106MB [2022-12-19 19:00:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][870/1519] eta 0:10:52 lr 0.000025 time 0.9285 (1.0061) model_time 0.9284 (1.0052) loss 0.7087 (0.8836) grad_norm 5.9902 (8.6416/1.7797) mem 68106MB [2022-12-19 19:00:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][880/1519] eta 0:10:42 lr 0.000025 time 0.9306 (1.0060) model_time 0.9304 (1.0051) loss 0.9572 (0.8843) grad_norm 6.1971 (8.6144/1.7774) mem 68106MB [2022-12-19 19:00:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][890/1519] eta 0:10:32 lr 0.000025 time 0.9284 (1.0060) model_time 0.9283 (1.0051) loss 0.7784 (0.8842) grad_norm 17.7273 (8.6362/1.8528) mem 68106MB [2022-12-19 19:00:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][900/1519] eta 0:10:22 lr 0.000025 time 0.9732 (1.0060) model_time 0.9730 (1.0052) loss 0.8268 (0.8833) grad_norm 9.0035 (8.6684/1.8718) mem 68106MB [2022-12-19 19:01:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][910/1519] eta 0:10:12 lr 0.000025 time 0.9256 (1.0060) model_time 0.9255 (1.0051) loss 0.6840 (0.8831) grad_norm 10.6890 (8.6648/1.8729) mem 68106MB [2022-12-19 19:01:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][920/1519] eta 0:10:02 lr 0.000025 time 0.9249 (1.0059) model_time 0.9248 (1.0051) loss 0.9223 (0.8828) grad_norm 7.6546 (8.6463/1.8803) mem 68106MB [2022-12-19 19:01:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][930/1519] eta 0:09:52 lr 0.000025 time 0.9249 (1.0058) model_time 0.9247 (1.0050) loss 1.0213 (0.8828) grad_norm 6.7746 (8.6297/1.8783) mem 68106MB [2022-12-19 19:01:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][940/1519] eta 0:09:42 lr 0.000025 time 0.9255 (1.0058) model_time 0.9254 (1.0050) loss 0.6936 (0.8827) grad_norm 6.4209 (8.6427/1.8828) mem 68106MB [2022-12-19 19:01:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][950/1519] eta 0:09:32 lr 0.000025 time 0.9303 (1.0057) model_time 0.9300 (1.0049) loss 0.9193 (0.8824) grad_norm 6.9583 (8.6231/1.8868) mem 68106MB [2022-12-19 19:01:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][960/1519] eta 0:09:22 lr 0.000025 time 0.9266 (1.0058) model_time 0.9264 (1.0050) loss 0.7080 (0.8822) grad_norm 8.4446 (8.6072/1.8786) mem 68106MB [2022-12-19 19:02:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][970/1519] eta 0:09:12 lr 0.000025 time 0.9245 (1.0057) model_time 0.9243 (1.0049) loss 0.8680 (0.8830) grad_norm 10.6003 (8.6330/1.8682) mem 68106MB [2022-12-19 19:02:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][980/1519] eta 0:09:02 lr 0.000025 time 0.9282 (1.0058) model_time 0.9281 (1.0050) loss 0.9657 (0.8838) grad_norm 9.4463 (8.6168/1.8659) mem 68106MB [2022-12-19 19:02:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][990/1519] eta 0:08:52 lr 0.000025 time 0.9276 (1.0058) model_time 0.9275 (1.0050) loss 1.1291 (0.8846) grad_norm 7.1946 (8.5787/1.8335) mem 68106MB [2022-12-19 19:02:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1000/1519] eta 0:08:42 lr 0.000025 time 1.0100 (1.0058) model_time 1.0099 (1.0050) loss 0.8586 (0.8847) grad_norm 9.5895 (8.5837/1.8387) mem 68106MB [2022-12-19 19:02:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1010/1519] eta 0:08:31 lr 0.000025 time 0.9281 (1.0057) model_time 0.9279 (1.0050) loss 0.7499 (0.8846) grad_norm 11.1509 (8.5791/1.8236) mem 68106MB [2022-12-19 19:02:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1020/1519] eta 0:08:21 lr 0.000025 time 0.9370 (1.0059) model_time 0.9369 (1.0051) loss 0.8291 (0.8843) grad_norm 8.7441 (8.5904/1.8361) mem 68106MB [2022-12-19 19:03:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1030/1519] eta 0:08:11 lr 0.000025 time 0.9242 (1.0060) model_time 0.9241 (1.0052) loss 0.6983 (0.8846) grad_norm 10.9951 (8.5856/1.8418) mem 68106MB [2022-12-19 19:03:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1040/1519] eta 0:08:01 lr 0.000025 time 0.9262 (1.0061) model_time 0.9261 (1.0053) loss 1.1171 (0.8846) grad_norm 5.7578 (8.6033/1.8463) mem 68106MB [2022-12-19 19:03:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1050/1519] eta 0:07:51 lr 0.000025 time 0.9309 (1.0061) model_time 0.9306 (1.0053) loss 0.7220 (0.8847) grad_norm 10.6280 (8.5962/1.8657) mem 68106MB [2022-12-19 19:03:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1060/1519] eta 0:07:41 lr 0.000025 time 0.9302 (1.0060) model_time 0.9300 (1.0053) loss 0.6756 (0.8845) grad_norm 8.4298 (8.5409/1.8341) mem 68106MB [2022-12-19 19:03:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1070/1519] eta 0:07:31 lr 0.000025 time 0.9308 (1.0061) model_time 0.9302 (1.0053) loss 0.7287 (0.8850) grad_norm 5.9278 (8.5220/1.8400) mem 68106MB [2022-12-19 19:03:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1080/1519] eta 0:07:21 lr 0.000025 time 0.9229 (1.0062) model_time 0.9227 (1.0055) loss 0.9562 (0.8846) grad_norm 11.2261 (8.5311/1.8499) mem 68106MB [2022-12-19 19:04:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1090/1519] eta 0:07:11 lr 0.000025 time 0.9284 (1.0062) model_time 0.9281 (1.0054) loss 0.7751 (0.8845) grad_norm 8.2445 (8.5174/1.8425) mem 68106MB [2022-12-19 19:04:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1100/1519] eta 0:07:01 lr 0.000025 time 0.9312 (1.0061) model_time 0.9310 (1.0054) loss 1.0692 (0.8853) grad_norm 9.9925 (8.5011/1.8306) mem 68106MB [2022-12-19 19:04:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1110/1519] eta 0:06:51 lr 0.000025 time 0.9257 (1.0061) model_time 0.9255 (1.0054) loss 0.7029 (0.8850) grad_norm 7.8939 (8.5193/1.8388) mem 68106MB [2022-12-19 19:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1120/1519] eta 0:06:41 lr 0.000025 time 0.9361 (1.0061) model_time 0.9358 (1.0053) loss 0.9146 (0.8854) grad_norm 10.4202 (8.5143/1.8442) mem 68106MB [2022-12-19 19:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1130/1519] eta 0:06:31 lr 0.000025 time 0.9207 (1.0060) model_time 0.9206 (1.0053) loss 0.7629 (0.8844) grad_norm 13.8100 (8.4888/1.8032) mem 68106MB [2022-12-19 19:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1140/1519] eta 0:06:21 lr 0.000025 time 1.0061 (1.0061) model_time 1.0060 (1.0053) loss 1.1012 (0.8848) grad_norm 9.9729 (8.5019/1.8087) mem 68106MB [2022-12-19 19:05:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1150/1519] eta 0:06:11 lr 0.000025 time 0.9626 (1.0061) model_time 0.9625 (1.0053) loss 0.6970 (0.8850) grad_norm 11.1058 (8.5133/1.8223) mem 68106MB [2022-12-19 19:05:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1160/1519] eta 0:06:01 lr 0.000025 time 0.9224 (1.0060) model_time 0.9223 (1.0053) loss 0.9152 (0.8855) grad_norm 7.0295 (8.4930/1.8043) mem 68106MB [2022-12-19 19:05:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1170/1519] eta 0:05:51 lr 0.000025 time 0.9208 (1.0060) model_time 0.9206 (1.0053) loss 1.2023 (0.8854) grad_norm 8.9938 (8.4906/1.8122) mem 68106MB [2022-12-19 19:05:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1180/1519] eta 0:05:41 lr 0.000025 time 1.0168 (1.0060) model_time 1.0166 (1.0053) loss 0.8274 (0.8853) grad_norm 7.2685 (8.5009/1.8284) mem 68106MB [2022-12-19 19:05:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1190/1519] eta 0:05:30 lr 0.000025 time 0.9223 (1.0060) model_time 0.9221 (1.0053) loss 0.8113 (0.8853) grad_norm 16.5193 (8.5849/2.1007) mem 68106MB [2022-12-19 19:05:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1200/1519] eta 0:05:20 lr 0.000025 time 0.9779 (1.0060) model_time 0.9777 (1.0052) loss 1.1641 (0.8862) grad_norm 10.7182 (8.6255/2.1213) mem 68106MB [2022-12-19 19:06:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1210/1519] eta 0:05:10 lr 0.000025 time 0.9242 (1.0060) model_time 0.9241 (1.0053) loss 0.7241 (0.8855) grad_norm 6.7612 (8.6228/2.1253) mem 68106MB [2022-12-19 19:06:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1220/1519] eta 0:05:00 lr 0.000025 time 0.9251 (1.0061) model_time 0.9249 (1.0054) loss 1.0028 (0.8848) grad_norm 18.0423 (8.6484/2.1967) mem 68106MB [2022-12-19 19:06:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1230/1519] eta 0:04:50 lr 0.000025 time 0.9282 (1.0060) model_time 0.9280 (1.0053) loss 0.8370 (0.8854) grad_norm 8.8727 (8.6770/2.1964) mem 68106MB [2022-12-19 19:06:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1240/1519] eta 0:04:40 lr 0.000025 time 0.9345 (1.0060) model_time 0.9344 (1.0053) loss 1.0140 (0.8853) grad_norm 7.6003 (8.6910/2.1923) mem 68106MB [2022-12-19 19:06:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1250/1519] eta 0:04:30 lr 0.000025 time 0.9295 (1.0062) model_time 0.9294 (1.0055) loss 1.0842 (0.8857) grad_norm 6.0150 (8.6727/2.1960) mem 68106MB [2022-12-19 19:06:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1260/1519] eta 0:04:20 lr 0.000025 time 0.9268 (1.0061) model_time 0.9267 (1.0054) loss 0.7310 (0.8855) grad_norm 10.0977 (8.6765/2.1874) mem 68106MB [2022-12-19 19:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1270/1519] eta 0:04:10 lr 0.000025 time 0.9294 (1.0061) model_time 0.9293 (1.0054) loss 0.9542 (0.8860) grad_norm 6.9379 (8.6548/2.1929) mem 68106MB [2022-12-19 19:07:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1280/1519] eta 0:04:00 lr 0.000025 time 0.9263 (1.0060) model_time 0.9261 (1.0053) loss 0.6951 (0.8867) grad_norm 11.0045 (8.6775/2.1968) mem 68106MB [2022-12-19 19:07:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1290/1519] eta 0:03:50 lr 0.000025 time 0.9231 (1.0061) model_time 0.9229 (1.0054) loss 0.8018 (0.8868) grad_norm 8.3483 (8.6868/2.1951) mem 68106MB [2022-12-19 19:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1300/1519] eta 0:03:40 lr 0.000025 time 0.9351 (1.0060) model_time 0.9349 (1.0054) loss 0.8345 (0.8864) grad_norm 7.5048 (8.6755/2.1832) mem 68106MB [2022-12-19 19:07:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1310/1519] eta 0:03:30 lr 0.000025 time 0.9213 (1.0060) model_time 0.9212 (1.0053) loss 1.0921 (0.8869) grad_norm 16.9334 (8.6907/2.2213) mem 68106MB [2022-12-19 19:07:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1320/1519] eta 0:03:20 lr 0.000025 time 1.0276 (1.0060) model_time 1.0274 (1.0053) loss 0.9104 (0.8866) grad_norm 7.9618 (8.7003/2.2280) mem 68106MB [2022-12-19 19:08:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1330/1519] eta 0:03:10 lr 0.000025 time 0.9040 (1.0060) model_time 0.9039 (1.0053) loss 0.9846 (0.8869) grad_norm 9.9211 (8.6963/2.2109) mem 68106MB [2022-12-19 19:08:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1340/1519] eta 0:03:00 lr 0.000025 time 0.9246 (1.0060) model_time 0.9244 (1.0053) loss 0.6778 (0.8871) grad_norm 7.4584 (8.7249/2.2268) mem 68106MB [2022-12-19 19:08:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1350/1519] eta 0:02:50 lr 0.000025 time 0.9266 (1.0060) model_time 0.9264 (1.0054) loss 0.8296 (0.8870) grad_norm 10.3332 (8.7659/2.3362) mem 68106MB [2022-12-19 19:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1360/1519] eta 0:02:39 lr 0.000025 time 0.9181 (1.0062) model_time 0.9179 (1.0055) loss 0.8586 (0.8873) grad_norm 7.6272 (8.7651/2.3296) mem 68106MB [2022-12-19 19:08:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1370/1519] eta 0:02:29 lr 0.000025 time 0.9227 (1.0061) model_time 0.9226 (1.0054) loss 1.0163 (0.8879) grad_norm 7.0717 (8.7719/2.3293) mem 68106MB [2022-12-19 19:08:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1380/1519] eta 0:02:19 lr 0.000025 time 0.9871 (1.0061) model_time 0.9870 (1.0055) loss 0.9690 (0.8874) grad_norm 8.3700 (8.7636/2.3299) mem 68106MB [2022-12-19 19:09:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1390/1519] eta 0:02:09 lr 0.000025 time 0.9363 (1.0061) model_time 0.9361 (1.0055) loss 0.7643 (0.8874) grad_norm 8.0195 (8.7712/2.3310) mem 68106MB [2022-12-19 19:09:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1400/1519] eta 0:01:59 lr 0.000025 time 0.9336 (1.0061) model_time 0.9335 (1.0054) loss 0.7461 (0.8875) grad_norm 7.8270 (8.7598/2.3268) mem 68106MB [2022-12-19 19:09:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1410/1519] eta 0:01:49 lr 0.000025 time 0.9242 (1.0061) model_time 0.9241 (1.0054) loss 1.0419 (0.8878) grad_norm 9.4450 (8.7492/2.3234) mem 68106MB [2022-12-19 19:09:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1420/1519] eta 0:01:39 lr 0.000025 time 0.9199 (1.0060) model_time 0.9197 (1.0054) loss 0.9795 (0.8878) grad_norm 8.6988 (8.7586/2.3284) mem 68106MB [2022-12-19 19:09:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1430/1519] eta 0:01:29 lr 0.000025 time 0.9288 (1.0060) model_time 0.9287 (1.0054) loss 0.8722 (0.8875) grad_norm 7.7231 (8.7490/2.3249) mem 68106MB [2022-12-19 19:09:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1440/1519] eta 0:01:19 lr 0.000025 time 0.9239 (1.0060) model_time 0.9237 (1.0053) loss 1.3460 (0.8877) grad_norm 9.7964 (8.7496/2.3252) mem 68106MB [2022-12-19 19:10:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1450/1519] eta 0:01:09 lr 0.000025 time 0.9428 (1.0059) model_time 0.9426 (1.0053) loss 0.7908 (0.8875) grad_norm 12.0514 (8.7546/2.3354) mem 68106MB [2022-12-19 19:10:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1460/1519] eta 0:00:59 lr 0.000025 time 0.9319 (1.0059) model_time 0.9317 (1.0053) loss 0.7627 (0.8871) grad_norm 8.7553 (8.7414/2.3307) mem 68106MB [2022-12-19 19:10:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1470/1519] eta 0:00:49 lr 0.000025 time 0.9308 (1.0059) model_time 0.9306 (1.0052) loss 0.6951 (0.8877) grad_norm 8.3715 (8.7179/2.3234) mem 68106MB [2022-12-19 19:10:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1480/1519] eta 0:00:39 lr 0.000025 time 0.9309 (1.0058) model_time 0.9307 (1.0052) loss 0.7793 (0.8872) grad_norm 7.3481 (8.7186/2.3260) mem 68106MB [2022-12-19 19:10:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1490/1519] eta 0:00:29 lr 0.000025 time 0.9341 (1.0058) model_time 0.9338 (1.0052) loss 0.8323 (0.8870) grad_norm 6.5071 (8.6826/2.2663) mem 68106MB [2022-12-19 19:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1500/1519] eta 0:00:19 lr 0.000025 time 0.9375 (1.0058) model_time 0.9370 (1.0052) loss 0.9382 (0.8874) grad_norm 8.9260 (8.7072/2.3317) mem 68106MB [2022-12-19 19:11:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [33/100][1510/1519] eta 0:00:09 lr 0.000025 time 0.9167 (1.0058) model_time 0.9163 (1.0051) loss 0.7970 (0.8870) grad_norm 11.2127 (8.7272/2.3840) mem 68106MB [2022-12-19 19:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 33 training takes 0:25:27 [2022-12-19 19:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_33.pth saving...... [2022-12-19 19:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_33.pth saved !!! [2022-12-19 19:11:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.667 (0.667) Loss 0.5183 (0.5183) Acc@1 90.972 (90.972) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 19:11:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.331) Loss 0.5138 (0.4991) Acc@1 92.708 (91.667) Acc@5 98.264 (98.516) Mem 68106MB [2022-12-19 19:11:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.310 (0.315) Loss 0.4627 (0.5001) Acc@1 92.014 (91.319) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-19 19:11:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.299 (0.310) Loss 0.6357 (0.5035) Acc@1 88.889 (91.297) Acc@5 96.181 (98.208) Mem 68106MB [2022-12-19 19:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.307) Loss 0.4520 (0.4943) Acc@1 91.319 (91.345) Acc@5 98.958 (98.315) Mem 68106MB [2022-12-19 19:11:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 0.4910 (0.4905) Acc@1 89.583 (91.394) Acc@5 99.306 (98.373) Mem 68106MB [2022-12-19 19:11:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.305) Loss 0.4811 (0.4895) Acc@1 90.972 (91.388) Acc@5 97.917 (98.366) Mem 68106MB [2022-12-19 19:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5435 (0.4911) Acc@1 91.667 (91.363) Acc@5 98.611 (98.347) Mem 68106MB [2022-12-19 19:12:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4187 (0.4892) Acc@1 92.708 (91.431) Acc@5 97.917 (98.354) Mem 68106MB [2022-12-19 19:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:33] * Acc@1 91.429 Acc@5 98.359 [2022-12-19 19:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.4% [2022-12-19 19:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.63% [2022-12-19 19:12:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][0/1519] eta 0:46:51 lr 0.000025 time 1.8508 (1.8508) model_time 1.0462 (1.0462) loss 0.6844 (0.6844) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 19:12:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][10/1519] eta 0:27:11 lr 0.000025 time 0.9293 (1.0814) model_time 0.9291 (1.0078) loss 0.8777 (0.8601) grad_norm 12.5475 (9.4114/2.2889) mem 68106MB [2022-12-19 19:12:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][20/1519] eta 0:26:10 lr 0.000025 time 0.9431 (1.0478) model_time 0.9429 (1.0091) loss 0.9528 (0.8521) grad_norm 8.7720 (10.2633/3.2003) mem 68106MB [2022-12-19 19:12:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][30/1519] eta 0:25:54 lr 0.000025 time 1.0324 (1.0442) model_time 1.0323 (1.0178) loss 1.0621 (0.8645) grad_norm 8.3997 (9.5231/2.8818) mem 68106MB [2022-12-19 19:12:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][40/1519] eta 0:25:29 lr 0.000025 time 0.9861 (1.0344) model_time 0.9859 (1.0143) loss 0.7040 (0.8812) grad_norm 10.5911 (9.2338/2.6471) mem 68106MB [2022-12-19 19:12:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][50/1519] eta 0:25:09 lr 0.000025 time 0.9274 (1.0275) model_time 0.9272 (1.0113) loss 0.6983 (0.8868) grad_norm 7.9798 (9.0630/2.4256) mem 68106MB [2022-12-19 19:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][60/1519] eta 0:24:55 lr 0.000025 time 0.9343 (1.0247) model_time 0.9341 (1.0111) loss 0.7637 (0.8881) grad_norm 12.1113 (9.0939/2.3377) mem 68106MB [2022-12-19 19:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][70/1519] eta 0:24:40 lr 0.000025 time 0.9142 (1.0217) model_time 0.9140 (1.0099) loss 1.1469 (0.8913) grad_norm 6.1952 (9.0401/2.2874) mem 68106MB [2022-12-19 19:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][80/1519] eta 0:24:27 lr 0.000025 time 0.9139 (1.0202) model_time 0.9137 (1.0098) loss 0.8349 (0.8958) grad_norm 7.3672 (9.0332/2.2399) mem 68106MB [2022-12-19 19:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][90/1519] eta 0:24:14 lr 0.000025 time 0.9321 (1.0181) model_time 0.9317 (1.0088) loss 1.0610 (0.9022) grad_norm 10.8078 (8.9915/2.1633) mem 68106MB [2022-12-19 19:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][100/1519] eta 0:24:01 lr 0.000025 time 0.9279 (1.0160) model_time 0.9278 (1.0076) loss 0.9047 (0.9025) grad_norm 5.2099 (8.9916/2.1474) mem 68106MB [2022-12-19 19:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][110/1519] eta 0:23:49 lr 0.000025 time 0.9287 (1.0143) model_time 0.9285 (1.0066) loss 0.8940 (0.9062) grad_norm 12.2128 (9.0652/2.1001) mem 68106MB [2022-12-19 19:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][120/1519] eta 0:23:38 lr 0.000025 time 0.9323 (1.0137) model_time 0.9322 (1.0066) loss 0.6975 (0.8998) grad_norm 6.8026 (9.0489/2.0841) mem 68106MB [2022-12-19 19:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][130/1519] eta 0:23:28 lr 0.000025 time 1.1204 (1.0139) model_time 1.1202 (1.0074) loss 1.0594 (0.8984) grad_norm 7.5722 (8.9794/2.0406) mem 68106MB [2022-12-19 19:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][140/1519] eta 0:23:17 lr 0.000025 time 0.9407 (1.0133) model_time 0.9404 (1.0072) loss 0.7098 (0.8915) grad_norm 6.7653 (8.9290/2.0139) mem 68106MB [2022-12-19 19:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][150/1519] eta 0:23:06 lr 0.000025 time 0.9355 (1.0127) model_time 0.9353 (1.0069) loss 0.7995 (0.8937) grad_norm 7.8485 (8.8611/1.9656) mem 68106MB [2022-12-19 19:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][160/1519] eta 0:22:55 lr 0.000025 time 0.9327 (1.0120) model_time 0.9326 (1.0066) loss 0.8050 (0.8978) grad_norm 8.6957 (8.8526/1.9209) mem 68106MB [2022-12-19 19:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][170/1519] eta 0:22:44 lr 0.000025 time 0.9300 (1.0117) model_time 0.9299 (1.0066) loss 1.0356 (0.9012) grad_norm 13.6523 (9.0094/2.0379) mem 68106MB [2022-12-19 19:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][180/1519] eta 0:22:34 lr 0.000025 time 0.9420 (1.0119) model_time 0.9418 (1.0071) loss 0.9109 (0.8964) grad_norm 6.9383 (8.9376/2.0547) mem 68106MB [2022-12-19 19:15:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][190/1519] eta 0:22:24 lr 0.000025 time 0.9714 (1.0117) model_time 0.9712 (1.0071) loss 0.8227 (0.8910) grad_norm 10.5890 (8.9790/2.0417) mem 68106MB [2022-12-19 19:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][200/1519] eta 0:22:13 lr 0.000025 time 0.9292 (1.0109) model_time 0.9290 (1.0065) loss 0.7463 (0.8906) grad_norm 10.9828 (8.9903/2.0034) mem 68106MB [2022-12-19 19:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][210/1519] eta 0:22:02 lr 0.000025 time 0.9377 (1.0106) model_time 0.9375 (1.0063) loss 0.8809 (0.8917) grad_norm 6.3231 (8.9516/1.9894) mem 68106MB [2022-12-19 19:15:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][220/1519] eta 0:21:54 lr 0.000025 time 1.0741 (1.0115) model_time 1.0739 (1.0075) loss 0.7159 (0.8879) grad_norm 8.6378 (8.8982/1.9686) mem 68106MB [2022-12-19 19:15:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][230/1519] eta 0:21:43 lr 0.000025 time 0.9320 (1.0110) model_time 0.9319 (1.0071) loss 0.7392 (0.8912) grad_norm 8.3928 (8.9018/1.9610) mem 68106MB [2022-12-19 19:16:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][240/1519] eta 0:21:32 lr 0.000025 time 0.9270 (1.0106) model_time 0.9267 (1.0068) loss 1.0799 (0.8922) grad_norm 5.1570 (8.8814/2.0019) mem 68106MB [2022-12-19 19:16:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][250/1519] eta 0:21:21 lr 0.000025 time 0.9346 (1.0101) model_time 0.9344 (1.0065) loss 0.7172 (0.8892) grad_norm 6.0883 (8.8452/1.9881) mem 68106MB [2022-12-19 19:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][260/1519] eta 0:21:11 lr 0.000025 time 0.9320 (1.0103) model_time 0.9317 (1.0068) loss 0.8137 (0.8892) grad_norm 8.7291 (8.8623/1.9885) mem 68106MB [2022-12-19 19:16:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][270/1519] eta 0:21:01 lr 0.000025 time 0.9340 (1.0099) model_time 0.9337 (1.0065) loss 0.8249 (0.8881) grad_norm 7.3511 (8.8583/1.9775) mem 68106MB [2022-12-19 19:16:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][280/1519] eta 0:20:50 lr 0.000025 time 0.9277 (1.0095) model_time 0.9275 (1.0062) loss 1.0612 (0.8899) grad_norm 12.1779 (8.8551/1.9813) mem 68106MB [2022-12-19 19:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][290/1519] eta 0:20:40 lr 0.000025 time 0.9297 (1.0091) model_time 0.9295 (1.0059) loss 0.7025 (0.8879) grad_norm 15.1869 (8.9356/2.0963) mem 68106MB [2022-12-19 19:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][300/1519] eta 0:20:30 lr 0.000025 time 0.9363 (1.0091) model_time 0.9361 (1.0060) loss 0.8649 (0.8874) grad_norm 6.9827 (8.9606/2.1154) mem 68106MB [2022-12-19 19:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][310/1519] eta 0:20:21 lr 0.000025 time 1.0164 (1.0100) model_time 1.0161 (1.0071) loss 0.8430 (0.8911) grad_norm 9.0845 (8.9669/2.0882) mem 68106MB [2022-12-19 19:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][320/1519] eta 0:20:10 lr 0.000025 time 0.9310 (1.0098) model_time 0.9309 (1.0069) loss 0.9763 (0.8896) grad_norm 6.9553 (8.9223/2.0766) mem 68106MB [2022-12-19 19:17:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][330/1519] eta 0:20:00 lr 0.000025 time 0.9309 (1.0098) model_time 0.9308 (1.0069) loss 0.8595 (0.8861) grad_norm 8.2806 (8.9118/2.0507) mem 68106MB [2022-12-19 19:17:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][340/1519] eta 0:19:50 lr 0.000025 time 0.9759 (1.0097) model_time 0.9757 (1.0069) loss 1.1402 (0.8862) grad_norm 7.5013 (8.8811/2.0477) mem 68106MB [2022-12-19 19:17:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][350/1519] eta 0:19:40 lr 0.000025 time 0.9375 (1.0099) model_time 0.9373 (1.0072) loss 0.7198 (0.8870) grad_norm 11.7675 (8.8973/2.0333) mem 68106MB [2022-12-19 19:18:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][360/1519] eta 0:19:30 lr 0.000025 time 0.9287 (1.0102) model_time 0.9285 (1.0076) loss 0.9389 (0.8885) grad_norm 7.7006 (8.9006/2.0418) mem 68106MB [2022-12-19 19:18:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][370/1519] eta 0:19:20 lr 0.000025 time 1.0142 (1.0101) model_time 1.0139 (1.0075) loss 0.9776 (0.8903) grad_norm 7.4117 (8.8827/2.0257) mem 68106MB [2022-12-19 19:18:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][380/1519] eta 0:19:10 lr 0.000025 time 0.9320 (1.0098) model_time 0.9318 (1.0073) loss 1.1222 (0.8913) grad_norm 10.7056 (8.8996/2.0188) mem 68106MB [2022-12-19 19:18:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][390/1519] eta 0:19:00 lr 0.000025 time 0.9336 (1.0098) model_time 0.9333 (1.0073) loss 0.7434 (0.8892) grad_norm 6.7000 (8.8915/2.0100) mem 68106MB [2022-12-19 19:18:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][400/1519] eta 0:18:49 lr 0.000025 time 0.9358 (1.0097) model_time 0.9357 (1.0073) loss 0.7583 (0.8893) grad_norm 6.3716 (8.8640/2.0079) mem 68106MB [2022-12-19 19:18:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][410/1519] eta 0:18:39 lr 0.000025 time 0.9348 (1.0095) model_time 0.9346 (1.0072) loss 0.6993 (0.8885) grad_norm 8.4486 (8.8480/2.0003) mem 68106MB [2022-12-19 19:19:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][420/1519] eta 0:18:29 lr 0.000025 time 0.9272 (1.0093) model_time 0.9271 (1.0070) loss 0.9574 (0.8886) grad_norm 9.3611 (8.8450/1.9872) mem 68106MB [2022-12-19 19:19:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][430/1519] eta 0:18:18 lr 0.000025 time 0.9300 (1.0090) model_time 0.9299 (1.0068) loss 0.8790 (0.8887) grad_norm 15.1960 (8.9021/2.0267) mem 68106MB [2022-12-19 19:19:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][440/1519] eta 0:18:08 lr 0.000025 time 0.9300 (1.0088) model_time 0.9299 (1.0066) loss 0.7484 (0.8872) grad_norm 6.6280 (8.8843/2.0196) mem 68106MB [2022-12-19 19:19:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][450/1519] eta 0:17:58 lr 0.000025 time 0.9358 (1.0087) model_time 0.9356 (1.0065) loss 0.9508 (0.8875) grad_norm 10.6991 (8.8526/2.0260) mem 68106MB [2022-12-19 19:19:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][460/1519] eta 0:17:48 lr 0.000025 time 0.9351 (1.0090) model_time 0.9349 (1.0069) loss 0.8081 (0.8859) grad_norm 11.3867 (8.8979/2.0512) mem 68106MB [2022-12-19 19:19:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][470/1519] eta 0:17:38 lr 0.000025 time 0.9285 (1.0088) model_time 0.9283 (1.0067) loss 0.6991 (0.8830) grad_norm 6.0194 (8.8647/2.0561) mem 68106MB [2022-12-19 19:20:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][480/1519] eta 0:17:28 lr 0.000025 time 0.9252 (1.0088) model_time 0.9250 (1.0067) loss 0.7885 (0.8822) grad_norm 6.2548 (8.8755/2.0748) mem 68106MB [2022-12-19 19:20:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][490/1519] eta 0:17:18 lr 0.000025 time 0.9375 (1.0090) model_time 0.9374 (1.0069) loss 0.7614 (0.8816) grad_norm 9.3645 (8.8519/2.0643) mem 68106MB [2022-12-19 19:20:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][500/1519] eta 0:17:07 lr 0.000025 time 0.9279 (1.0087) model_time 0.9277 (1.0068) loss 0.9620 (0.8836) grad_norm 7.9941 (8.8321/2.0641) mem 68106MB [2022-12-19 19:20:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][510/1519] eta 0:16:57 lr 0.000025 time 0.9343 (1.0089) model_time 0.9342 (1.0070) loss 1.0880 (0.8820) grad_norm 8.8273 (8.8124/2.0513) mem 68106MB [2022-12-19 19:20:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][520/1519] eta 0:16:47 lr 0.000025 time 0.9943 (1.0089) model_time 0.9941 (1.0069) loss 0.8903 (0.8834) grad_norm 7.8327 (8.7993/2.0351) mem 68106MB [2022-12-19 19:20:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][530/1519] eta 0:16:38 lr 0.000025 time 0.9318 (1.0092) model_time 0.9316 (1.0073) loss 0.7935 (0.8823) grad_norm 9.1753 (8.7787/2.0337) mem 68106MB [2022-12-19 19:21:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][540/1519] eta 0:16:27 lr 0.000025 time 0.9292 (1.0091) model_time 0.9291 (1.0072) loss 0.9678 (0.8817) grad_norm 7.6078 (8.7577/2.0223) mem 68106MB [2022-12-19 19:21:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][550/1519] eta 0:16:17 lr 0.000025 time 0.9277 (1.0089) model_time 0.9275 (1.0070) loss 0.8800 (0.8814) grad_norm 11.1563 (8.7488/2.0180) mem 68106MB [2022-12-19 19:21:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][560/1519] eta 0:16:07 lr 0.000025 time 0.9373 (1.0087) model_time 0.9371 (1.0069) loss 1.1735 (0.8814) grad_norm 7.3701 (8.7282/2.0095) mem 68106MB [2022-12-19 19:21:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][570/1519] eta 0:15:57 lr 0.000025 time 0.9284 (1.0085) model_time 0.9282 (1.0067) loss 0.9825 (0.8806) grad_norm 10.4304 (8.7221/2.0032) mem 68106MB [2022-12-19 19:21:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][580/1519] eta 0:15:46 lr 0.000025 time 0.9382 (1.0085) model_time 0.9379 (1.0067) loss 0.6969 (0.8804) grad_norm 9.8761 (8.7219/1.9904) mem 68106MB [2022-12-19 19:21:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][590/1519] eta 0:15:36 lr 0.000025 time 0.9370 (1.0083) model_time 0.9369 (1.0066) loss 0.8126 (0.8798) grad_norm 7.1725 (8.7339/1.9885) mem 68106MB [2022-12-19 19:22:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][600/1519] eta 0:15:26 lr 0.000025 time 0.9296 (1.0081) model_time 0.9294 (1.0064) loss 0.9205 (0.8786) grad_norm 8.3415 (8.7540/2.0056) mem 68106MB [2022-12-19 19:22:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][610/1519] eta 0:15:16 lr 0.000025 time 0.9323 (1.0081) model_time 0.9322 (1.0065) loss 1.1201 (0.8794) grad_norm 8.8555 (8.7404/1.9865) mem 68106MB [2022-12-19 19:22:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][620/1519] eta 0:15:06 lr 0.000025 time 0.9416 (1.0082) model_time 0.9415 (1.0065) loss 0.9637 (0.8790) grad_norm 7.8233 (8.6989/1.9124) mem 68106MB [2022-12-19 19:22:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][630/1519] eta 0:14:56 lr 0.000025 time 0.9289 (1.0084) model_time 0.9287 (1.0068) loss 0.8578 (0.8794) grad_norm 10.2122 (8.7073/1.9172) mem 68106MB [2022-12-19 19:22:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][640/1519] eta 0:14:46 lr 0.000025 time 0.9347 (1.0083) model_time 0.9345 (1.0067) loss 0.8486 (0.8796) grad_norm 6.2284 (8.7060/1.9181) mem 68106MB [2022-12-19 19:22:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][650/1519] eta 0:14:36 lr 0.000025 time 0.9472 (1.0082) model_time 0.9470 (1.0066) loss 0.7264 (0.8789) grad_norm 8.1537 (8.7096/1.9231) mem 68106MB [2022-12-19 19:23:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][660/1519] eta 0:14:25 lr 0.000025 time 0.9377 (1.0081) model_time 0.9375 (1.0065) loss 0.9942 (0.8792) grad_norm 10.3143 (8.6944/1.9154) mem 68106MB [2022-12-19 19:23:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][670/1519] eta 0:14:15 lr 0.000025 time 0.9337 (1.0081) model_time 0.9335 (1.0065) loss 0.9350 (0.8785) grad_norm 7.1749 (8.7030/1.9225) mem 68106MB [2022-12-19 19:23:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][680/1519] eta 0:14:05 lr 0.000025 time 0.9343 (1.0080) model_time 0.9342 (1.0065) loss 0.9216 (0.8781) grad_norm 8.2692 (8.7054/1.9111) mem 68106MB [2022-12-19 19:23:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][690/1519] eta 0:13:55 lr 0.000025 time 0.9338 (1.0081) model_time 0.9336 (1.0065) loss 0.8603 (0.8769) grad_norm 8.9896 (8.7053/1.9128) mem 68106MB [2022-12-19 19:23:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][700/1519] eta 0:13:45 lr 0.000025 time 0.9427 (1.0080) model_time 0.9425 (1.0065) loss 0.7795 (0.8762) grad_norm 6.7656 (8.6780/1.9086) mem 68106MB [2022-12-19 19:23:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][710/1519] eta 0:13:35 lr 0.000025 time 0.9300 (1.0080) model_time 0.9299 (1.0065) loss 0.7344 (0.8765) grad_norm 10.3901 (8.6510/1.9018) mem 68106MB [2022-12-19 19:24:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][720/1519] eta 0:13:25 lr 0.000025 time 0.9321 (1.0078) model_time 0.9319 (1.0063) loss 0.8920 (0.8770) grad_norm 8.0107 (8.6721/1.9072) mem 68106MB [2022-12-19 19:24:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][730/1519] eta 0:13:15 lr 0.000025 time 0.9312 (1.0078) model_time 0.9311 (1.0063) loss 1.0167 (0.8764) grad_norm 10.1933 (8.6826/1.9119) mem 68106MB [2022-12-19 19:24:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][740/1519] eta 0:13:04 lr 0.000025 time 0.9319 (1.0076) model_time 0.9318 (1.0062) loss 0.9939 (0.8774) grad_norm 8.3959 (8.6854/1.9242) mem 68106MB [2022-12-19 19:24:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][750/1519] eta 0:12:54 lr 0.000025 time 0.9319 (1.0075) model_time 0.9318 (1.0061) loss 0.9047 (0.8778) grad_norm 6.7605 (8.6917/1.9320) mem 68106MB [2022-12-19 19:24:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][760/1519] eta 0:12:44 lr 0.000025 time 0.9275 (1.0074) model_time 0.9273 (1.0060) loss 0.8957 (0.8775) grad_norm 7.7686 (8.6730/1.9348) mem 68106MB [2022-12-19 19:24:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][770/1519] eta 0:12:34 lr 0.000025 time 0.9292 (1.0073) model_time 0.9291 (1.0058) loss 0.7927 (0.8783) grad_norm 7.4205 (8.6471/1.9168) mem 68106MB [2022-12-19 19:25:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][780/1519] eta 0:12:24 lr 0.000025 time 0.9279 (1.0071) model_time 0.9278 (1.0057) loss 0.7736 (0.8786) grad_norm 8.6109 (8.6789/1.9102) mem 68106MB [2022-12-19 19:25:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][790/1519] eta 0:12:14 lr 0.000025 time 0.9341 (1.0070) model_time 0.9339 (1.0056) loss 0.8732 (0.8788) grad_norm 8.1477 (8.6608/1.8967) mem 68106MB [2022-12-19 19:25:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][800/1519] eta 0:12:04 lr 0.000025 time 0.9256 (1.0072) model_time 0.9255 (1.0058) loss 1.0432 (0.8794) grad_norm 9.3823 (8.6447/1.9022) mem 68106MB [2022-12-19 19:25:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][810/1519] eta 0:11:54 lr 0.000025 time 0.9298 (1.0073) model_time 0.9297 (1.0059) loss 0.9047 (0.8787) grad_norm 8.1214 (8.6449/1.8998) mem 68106MB [2022-12-19 19:25:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][820/1519] eta 0:11:44 lr 0.000025 time 0.9359 (1.0072) model_time 0.9358 (1.0059) loss 0.8264 (0.8779) grad_norm 6.4341 (8.6480/1.9197) mem 68106MB [2022-12-19 19:25:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][830/1519] eta 0:11:34 lr 0.000025 time 0.9373 (1.0073) model_time 0.9371 (1.0060) loss 1.1717 (0.8793) grad_norm 8.5533 (8.6267/1.9110) mem 68106MB [2022-12-19 19:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][840/1519] eta 0:11:24 lr 0.000025 time 0.9293 (1.0077) model_time 0.9291 (1.0063) loss 0.7491 (0.8790) grad_norm 9.5819 (8.6265/1.8801) mem 68106MB [2022-12-19 19:26:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][850/1519] eta 0:11:14 lr 0.000025 time 0.9326 (1.0076) model_time 0.9324 (1.0063) loss 1.1030 (0.8794) grad_norm 6.9657 (8.6154/1.8844) mem 68106MB [2022-12-19 19:26:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][860/1519] eta 0:11:03 lr 0.000025 time 0.9282 (1.0075) model_time 0.9281 (1.0062) loss 0.9741 (0.8798) grad_norm 7.9430 (8.6106/1.8856) mem 68106MB [2022-12-19 19:26:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][870/1519] eta 0:10:53 lr 0.000025 time 0.9278 (1.0076) model_time 0.9277 (1.0063) loss 1.1377 (0.8795) grad_norm 10.2764 (8.6156/1.8877) mem 68106MB [2022-12-19 19:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][880/1519] eta 0:10:43 lr 0.000025 time 0.9740 (1.0076) model_time 0.9739 (1.0063) loss 0.9223 (0.8803) grad_norm 7.2069 (8.6062/1.8736) mem 68106MB [2022-12-19 19:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][890/1519] eta 0:10:33 lr 0.000025 time 0.9366 (1.0077) model_time 0.9364 (1.0064) loss 0.8653 (0.8809) grad_norm 9.0001 (8.5545/1.7896) mem 68106MB [2022-12-19 19:27:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][900/1519] eta 0:10:23 lr 0.000025 time 0.9246 (1.0076) model_time 0.9244 (1.0063) loss 0.7147 (0.8810) grad_norm 7.6847 (8.5172/1.7605) mem 68106MB [2022-12-19 19:27:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][910/1519] eta 0:10:13 lr 0.000025 time 0.9320 (1.0075) model_time 0.9318 (1.0062) loss 1.0101 (0.8823) grad_norm 7.3061 (8.4969/1.7628) mem 68106MB [2022-12-19 19:27:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][920/1519] eta 0:10:03 lr 0.000025 time 1.0228 (1.0076) model_time 1.0227 (1.0063) loss 0.7648 (0.8819) grad_norm 8.6653 (8.5166/1.7651) mem 68106MB [2022-12-19 19:27:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][930/1519] eta 0:09:53 lr 0.000025 time 0.9281 (1.0075) model_time 0.9280 (1.0062) loss 0.8720 (0.8814) grad_norm 6.7735 (8.5115/1.7793) mem 68106MB [2022-12-19 19:27:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][940/1519] eta 0:09:43 lr 0.000025 time 0.9239 (1.0075) model_time 0.9238 (1.0063) loss 1.1141 (0.8810) grad_norm 7.9288 (8.5099/1.7719) mem 68106MB [2022-12-19 19:28:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][950/1519] eta 0:09:33 lr 0.000025 time 0.9336 (1.0075) model_time 0.9333 (1.0063) loss 0.9440 (0.8806) grad_norm 6.7912 (8.4949/1.7830) mem 68106MB [2022-12-19 19:28:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][960/1519] eta 0:09:23 lr 0.000025 time 0.9763 (1.0074) model_time 0.9760 (1.0062) loss 0.7222 (0.8812) grad_norm 6.7048 (8.4686/1.7625) mem 68106MB [2022-12-19 19:28:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][970/1519] eta 0:09:13 lr 0.000025 time 0.9277 (1.0074) model_time 0.9276 (1.0061) loss 0.7804 (0.8803) grad_norm 4.4956 (8.4508/1.7779) mem 68106MB [2022-12-19 19:28:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][980/1519] eta 0:09:02 lr 0.000025 time 0.9628 (1.0073) model_time 0.9626 (1.0061) loss 0.7394 (0.8810) grad_norm 8.8483 (8.4466/1.7814) mem 68106MB [2022-12-19 19:28:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][990/1519] eta 0:08:52 lr 0.000025 time 0.9245 (1.0074) model_time 0.9244 (1.0062) loss 0.7629 (0.8813) grad_norm 15.0544 (8.4730/1.8139) mem 68106MB [2022-12-19 19:28:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1000/1519] eta 0:08:42 lr 0.000025 time 0.9310 (1.0073) model_time 0.9309 (1.0061) loss 0.8195 (0.8805) grad_norm 8.1077 (8.5022/1.8121) mem 68106MB [2022-12-19 19:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1010/1519] eta 0:08:32 lr 0.000025 time 0.9185 (1.0073) model_time 0.9182 (1.0061) loss 0.8091 (0.8804) grad_norm 9.5930 (8.4950/1.8105) mem 68106MB [2022-12-19 19:29:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1020/1519] eta 0:08:22 lr 0.000025 time 0.9322 (1.0073) model_time 0.9320 (1.0061) loss 0.7788 (0.8805) grad_norm 7.4622 (8.4796/1.8103) mem 68106MB [2022-12-19 19:29:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1030/1519] eta 0:08:12 lr 0.000025 time 0.9339 (1.0072) model_time 0.9338 (1.0060) loss 1.0406 (0.8809) grad_norm 10.2066 (8.4248/1.7706) mem 68106MB [2022-12-19 19:29:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1040/1519] eta 0:08:02 lr 0.000025 time 0.9204 (1.0071) model_time 0.9203 (1.0059) loss 0.9913 (0.8817) grad_norm 7.8931 (8.4167/1.7645) mem 68106MB [2022-12-19 19:29:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1050/1519] eta 0:07:52 lr 0.000025 time 0.9356 (1.0070) model_time 0.9355 (1.0059) loss 0.7103 (0.8817) grad_norm 6.9099 (8.4273/1.7473) mem 68106MB [2022-12-19 19:29:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1060/1519] eta 0:07:42 lr 0.000025 time 1.0442 (1.0070) model_time 1.0441 (1.0059) loss 0.8764 (0.8814) grad_norm 5.9248 (8.3908/1.7116) mem 68106MB [2022-12-19 19:30:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1070/1519] eta 0:07:32 lr 0.000025 time 0.9246 (1.0070) model_time 0.9244 (1.0058) loss 0.6730 (0.8807) grad_norm 7.9675 (8.4123/1.7014) mem 68106MB [2022-12-19 19:30:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1080/1519] eta 0:07:22 lr 0.000025 time 0.9305 (1.0069) model_time 0.9304 (1.0057) loss 0.7419 (0.8814) grad_norm 6.0967 (8.3771/1.6690) mem 68106MB [2022-12-19 19:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1090/1519] eta 0:07:11 lr 0.000025 time 0.9362 (1.0068) model_time 0.9361 (1.0057) loss 0.8012 (0.8817) grad_norm 9.1472 (8.3915/1.6762) mem 68106MB [2022-12-19 19:30:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1100/1519] eta 0:07:01 lr 0.000025 time 0.9300 (1.0067) model_time 0.9298 (1.0056) loss 1.0447 (0.8819) grad_norm 5.5739 (8.3860/1.6731) mem 68106MB [2022-12-19 19:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1110/1519] eta 0:06:51 lr 0.000025 time 0.9374 (1.0068) model_time 0.9373 (1.0057) loss 0.7113 (0.8821) grad_norm 6.0338 (8.3975/1.6871) mem 68106MB [2022-12-19 19:30:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1120/1519] eta 0:06:41 lr 0.000025 time 0.9300 (1.0070) model_time 0.9299 (1.0059) loss 1.2116 (0.8826) grad_norm 6.2987 (8.3987/1.6969) mem 68106MB [2022-12-19 19:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1130/1519] eta 0:06:31 lr 0.000025 time 0.9269 (1.0070) model_time 0.9268 (1.0059) loss 1.2587 (0.8829) grad_norm 11.2091 (8.4074/1.6926) mem 68106MB [2022-12-19 19:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1140/1519] eta 0:06:21 lr 0.000025 time 0.9886 (1.0070) model_time 0.9884 (1.0059) loss 1.1328 (0.8832) grad_norm 9.0579 (8.4208/1.6957) mem 68106MB [2022-12-19 19:31:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1150/1519] eta 0:06:11 lr 0.000025 time 0.9265 (1.0070) model_time 0.9264 (1.0059) loss 0.9277 (0.8828) grad_norm 8.7726 (8.4226/1.6870) mem 68106MB [2022-12-19 19:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1160/1519] eta 0:06:01 lr 0.000025 time 0.9336 (1.0070) model_time 0.9335 (1.0059) loss 0.7578 (0.8823) grad_norm 7.9695 (8.4405/1.7045) mem 68106MB [2022-12-19 19:31:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1170/1519] eta 0:05:51 lr 0.000025 time 0.9332 (1.0069) model_time 0.9331 (1.0059) loss 0.6999 (0.8821) grad_norm 11.3952 (8.4390/1.7066) mem 68106MB [2022-12-19 19:31:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1180/1519] eta 0:05:41 lr 0.000025 time 0.9369 (1.0070) model_time 0.9367 (1.0059) loss 0.6812 (0.8815) grad_norm 7.4019 (8.4391/1.7137) mem 68106MB [2022-12-19 19:32:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1190/1519] eta 0:05:31 lr 0.000025 time 0.9324 (1.0070) model_time 0.9323 (1.0059) loss 0.7812 (0.8820) grad_norm 8.3712 (8.4123/1.7009) mem 68106MB [2022-12-19 19:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1200/1519] eta 0:05:21 lr 0.000025 time 0.9359 (1.0070) model_time 0.9357 (1.0060) loss 1.2447 (0.8821) grad_norm 5.9633 (8.3815/1.6646) mem 68106MB [2022-12-19 19:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1210/1519] eta 0:05:11 lr 0.000025 time 0.9325 (1.0070) model_time 0.9323 (1.0059) loss 0.8437 (0.8820) grad_norm 7.8283 (8.3642/1.6723) mem 68106MB [2022-12-19 19:32:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1220/1519] eta 0:05:01 lr 0.000025 time 0.9429 (1.0069) model_time 0.9428 (1.0059) loss 1.0401 (0.8823) grad_norm 11.2884 (8.3664/1.6745) mem 68106MB [2022-12-19 19:32:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1230/1519] eta 0:04:50 lr 0.000025 time 0.9342 (1.0069) model_time 0.9340 (1.0059) loss 0.7759 (0.8824) grad_norm 9.0340 (8.3545/1.6658) mem 68106MB [2022-12-19 19:32:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1240/1519] eta 0:04:40 lr 0.000025 time 0.9355 (1.0069) model_time 0.9354 (1.0059) loss 0.7360 (0.8819) grad_norm 8.4396 (8.3602/1.6613) mem 68106MB [2022-12-19 19:33:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1250/1519] eta 0:04:30 lr 0.000025 time 0.9881 (1.0069) model_time 0.9880 (1.0059) loss 0.7540 (0.8812) grad_norm 5.8428 (8.3618/1.6760) mem 68106MB [2022-12-19 19:33:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1260/1519] eta 0:04:20 lr 0.000025 time 0.9835 (1.0069) model_time 0.9834 (1.0059) loss 0.8249 (0.8816) grad_norm 11.6735 (8.3704/1.6882) mem 68106MB [2022-12-19 19:33:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1270/1519] eta 0:04:10 lr 0.000025 time 0.9287 (1.0069) model_time 0.9286 (1.0059) loss 0.7786 (0.8817) grad_norm 8.0556 (8.3610/1.6663) mem 68106MB [2022-12-19 19:33:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1280/1519] eta 0:04:00 lr 0.000025 time 0.9279 (1.0068) model_time 0.9277 (1.0058) loss 0.6995 (0.8820) grad_norm 9.2823 (8.3496/1.6612) mem 68106MB [2022-12-19 19:33:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1290/1519] eta 0:03:50 lr 0.000025 time 0.9287 (1.0068) model_time 0.9285 (1.0058) loss 0.8041 (0.8815) grad_norm 6.7782 (8.3277/1.6607) mem 68106MB [2022-12-19 19:33:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1300/1519] eta 0:03:40 lr 0.000025 time 0.9321 (1.0069) model_time 0.9320 (1.0059) loss 0.9659 (0.8816) grad_norm 8.2116 (8.3469/1.6653) mem 68106MB [2022-12-19 19:34:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1310/1519] eta 0:03:30 lr 0.000025 time 0.9330 (1.0068) model_time 0.9329 (1.0059) loss 0.7078 (0.8819) grad_norm 9.2414 (8.3690/1.6777) mem 68106MB [2022-12-19 19:34:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1320/1519] eta 0:03:20 lr 0.000025 time 0.9355 (1.0068) model_time 0.9352 (1.0058) loss 0.8826 (0.8823) grad_norm 6.4294 (8.3444/1.6897) mem 68106MB [2022-12-19 19:34:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1330/1519] eta 0:03:10 lr 0.000025 time 0.9294 (1.0068) model_time 0.9292 (1.0058) loss 0.9143 (0.8827) grad_norm 8.0740 (8.3349/1.6828) mem 68106MB [2022-12-19 19:34:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1340/1519] eta 0:03:00 lr 0.000025 time 0.9315 (1.0068) model_time 0.9314 (1.0059) loss 0.7745 (0.8827) grad_norm 10.5088 (8.3513/1.6732) mem 68106MB [2022-12-19 19:34:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1350/1519] eta 0:02:50 lr 0.000025 time 0.9324 (1.0068) model_time 0.9323 (1.0058) loss 0.8413 (0.8826) grad_norm 10.0847 (8.3561/1.6707) mem 68106MB [2022-12-19 19:34:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1360/1519] eta 0:02:40 lr 0.000025 time 0.9380 (1.0067) model_time 0.9378 (1.0058) loss 1.0056 (0.8827) grad_norm 9.9679 (8.3938/1.7013) mem 68106MB [2022-12-19 19:35:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1370/1519] eta 0:02:29 lr 0.000025 time 0.9368 (1.0067) model_time 0.9366 (1.0057) loss 0.7599 (0.8829) grad_norm 8.4383 (8.3780/1.6820) mem 68106MB [2022-12-19 19:35:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1380/1519] eta 0:02:19 lr 0.000025 time 0.9324 (1.0068) model_time 0.9323 (1.0059) loss 0.9342 (0.8829) grad_norm 8.9438 (8.3635/1.6693) mem 68106MB [2022-12-19 19:35:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1390/1519] eta 0:02:09 lr 0.000025 time 0.9293 (1.0068) model_time 0.9292 (1.0058) loss 1.0445 (0.8827) grad_norm 7.1885 (8.3457/1.6739) mem 68106MB [2022-12-19 19:35:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1400/1519] eta 0:01:59 lr 0.000025 time 0.9312 (1.0067) model_time 0.9311 (1.0058) loss 0.8263 (0.8825) grad_norm 8.4047 (8.3599/1.7278) mem 68106MB [2022-12-19 19:35:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1410/1519] eta 0:01:49 lr 0.000025 time 0.9307 (1.0067) model_time 0.9305 (1.0057) loss 1.0216 (0.8829) grad_norm 13.4225 (8.3862/1.7543) mem 68106MB [2022-12-19 19:35:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1420/1519] eta 0:01:39 lr 0.000025 time 0.9293 (1.0066) model_time 0.9292 (1.0056) loss 0.8188 (0.8832) grad_norm 10.2821 (8.3901/1.7392) mem 68106MB [2022-12-19 19:36:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1430/1519] eta 0:01:29 lr 0.000025 time 0.9411 (1.0067) model_time 0.9410 (1.0058) loss 0.8725 (0.8831) grad_norm 6.1501 (8.4151/1.7495) mem 68106MB [2022-12-19 19:36:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1440/1519] eta 0:01:19 lr 0.000025 time 0.9343 (1.0067) model_time 0.9341 (1.0058) loss 0.9567 (0.8830) grad_norm 7.8947 (8.4282/1.7556) mem 68106MB [2022-12-19 19:36:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1450/1519] eta 0:01:09 lr 0.000025 time 0.9342 (1.0068) model_time 0.9341 (1.0058) loss 0.8400 (0.8834) grad_norm 7.7650 (8.4393/1.7479) mem 68106MB [2022-12-19 19:36:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1460/1519] eta 0:00:59 lr 0.000025 time 0.9840 (1.0068) model_time 0.9838 (1.0059) loss 0.8629 (0.8834) grad_norm 7.4719 (8.4280/1.7341) mem 68106MB [2022-12-19 19:36:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1470/1519] eta 0:00:49 lr 0.000025 time 0.9312 (1.0068) model_time 0.9310 (1.0059) loss 0.9293 (0.8829) grad_norm 8.8344 (8.4635/1.7675) mem 68106MB [2022-12-19 19:36:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1480/1519] eta 0:00:39 lr 0.000025 time 0.9343 (1.0070) model_time 0.9342 (1.0060) loss 0.9312 (0.8831) grad_norm 6.9219 (8.4698/1.7828) mem 68106MB [2022-12-19 19:37:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1490/1519] eta 0:00:29 lr 0.000025 time 0.9373 (1.0070) model_time 0.9372 (1.0061) loss 0.8711 (0.8835) grad_norm 7.2368 (8.4725/1.7932) mem 68106MB [2022-12-19 19:37:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1500/1519] eta 0:00:19 lr 0.000025 time 0.9105 (1.0070) model_time 0.9103 (1.0061) loss 1.0832 (0.8836) grad_norm 7.3131 (8.4736/1.7916) mem 68106MB [2022-12-19 19:37:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [34/100][1510/1519] eta 0:00:09 lr 0.000025 time 0.9657 (1.0070) model_time 0.9656 (1.0060) loss 1.1435 (0.8843) grad_norm 9.9063 (8.4831/1.7956) mem 68106MB [2022-12-19 19:37:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 34 training takes 0:25:29 [2022-12-19 19:37:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_34.pth saving...... [2022-12-19 19:37:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_34.pth saved !!! [2022-12-19 19:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.653 (0.653) Loss 0.5203 (0.5203) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 19:38:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.329) Loss 0.5141 (0.4887) Acc@1 92.014 (91.825) Acc@5 97.917 (98.453) Mem 68106MB [2022-12-19 19:38:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.313) Loss 0.4449 (0.4884) Acc@1 92.014 (91.634) Acc@5 98.958 (98.330) Mem 68106MB [2022-12-19 19:38:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.308) Loss 0.5988 (0.4926) Acc@1 88.194 (91.409) Acc@5 97.569 (98.297) Mem 68106MB [2022-12-19 19:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.306) Loss 0.4417 (0.4849) Acc@1 92.708 (91.565) Acc@5 99.306 (98.382) Mem 68106MB [2022-12-19 19:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.4948 (0.4829) Acc@1 89.931 (91.605) Acc@5 99.306 (98.427) Mem 68106MB [2022-12-19 19:38:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.4930 (0.4820) Acc@1 90.625 (91.667) Acc@5 98.264 (98.423) Mem 68106MB [2022-12-19 19:38:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.296 (0.303) Loss 0.5535 (0.4838) Acc@1 91.319 (91.637) Acc@5 98.264 (98.411) Mem 68106MB [2022-12-19 19:38:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.303) Loss 0.3879 (0.4816) Acc@1 92.708 (91.650) Acc@5 98.611 (98.440) Mem 68106MB [2022-12-19 19:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:34] * Acc@1 91.613 Acc@5 98.449 [2022-12-19 19:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.6% [2022-12-19 19:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.63% [2022-12-19 19:38:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][0/1519] eta 0:47:11 lr 0.000025 time 1.8643 (1.8643) model_time 1.0294 (1.0294) loss 0.9109 (0.9109) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 19:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][10/1519] eta 0:27:05 lr 0.000025 time 0.9227 (1.0770) model_time 0.9226 (1.0007) loss 0.7761 (0.8515) grad_norm 8.5963 (8.7046/1.5271) mem 68106MB [2022-12-19 19:38:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][20/1519] eta 0:25:59 lr 0.000025 time 0.9393 (1.0405) model_time 0.9392 (1.0004) loss 1.0341 (0.8689) grad_norm 12.9781 (8.2726/2.1908) mem 68106MB [2022-12-19 19:38:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][30/1519] eta 0:25:32 lr 0.000025 time 0.9420 (1.0289) model_time 0.9418 (1.0016) loss 1.2189 (0.8621) grad_norm 7.4595 (8.8171/2.1489) mem 68106MB [2022-12-19 19:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][40/1519] eta 0:25:10 lr 0.000025 time 0.9302 (1.0211) model_time 0.9301 (1.0004) loss 0.7668 (0.8591) grad_norm 6.8042 (8.6615/2.0560) mem 68106MB [2022-12-19 19:39:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][50/1519] eta 0:24:55 lr 0.000025 time 0.9320 (1.0182) model_time 0.9319 (1.0014) loss 1.2557 (0.8585) grad_norm 10.5849 (8.6273/1.9422) mem 68106MB [2022-12-19 19:39:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][60/1519] eta 0:24:43 lr 0.000025 time 0.9164 (1.0166) model_time 0.9158 (1.0025) loss 0.8832 (0.8477) grad_norm 11.2102 (8.6584/1.9186) mem 68106MB [2022-12-19 19:39:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][70/1519] eta 0:24:31 lr 0.000025 time 0.9751 (1.0158) model_time 0.9750 (1.0036) loss 1.0439 (0.8469) grad_norm 6.2442 (8.4941/1.9731) mem 68106MB [2022-12-19 19:39:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][80/1519] eta 0:24:20 lr 0.000025 time 0.9283 (1.0147) model_time 0.9281 (1.0040) loss 0.8988 (0.8497) grad_norm 8.4237 (8.4825/1.8763) mem 68106MB [2022-12-19 19:39:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][90/1519] eta 0:24:07 lr 0.000025 time 0.9284 (1.0129) model_time 0.9282 (1.0034) loss 0.8562 (0.8518) grad_norm 5.5346 (8.4952/1.9995) mem 68106MB [2022-12-19 19:40:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][100/1519] eta 0:23:56 lr 0.000025 time 0.9753 (1.0122) model_time 0.9751 (1.0035) loss 0.9202 (0.8529) grad_norm 10.6440 (8.6431/2.0189) mem 68106MB [2022-12-19 19:40:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][110/1519] eta 0:23:44 lr 0.000025 time 0.9219 (1.0109) model_time 0.9215 (1.0030) loss 1.2222 (0.8507) grad_norm 11.1490 (8.6179/1.9807) mem 68106MB [2022-12-19 19:40:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][120/1519] eta 0:23:36 lr 0.000025 time 1.1885 (1.0124) model_time 1.1884 (1.0051) loss 0.7129 (0.8548) grad_norm 8.4038 (8.5812/1.9202) mem 68106MB [2022-12-19 19:40:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][130/1519] eta 0:23:24 lr 0.000025 time 0.9241 (1.0113) model_time 0.9239 (1.0045) loss 0.7184 (0.8570) grad_norm 9.5183 (8.6223/1.9117) mem 68106MB [2022-12-19 19:40:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][140/1519] eta 0:23:14 lr 0.000025 time 0.9232 (1.0111) model_time 0.9230 (1.0048) loss 0.7789 (0.8552) grad_norm 8.6967 (8.5946/1.8862) mem 68106MB [2022-12-19 19:40:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][150/1519] eta 0:23:03 lr 0.000025 time 0.9218 (1.0104) model_time 0.9217 (1.0044) loss 0.6999 (0.8543) grad_norm 9.5195 (8.6046/1.8367) mem 68106MB [2022-12-19 19:41:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][160/1519] eta 0:22:52 lr 0.000025 time 0.9281 (1.0100) model_time 0.9279 (1.0044) loss 0.9748 (0.8576) grad_norm 5.9652 (8.5277/1.8325) mem 68106MB [2022-12-19 19:41:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][170/1519] eta 0:22:42 lr 0.000025 time 1.0175 (1.0102) model_time 1.0173 (1.0049) loss 0.7986 (0.8601) grad_norm 7.9110 (8.4803/1.8149) mem 68106MB [2022-12-19 19:41:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][180/1519] eta 0:22:31 lr 0.000025 time 0.9302 (1.0097) model_time 0.9301 (1.0047) loss 0.8221 (0.8621) grad_norm 6.4113 (8.4745/1.8144) mem 68106MB [2022-12-19 19:41:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][190/1519] eta 0:22:21 lr 0.000025 time 0.9281 (1.0097) model_time 0.9279 (1.0050) loss 0.8907 (0.8630) grad_norm 7.0827 (8.4847/1.7819) mem 68106MB [2022-12-19 19:41:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][200/1519] eta 0:22:11 lr 0.000025 time 0.9273 (1.0092) model_time 0.9271 (1.0047) loss 1.0510 (0.8631) grad_norm 7.7493 (8.4939/1.7701) mem 68106MB [2022-12-19 19:41:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][210/1519] eta 0:22:01 lr 0.000025 time 0.9070 (1.0094) model_time 0.9069 (1.0050) loss 0.7814 (0.8616) grad_norm 10.7210 (8.5487/1.7825) mem 68106MB [2022-12-19 19:42:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][220/1519] eta 0:21:50 lr 0.000025 time 0.9225 (1.0090) model_time 0.9222 (1.0048) loss 1.1169 (0.8621) grad_norm 8.0705 (8.4901/1.7728) mem 68106MB [2022-12-19 19:42:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][230/1519] eta 0:21:40 lr 0.000025 time 0.9330 (1.0089) model_time 0.9328 (1.0049) loss 0.6890 (0.8612) grad_norm 6.7662 (8.4633/1.7785) mem 68106MB [2022-12-19 19:42:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][240/1519] eta 0:21:29 lr 0.000025 time 0.9319 (1.0085) model_time 0.9318 (1.0046) loss 0.7234 (0.8606) grad_norm 6.4309 (8.4364/1.7793) mem 68106MB [2022-12-19 19:42:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][250/1519] eta 0:21:20 lr 0.000025 time 0.9354 (1.0088) model_time 0.9353 (1.0051) loss 0.9944 (0.8637) grad_norm 18.3006 (8.5004/1.9748) mem 68106MB [2022-12-19 19:42:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][260/1519] eta 0:21:10 lr 0.000025 time 0.9314 (1.0089) model_time 0.9312 (1.0053) loss 1.0560 (0.8630) grad_norm 13.1558 (8.5088/1.9931) mem 68106MB [2022-12-19 19:42:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][270/1519] eta 0:21:00 lr 0.000025 time 0.9856 (1.0090) model_time 0.9855 (1.0055) loss 0.9272 (0.8618) grad_norm 5.8470 (8.4678/1.9886) mem 68106MB [2022-12-19 19:43:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][280/1519] eta 0:20:49 lr 0.000025 time 0.9768 (1.0089) model_time 0.9766 (1.0055) loss 0.7338 (0.8642) grad_norm 7.5595 (8.5172/2.0111) mem 68106MB [2022-12-19 19:43:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][290/1519] eta 0:20:40 lr 0.000025 time 0.9197 (1.0091) model_time 0.9194 (1.0058) loss 0.9437 (0.8652) grad_norm 6.1731 (8.5056/1.9912) mem 68106MB [2022-12-19 19:43:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][300/1519] eta 0:20:30 lr 0.000025 time 1.0046 (1.0090) model_time 1.0045 (1.0059) loss 0.7239 (0.8650) grad_norm 6.9416 (8.5098/2.0071) mem 68106MB [2022-12-19 19:43:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][310/1519] eta 0:20:19 lr 0.000025 time 0.9297 (1.0089) model_time 0.9295 (1.0058) loss 1.0304 (0.8668) grad_norm 8.7548 (8.4914/1.9915) mem 68106MB [2022-12-19 19:43:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][320/1519] eta 0:20:10 lr 0.000025 time 0.9176 (1.0093) model_time 0.9174 (1.0063) loss 0.8386 (0.8684) grad_norm 8.4806 (8.4883/1.9699) mem 68106MB [2022-12-19 19:43:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][330/1519] eta 0:19:59 lr 0.000025 time 0.9357 (1.0090) model_time 0.9356 (1.0060) loss 0.8405 (0.8689) grad_norm 7.2115 (8.4748/1.9538) mem 68106MB [2022-12-19 19:44:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][340/1519] eta 0:19:49 lr 0.000025 time 0.9320 (1.0087) model_time 0.9318 (1.0058) loss 0.9263 (0.8700) grad_norm 7.0171 (8.4563/1.9423) mem 68106MB [2022-12-19 19:44:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][350/1519] eta 0:19:38 lr 0.000025 time 0.9801 (1.0085) model_time 0.9800 (1.0057) loss 0.8312 (0.8703) grad_norm 6.1594 (8.4521/1.9368) mem 68106MB [2022-12-19 19:44:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][360/1519] eta 0:19:28 lr 0.000025 time 0.9272 (1.0084) model_time 0.9270 (1.0057) loss 0.7626 (0.8734) grad_norm 8.0994 (8.4099/1.9320) mem 68106MB [2022-12-19 19:44:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][370/1519] eta 0:19:18 lr 0.000025 time 0.9355 (1.0083) model_time 0.9353 (1.0056) loss 1.3050 (0.8787) grad_norm 6.9939 (8.3978/1.9227) mem 68106MB [2022-12-19 19:44:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][380/1519] eta 0:19:08 lr 0.000025 time 0.9372 (1.0081) model_time 0.9370 (1.0055) loss 1.1176 (0.8804) grad_norm 8.3570 (8.3970/1.9194) mem 68106MB [2022-12-19 19:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][390/1519] eta 0:18:58 lr 0.000025 time 0.9282 (1.0082) model_time 0.9281 (1.0056) loss 0.7525 (0.8797) grad_norm 13.5020 (8.5263/2.0624) mem 68106MB [2022-12-19 19:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][400/1519] eta 0:18:48 lr 0.000025 time 0.9247 (1.0081) model_time 0.9246 (1.0056) loss 0.9473 (0.8791) grad_norm 7.4871 (8.5074/2.0441) mem 68106MB [2022-12-19 19:45:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][410/1519] eta 0:18:37 lr 0.000025 time 0.9268 (1.0081) model_time 0.9266 (1.0056) loss 1.1956 (0.8790) grad_norm 7.3441 (8.4999/2.0322) mem 68106MB [2022-12-19 19:45:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][420/1519] eta 0:18:27 lr 0.000024 time 0.9295 (1.0080) model_time 0.9293 (1.0056) loss 1.0186 (0.8801) grad_norm 9.8822 (8.5082/2.0324) mem 68106MB [2022-12-19 19:45:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][430/1519] eta 0:18:17 lr 0.000024 time 0.9358 (1.0078) model_time 0.9356 (1.0055) loss 0.9562 (0.8815) grad_norm 11.2537 (8.5331/2.0297) mem 68106MB [2022-12-19 19:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][440/1519] eta 0:18:07 lr 0.000024 time 0.9379 (1.0077) model_time 0.9377 (1.0054) loss 0.8389 (0.8814) grad_norm 19.3161 (8.5833/2.1418) mem 68106MB [2022-12-19 19:45:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][450/1519] eta 0:17:57 lr 0.000024 time 0.9787 (1.0077) model_time 0.9786 (1.0055) loss 0.8908 (0.8807) grad_norm 7.6422 (8.5833/2.1197) mem 68106MB [2022-12-19 19:46:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][460/1519] eta 0:17:47 lr 0.000024 time 0.9166 (1.0076) model_time 0.9164 (1.0054) loss 0.6728 (0.8807) grad_norm 9.9536 (8.6280/2.1398) mem 68106MB [2022-12-19 19:46:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][470/1519] eta 0:17:37 lr 0.000024 time 0.9429 (1.0077) model_time 0.9428 (1.0055) loss 0.6874 (0.8792) grad_norm 8.5110 (8.6247/2.1191) mem 68106MB [2022-12-19 19:46:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][480/1519] eta 0:17:26 lr 0.000024 time 0.9026 (1.0077) model_time 0.9024 (1.0055) loss 0.9000 (0.8792) grad_norm 8.6754 (8.6205/2.1043) mem 68106MB [2022-12-19 19:46:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][490/1519] eta 0:17:16 lr 0.000024 time 0.9413 (1.0077) model_time 0.9411 (1.0056) loss 0.9559 (0.8784) grad_norm 7.8945 (8.6175/2.0894) mem 68106MB [2022-12-19 19:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][500/1519] eta 0:17:06 lr 0.000024 time 0.9267 (1.0077) model_time 0.9265 (1.0057) loss 0.8018 (0.8781) grad_norm 8.5382 (8.5950/2.0874) mem 68106MB [2022-12-19 19:46:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][510/1519] eta 0:16:56 lr 0.000024 time 0.9234 (1.0076) model_time 0.9233 (1.0056) loss 0.7226 (0.8771) grad_norm 11.2261 (8.6220/2.1052) mem 68106MB [2022-12-19 19:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][520/1519] eta 0:16:46 lr 0.000024 time 0.9346 (1.0075) model_time 0.9344 (1.0055) loss 0.8911 (0.8769) grad_norm 9.4174 (8.6447/2.1103) mem 68106MB [2022-12-19 19:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][530/1519] eta 0:16:36 lr 0.000024 time 0.9363 (1.0079) model_time 0.9361 (1.0059) loss 0.9366 (0.8776) grad_norm 10.7448 (8.6564/2.1013) mem 68106MB [2022-12-19 19:47:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][540/1519] eta 0:16:26 lr 0.000024 time 0.9308 (1.0078) model_time 0.9286 (1.0058) loss 0.9033 (0.8765) grad_norm 7.7646 (8.6355/2.0916) mem 68106MB [2022-12-19 19:47:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][550/1519] eta 0:16:16 lr 0.000024 time 0.9248 (1.0076) model_time 0.9246 (1.0057) loss 0.8794 (0.8766) grad_norm 10.6916 (8.6512/2.0797) mem 68106MB [2022-12-19 19:47:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][560/1519] eta 0:16:06 lr 0.000024 time 0.9294 (1.0075) model_time 0.9292 (1.0056) loss 0.6686 (0.8762) grad_norm 8.1032 (8.6431/2.0713) mem 68106MB [2022-12-19 19:47:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][570/1519] eta 0:15:56 lr 0.000024 time 0.9298 (1.0076) model_time 0.9296 (1.0057) loss 0.7863 (0.8772) grad_norm 7.0118 (8.6321/2.0584) mem 68106MB [2022-12-19 19:48:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][580/1519] eta 0:15:46 lr 0.000024 time 0.9482 (1.0076) model_time 0.9481 (1.0058) loss 0.6908 (0.8763) grad_norm 9.2477 (8.6260/2.0425) mem 68106MB [2022-12-19 19:48:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][590/1519] eta 0:15:35 lr 0.000024 time 0.9311 (1.0075) model_time 0.9309 (1.0057) loss 0.8025 (0.8766) grad_norm 11.0734 (8.6256/2.0439) mem 68106MB [2022-12-19 19:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][600/1519] eta 0:15:25 lr 0.000024 time 0.9304 (1.0076) model_time 0.9302 (1.0058) loss 0.8072 (0.8782) grad_norm 8.8187 (8.6145/2.0336) mem 68106MB [2022-12-19 19:48:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][610/1519] eta 0:15:15 lr 0.000024 time 0.9294 (1.0074) model_time 0.9292 (1.0057) loss 0.6860 (0.8778) grad_norm 7.6560 (8.5960/2.0314) mem 68106MB [2022-12-19 19:48:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][620/1519] eta 0:15:05 lr 0.000024 time 0.9306 (1.0074) model_time 0.9305 (1.0056) loss 0.7214 (0.8769) grad_norm 7.1181 (8.6309/2.0353) mem 68106MB [2022-12-19 19:48:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][630/1519] eta 0:14:55 lr 0.000024 time 0.9261 (1.0075) model_time 0.9259 (1.0057) loss 0.7629 (0.8755) grad_norm 9.9055 (8.6017/2.0260) mem 68106MB [2022-12-19 19:49:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][640/1519] eta 0:14:45 lr 0.000024 time 0.9267 (1.0073) model_time 0.9265 (1.0056) loss 0.8598 (0.8758) grad_norm 11.5435 (8.6240/2.0246) mem 68106MB [2022-12-19 19:49:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][650/1519] eta 0:14:35 lr 0.000024 time 0.9247 (1.0073) model_time 0.9245 (1.0056) loss 0.8011 (0.8754) grad_norm 7.7140 (8.6047/2.0249) mem 68106MB [2022-12-19 19:49:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][660/1519] eta 0:14:25 lr 0.000024 time 0.9287 (1.0072) model_time 0.9284 (1.0055) loss 0.8701 (0.8749) grad_norm 11.4761 (8.6052/2.0214) mem 68106MB [2022-12-19 19:49:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][670/1519] eta 0:14:15 lr 0.000024 time 0.9211 (1.0072) model_time 0.9209 (1.0056) loss 0.9553 (0.8740) grad_norm 10.1288 (8.6153/2.0080) mem 68106MB [2022-12-19 19:49:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][680/1519] eta 0:14:05 lr 0.000024 time 0.9349 (1.0072) model_time 0.9346 (1.0056) loss 0.7046 (0.8738) grad_norm 8.1955 (8.5986/2.0114) mem 68106MB [2022-12-19 19:49:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][690/1519] eta 0:13:54 lr 0.000024 time 0.9318 (1.0071) model_time 0.9317 (1.0055) loss 0.8198 (0.8732) grad_norm 8.0800 (8.5956/1.9869) mem 68106MB [2022-12-19 19:50:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][700/1519] eta 0:13:44 lr 0.000024 time 0.9155 (1.0073) model_time 0.9153 (1.0057) loss 0.7654 (0.8728) grad_norm 10.3017 (8.5828/1.9715) mem 68106MB [2022-12-19 19:50:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][710/1519] eta 0:13:34 lr 0.000024 time 0.9305 (1.0073) model_time 0.9303 (1.0057) loss 1.0202 (0.8725) grad_norm 6.1639 (8.5735/1.9711) mem 68106MB [2022-12-19 19:50:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][720/1519] eta 0:13:24 lr 0.000024 time 0.9594 (1.0073) model_time 0.9592 (1.0057) loss 0.8569 (0.8724) grad_norm 7.3044 (8.6101/2.0655) mem 68106MB [2022-12-19 19:50:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][730/1519] eta 0:13:14 lr 0.000024 time 0.9386 (1.0072) model_time 0.9384 (1.0056) loss 0.7011 (0.8731) grad_norm 6.9264 (8.6044/2.0605) mem 68106MB [2022-12-19 19:50:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][740/1519] eta 0:13:04 lr 0.000024 time 0.9336 (1.0071) model_time 0.9334 (1.0055) loss 0.9190 (0.8717) grad_norm 7.2495 (8.6172/2.0804) mem 68106MB [2022-12-19 19:50:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][750/1519] eta 0:12:54 lr 0.000024 time 0.9339 (1.0070) model_time 0.9337 (1.0055) loss 0.8023 (0.8717) grad_norm 6.9053 (8.6254/2.0911) mem 68106MB [2022-12-19 19:51:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][760/1519] eta 0:12:44 lr 0.000024 time 0.9158 (1.0069) model_time 0.9156 (1.0054) loss 0.8704 (0.8713) grad_norm 6.8506 (8.6299/2.0974) mem 68106MB [2022-12-19 19:51:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][770/1519] eta 0:12:34 lr 0.000024 time 0.9544 (1.0069) model_time 0.9541 (1.0054) loss 0.9025 (0.8709) grad_norm 9.9176 (8.6514/2.0949) mem 68106MB [2022-12-19 19:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][780/1519] eta 0:12:24 lr 0.000024 time 0.9330 (1.0069) model_time 0.9329 (1.0054) loss 0.8462 (0.8705) grad_norm 9.2221 (8.6668/2.0840) mem 68106MB [2022-12-19 19:51:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][790/1519] eta 0:12:13 lr 0.000024 time 0.9343 (1.0068) model_time 0.9341 (1.0053) loss 1.1387 (0.8705) grad_norm 6.7245 (8.6456/2.0938) mem 68106MB [2022-12-19 19:51:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][800/1519] eta 0:12:03 lr 0.000024 time 0.9502 (1.0069) model_time 0.9501 (1.0054) loss 1.1713 (0.8708) grad_norm 16.5001 (8.6786/2.1389) mem 68106MB [2022-12-19 19:51:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][810/1519] eta 0:11:53 lr 0.000024 time 0.9367 (1.0068) model_time 0.9366 (1.0054) loss 1.0462 (0.8708) grad_norm 7.4983 (8.6689/2.1330) mem 68106MB [2022-12-19 19:52:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][820/1519] eta 0:11:43 lr 0.000024 time 0.9297 (1.0067) model_time 0.9295 (1.0053) loss 0.8028 (0.8708) grad_norm 7.5645 (8.6844/2.1287) mem 68106MB [2022-12-19 19:52:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][830/1519] eta 0:11:33 lr 0.000024 time 0.9364 (1.0067) model_time 0.9363 (1.0053) loss 0.6944 (0.8713) grad_norm 11.7218 (8.7502/2.1594) mem 68106MB [2022-12-19 19:52:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][840/1519] eta 0:11:23 lr 0.000024 time 0.9360 (1.0067) model_time 0.9359 (1.0053) loss 0.6973 (0.8704) grad_norm 7.8718 (8.7577/2.1484) mem 68106MB [2022-12-19 19:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][850/1519] eta 0:11:13 lr 0.000024 time 0.9493 (1.0067) model_time 0.9492 (1.0053) loss 0.6736 (0.8715) grad_norm 14.1792 (8.7492/2.0993) mem 68106MB [2022-12-19 19:52:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][860/1519] eta 0:11:03 lr 0.000024 time 0.9313 (1.0070) model_time 0.9312 (1.0056) loss 0.9107 (0.8712) grad_norm 6.1505 (8.7309/2.0858) mem 68106MB [2022-12-19 19:52:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][870/1519] eta 0:10:53 lr 0.000024 time 0.9397 (1.0070) model_time 0.9391 (1.0056) loss 0.9279 (0.8715) grad_norm 7.0791 (8.7652/2.0886) mem 68106MB [2022-12-19 19:53:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][880/1519] eta 0:10:43 lr 0.000024 time 0.9374 (1.0070) model_time 0.9373 (1.0056) loss 0.9549 (0.8712) grad_norm 8.1935 (8.7413/2.0661) mem 68106MB [2022-12-19 19:53:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][890/1519] eta 0:10:33 lr 0.000024 time 0.9381 (1.0070) model_time 0.9380 (1.0057) loss 0.9026 (0.8708) grad_norm 7.9213 (8.7362/2.0664) mem 68106MB [2022-12-19 19:53:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][900/1519] eta 0:10:23 lr 0.000024 time 0.9453 (1.0069) model_time 0.9452 (1.0056) loss 0.9417 (0.8711) grad_norm 9.5143 (8.7366/2.0538) mem 68106MB [2022-12-19 19:53:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][910/1519] eta 0:10:13 lr 0.000024 time 0.9402 (1.0070) model_time 0.9400 (1.0057) loss 0.7021 (0.8706) grad_norm 7.4969 (8.7681/2.0628) mem 68106MB [2022-12-19 19:53:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][920/1519] eta 0:10:03 lr 0.000024 time 0.9297 (1.0069) model_time 0.9295 (1.0056) loss 0.7879 (0.8706) grad_norm 6.9136 (8.7579/2.0770) mem 68106MB [2022-12-19 19:53:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][930/1519] eta 0:09:53 lr 0.000024 time 0.9351 (1.0069) model_time 0.9350 (1.0056) loss 1.1706 (0.8710) grad_norm 7.6748 (8.7540/2.0727) mem 68106MB [2022-12-19 19:54:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][940/1519] eta 0:09:42 lr 0.000024 time 0.9334 (1.0069) model_time 0.9332 (1.0056) loss 0.8158 (0.8713) grad_norm 7.7352 (8.7674/2.0713) mem 68106MB [2022-12-19 19:54:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][950/1519] eta 0:09:32 lr 0.000024 time 0.9314 (1.0068) model_time 0.9312 (1.0055) loss 0.7948 (0.8708) grad_norm 10.8072 (8.7972/2.0858) mem 68106MB [2022-12-19 19:54:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][960/1519] eta 0:09:22 lr 0.000024 time 0.9134 (1.0068) model_time 0.9133 (1.0055) loss 1.0656 (0.8712) grad_norm 8.5306 (8.8294/2.0862) mem 68106MB [2022-12-19 19:54:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][970/1519] eta 0:09:12 lr 0.000024 time 0.9374 (1.0067) model_time 0.9372 (1.0055) loss 0.8774 (0.8707) grad_norm 10.4826 (8.8183/2.0945) mem 68106MB [2022-12-19 19:54:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][980/1519] eta 0:09:02 lr 0.000024 time 0.9337 (1.0067) model_time 0.9336 (1.0054) loss 0.9074 (0.8717) grad_norm 7.8455 (8.8144/2.0840) mem 68106MB [2022-12-19 19:54:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][990/1519] eta 0:08:52 lr 0.000024 time 0.9299 (1.0066) model_time 0.9298 (1.0054) loss 0.7075 (0.8713) grad_norm 11.8982 (8.7559/2.0050) mem 68106MB [2022-12-19 19:55:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1000/1519] eta 0:08:42 lr 0.000024 time 0.9327 (1.0067) model_time 0.9326 (1.0054) loss 0.6959 (0.8723) grad_norm 10.4534 (8.7652/2.0059) mem 68106MB [2022-12-19 19:55:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1010/1519] eta 0:08:32 lr 0.000024 time 0.9115 (1.0067) model_time 0.9114 (1.0055) loss 1.0558 (0.8728) grad_norm 8.8458 (8.7578/2.0014) mem 68106MB [2022-12-19 19:55:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1020/1519] eta 0:08:22 lr 0.000024 time 0.9420 (1.0069) model_time 0.9419 (1.0057) loss 0.7208 (0.8721) grad_norm 9.3028 (8.7537/1.9889) mem 68106MB [2022-12-19 19:55:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1030/1519] eta 0:08:12 lr 0.000024 time 0.9314 (1.0069) model_time 0.9312 (1.0057) loss 0.8380 (0.8718) grad_norm 8.6446 (8.7363/1.9899) mem 68106MB [2022-12-19 19:55:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1040/1519] eta 0:08:02 lr 0.000024 time 0.9333 (1.0069) model_time 0.9331 (1.0057) loss 0.6906 (0.8721) grad_norm 5.7873 (8.6900/1.9060) mem 68106MB [2022-12-19 19:56:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1050/1519] eta 0:07:52 lr 0.000024 time 0.9255 (1.0069) model_time 0.9253 (1.0057) loss 0.6846 (0.8715) grad_norm 7.0755 (8.6889/1.9272) mem 68106MB [2022-12-19 19:56:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1060/1519] eta 0:07:42 lr 0.000024 time 0.9311 (1.0068) model_time 0.9309 (1.0056) loss 0.9181 (0.8716) grad_norm 7.2659 (8.6439/1.8984) mem 68106MB [2022-12-19 19:56:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1070/1519] eta 0:07:32 lr 0.000024 time 0.9359 (1.0067) model_time 0.9357 (1.0056) loss 1.0527 (0.8721) grad_norm 8.6663 (8.6295/1.9042) mem 68106MB [2022-12-19 19:56:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1080/1519] eta 0:07:21 lr 0.000024 time 0.9314 (1.0067) model_time 0.9313 (1.0055) loss 0.8049 (0.8721) grad_norm 8.3518 (8.6413/1.9052) mem 68106MB [2022-12-19 19:56:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1090/1519] eta 0:07:11 lr 0.000024 time 0.9337 (1.0068) model_time 0.9334 (1.0056) loss 0.7896 (0.8717) grad_norm 6.6233 (8.6300/1.9136) mem 68106MB [2022-12-19 19:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1100/1519] eta 0:07:01 lr 0.000024 time 0.9327 (1.0067) model_time 0.9326 (1.0055) loss 0.9333 (0.8712) grad_norm 4.9961 (8.6334/1.9099) mem 68106MB [2022-12-19 19:57:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1110/1519] eta 0:06:51 lr 0.000024 time 0.9351 (1.0069) model_time 0.9350 (1.0058) loss 1.1467 (0.8725) grad_norm 14.2428 (8.6227/1.9086) mem 68106MB [2022-12-19 19:57:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1120/1519] eta 0:06:41 lr 0.000024 time 0.9381 (1.0069) model_time 0.9379 (1.0057) loss 0.7415 (0.8732) grad_norm 6.7755 (8.6080/1.9013) mem 68106MB [2022-12-19 19:57:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1130/1519] eta 0:06:31 lr 0.000024 time 0.9330 (1.0068) model_time 0.9329 (1.0057) loss 0.7448 (0.8732) grad_norm 9.2836 (8.5863/1.9000) mem 68106MB [2022-12-19 19:57:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1140/1519] eta 0:06:21 lr 0.000024 time 0.9322 (1.0067) model_time 0.9321 (1.0056) loss 1.1894 (0.8725) grad_norm 6.7311 (8.5969/1.9056) mem 68106MB [2022-12-19 19:57:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1150/1519] eta 0:06:11 lr 0.000024 time 0.9318 (1.0066) model_time 0.9317 (1.0055) loss 1.1337 (0.8725) grad_norm 6.5754 (8.5837/1.9077) mem 68106MB [2022-12-19 19:57:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1160/1519] eta 0:06:01 lr 0.000024 time 0.9388 (1.0066) model_time 0.9386 (1.0055) loss 0.7563 (0.8724) grad_norm 6.8500 (8.5771/1.9038) mem 68106MB [2022-12-19 19:58:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1170/1519] eta 0:05:51 lr 0.000024 time 0.9405 (1.0066) model_time 0.9404 (1.0055) loss 1.0063 (0.8719) grad_norm 8.8117 (8.5891/1.9022) mem 68106MB [2022-12-19 19:58:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1180/1519] eta 0:05:41 lr 0.000024 time 0.9324 (1.0066) model_time 0.9323 (1.0055) loss 0.8177 (0.8724) grad_norm 6.4825 (8.5716/1.9094) mem 68106MB [2022-12-19 19:58:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1190/1519] eta 0:05:31 lr 0.000024 time 0.9319 (1.0067) model_time 0.9317 (1.0056) loss 0.9200 (0.8733) grad_norm 8.1703 (8.5911/1.8991) mem 68106MB [2022-12-19 19:58:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1200/1519] eta 0:05:21 lr 0.000024 time 0.9255 (1.0067) model_time 0.9254 (1.0056) loss 0.7415 (0.8732) grad_norm 8.0020 (8.6010/1.9125) mem 68106MB [2022-12-19 19:58:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1210/1519] eta 0:05:11 lr 0.000024 time 0.9366 (1.0066) model_time 0.9364 (1.0055) loss 1.0404 (0.8740) grad_norm 6.3888 (8.6047/1.9112) mem 68106MB [2022-12-19 19:58:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1220/1519] eta 0:05:00 lr 0.000024 time 0.9314 (1.0066) model_time 0.9313 (1.0056) loss 0.6763 (0.8741) grad_norm 13.4930 (8.6035/1.9003) mem 68106MB [2022-12-19 19:59:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1230/1519] eta 0:04:50 lr 0.000024 time 0.9368 (1.0066) model_time 0.9366 (1.0055) loss 1.1635 (0.8740) grad_norm 7.1707 (8.6041/1.9119) mem 68106MB [2022-12-19 19:59:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1240/1519] eta 0:04:40 lr 0.000024 time 0.9265 (1.0066) model_time 0.9264 (1.0055) loss 0.9773 (0.8745) grad_norm 8.7655 (8.5799/1.9083) mem 68106MB [2022-12-19 19:59:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1250/1519] eta 0:04:30 lr 0.000024 time 1.0017 (1.0066) model_time 1.0016 (1.0055) loss 0.7169 (0.8746) grad_norm 8.2511 (8.6028/1.9068) mem 68106MB [2022-12-19 19:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1260/1519] eta 0:04:20 lr 0.000024 time 0.9305 (1.0065) model_time 0.9304 (1.0054) loss 0.7520 (0.8747) grad_norm 6.3493 (8.6063/1.9166) mem 68106MB [2022-12-19 19:59:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1270/1519] eta 0:04:10 lr 0.000024 time 0.9516 (1.0065) model_time 0.9515 (1.0055) loss 0.8523 (0.8745) grad_norm 6.0770 (8.5940/1.9202) mem 68106MB [2022-12-19 19:59:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1280/1519] eta 0:04:00 lr 0.000024 time 0.9300 (1.0065) model_time 0.9299 (1.0054) loss 0.9600 (0.8740) grad_norm 6.8052 (8.6229/1.9704) mem 68106MB [2022-12-19 20:00:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1290/1519] eta 0:03:50 lr 0.000024 time 0.9405 (1.0064) model_time 0.9404 (1.0054) loss 0.8721 (0.8741) grad_norm 7.1728 (8.6237/1.9695) mem 68106MB [2022-12-19 20:00:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1300/1519] eta 0:03:40 lr 0.000024 time 0.9276 (1.0064) model_time 0.9275 (1.0053) loss 0.7512 (0.8739) grad_norm 7.4446 (8.6055/1.9751) mem 68106MB [2022-12-19 20:00:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1310/1519] eta 0:03:30 lr 0.000024 time 0.9285 (1.0064) model_time 0.9283 (1.0054) loss 0.8597 (0.8742) grad_norm 8.1414 (8.6170/1.9864) mem 68106MB [2022-12-19 20:00:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1320/1519] eta 0:03:20 lr 0.000024 time 0.9334 (1.0063) model_time 0.9333 (1.0053) loss 0.6983 (0.8740) grad_norm 6.5381 (8.5663/1.8903) mem 68106MB [2022-12-19 20:00:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1330/1519] eta 0:03:10 lr 0.000024 time 0.9940 (1.0064) model_time 0.9939 (1.0054) loss 0.9580 (0.8738) grad_norm 9.2489 (8.5577/1.8901) mem 68106MB [2022-12-19 20:00:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1340/1519] eta 0:03:00 lr 0.000024 time 1.1943 (1.0066) model_time 1.1942 (1.0056) loss 0.7324 (0.8742) grad_norm 12.8380 (8.5704/1.8747) mem 68106MB [2022-12-19 20:01:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1350/1519] eta 0:02:50 lr 0.000024 time 0.9277 (1.0065) model_time 0.9276 (1.0055) loss 0.8717 (0.8735) grad_norm 10.6863 (8.5577/1.8669) mem 68106MB [2022-12-19 20:01:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1360/1519] eta 0:02:40 lr 0.000024 time 0.9374 (1.0065) model_time 0.9373 (1.0054) loss 0.8373 (0.8741) grad_norm 6.7364 (8.5554/1.8518) mem 68106MB [2022-12-19 20:01:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1370/1519] eta 0:02:29 lr 0.000024 time 0.9383 (1.0065) model_time 0.9381 (1.0055) loss 1.3502 (0.8742) grad_norm 7.1192 (8.5499/1.8469) mem 68106MB [2022-12-19 20:01:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1380/1519] eta 0:02:19 lr 0.000024 time 0.9277 (1.0065) model_time 0.9275 (1.0055) loss 0.9263 (0.8741) grad_norm 7.9110 (8.5370/1.8452) mem 68106MB [2022-12-19 20:01:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1390/1519] eta 0:02:09 lr 0.000024 time 0.9306 (1.0065) model_time 0.9304 (1.0055) loss 0.9360 (0.8744) grad_norm 6.0956 (8.5363/1.8403) mem 68106MB [2022-12-19 20:01:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1400/1519] eta 0:01:59 lr 0.000024 time 0.9659 (1.0066) model_time 0.9657 (1.0056) loss 0.8217 (0.8747) grad_norm 14.2921 (8.5677/1.8774) mem 68106MB [2022-12-19 20:02:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1410/1519] eta 0:01:49 lr 0.000024 time 0.9315 (1.0065) model_time 0.9313 (1.0056) loss 0.8359 (0.8744) grad_norm 8.5290 (8.5273/1.8276) mem 68106MB [2022-12-19 20:02:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1420/1519] eta 0:01:39 lr 0.000024 time 0.9290 (1.0065) model_time 0.9288 (1.0055) loss 0.7426 (0.8743) grad_norm 5.9594 (8.5501/1.8630) mem 68106MB [2022-12-19 20:02:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1430/1519] eta 0:01:29 lr 0.000024 time 1.1757 (1.0066) model_time 1.1756 (1.0057) loss 1.1893 (0.8746) grad_norm 10.4159 (8.5183/1.8254) mem 68106MB [2022-12-19 20:02:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1440/1519] eta 0:01:19 lr 0.000024 time 0.9275 (1.0066) model_time 0.9274 (1.0056) loss 0.9440 (0.8746) grad_norm 6.8193 (8.5032/1.8216) mem 68106MB [2022-12-19 20:02:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1450/1519] eta 0:01:09 lr 0.000024 time 0.9324 (1.0065) model_time 0.9323 (1.0056) loss 0.7634 (0.8741) grad_norm 9.8772 (8.5033/1.8184) mem 68106MB [2022-12-19 20:02:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1460/1519] eta 0:00:59 lr 0.000024 time 0.9345 (1.0065) model_time 0.9343 (1.0055) loss 1.0665 (0.8748) grad_norm 6.9178 (8.5052/1.8289) mem 68106MB [2022-12-19 20:03:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1470/1519] eta 0:00:49 lr 0.000024 time 0.9179 (1.0065) model_time 0.9177 (1.0055) loss 0.7899 (0.8749) grad_norm 10.4679 (8.4880/1.8150) mem 68106MB [2022-12-19 20:03:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1480/1519] eta 0:00:39 lr 0.000024 time 0.9340 (1.0064) model_time 0.9339 (1.0055) loss 0.9357 (0.8748) grad_norm 9.8543 (8.4941/1.8154) mem 68106MB [2022-12-19 20:03:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1490/1519] eta 0:00:29 lr 0.000024 time 0.9289 (1.0066) model_time 0.9287 (1.0056) loss 0.7032 (0.8748) grad_norm 10.2925 (8.5264/1.8238) mem 68106MB [2022-12-19 20:03:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1500/1519] eta 0:00:19 lr 0.000024 time 0.9276 (1.0065) model_time 0.9274 (1.0056) loss 0.7865 (0.8751) grad_norm 5.3794 (8.5146/1.8310) mem 68106MB [2022-12-19 20:03:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [35/100][1510/1519] eta 0:00:09 lr 0.000024 time 0.9247 (1.0065) model_time 0.9246 (1.0056) loss 0.9659 (0.8753) grad_norm 11.7047 (8.4998/1.8173) mem 68106MB [2022-12-19 20:03:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 35 training takes 0:25:28 [2022-12-19 20:03:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_35.pth saving...... [2022-12-19 20:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_35.pth saved !!! [2022-12-19 20:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.643 (0.643) Loss 0.5321 (0.5321) Acc@1 90.625 (90.625) Acc@5 98.264 (98.264) Mem 68106MB [2022-12-19 20:04:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.305 (0.331) Loss 0.4952 (0.4886) Acc@1 92.014 (91.856) Acc@5 97.569 (98.359) Mem 68106MB [2022-12-19 20:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.315) Loss 0.4255 (0.4874) Acc@1 92.708 (91.700) Acc@5 99.306 (98.363) Mem 68106MB [2022-12-19 20:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.5980 (0.4908) Acc@1 89.236 (91.543) Acc@5 97.569 (98.320) Mem 68106MB [2022-12-19 20:04:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.307) Loss 0.4455 (0.4835) Acc@1 92.361 (91.574) Acc@5 98.958 (98.408) Mem 68106MB [2022-12-19 20:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.305) Loss 0.4980 (0.4820) Acc@1 90.625 (91.537) Acc@5 99.653 (98.448) Mem 68106MB [2022-12-19 20:04:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.305 (0.305) Loss 0.4845 (0.4816) Acc@1 89.931 (91.570) Acc@5 98.264 (98.457) Mem 68106MB [2022-12-19 20:04:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5395 (0.4837) Acc@1 90.625 (91.535) Acc@5 97.917 (98.406) Mem 68106MB [2022-12-19 20:04:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.303) Loss 0.3827 (0.4809) Acc@1 94.097 (91.641) Acc@5 98.611 (98.448) Mem 68106MB [2022-12-19 20:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:35] * Acc@1 91.618 Acc@5 98.453 [2022-12-19 20:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.6% [2022-12-19 20:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.63% [2022-12-19 20:04:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][0/1519] eta 0:45:29 lr 0.000024 time 1.7968 (1.7968) model_time 1.0541 (1.0541) loss 0.8429 (0.8429) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 20:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][10/1519] eta 0:27:10 lr 0.000024 time 0.9270 (1.0805) model_time 0.9269 (1.0126) loss 0.8150 (0.8650) grad_norm 7.0547 (8.3156/0.7894) mem 68106MB [2022-12-19 20:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][20/1519] eta 0:26:02 lr 0.000024 time 0.9301 (1.0421) model_time 0.9299 (1.0064) loss 0.7829 (0.8537) grad_norm 6.0032 (7.9214/1.2390) mem 68106MB [2022-12-19 20:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][30/1519] eta 0:25:31 lr 0.000024 time 0.9227 (1.0283) model_time 0.9226 (1.0041) loss 0.8211 (0.8593) grad_norm 6.4116 (8.1693/1.6754) mem 68106MB [2022-12-19 20:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][40/1519] eta 0:25:10 lr 0.000024 time 0.9256 (1.0216) model_time 0.9254 (1.0032) loss 0.7061 (0.8556) grad_norm 8.2421 (8.2488/1.4892) mem 68106MB [2022-12-19 20:05:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][50/1519] eta 0:24:54 lr 0.000024 time 0.9249 (1.0171) model_time 0.9248 (1.0022) loss 1.0304 (0.8710) grad_norm 9.5141 (8.2956/1.5111) mem 68106MB [2022-12-19 20:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][60/1519] eta 0:24:41 lr 0.000024 time 0.9181 (1.0151) model_time 0.9179 (1.0026) loss 0.8869 (0.8708) grad_norm 7.9266 (9.3271/5.4304) mem 68106MB [2022-12-19 20:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][70/1519] eta 0:24:29 lr 0.000024 time 0.9344 (1.0140) model_time 0.9343 (1.0032) loss 0.7130 (0.8795) grad_norm 6.2845 (9.1863/5.0610) mem 68106MB [2022-12-19 20:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][80/1519] eta 0:24:19 lr 0.000024 time 0.9346 (1.0144) model_time 0.9345 (1.0049) loss 0.6963 (0.8760) grad_norm 12.4909 (9.2227/4.8140) mem 68106MB [2022-12-19 20:06:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][90/1519] eta 0:24:08 lr 0.000024 time 0.9292 (1.0137) model_time 0.9290 (1.0052) loss 0.6693 (0.8752) grad_norm 9.5108 (9.1961/4.5460) mem 68106MB [2022-12-19 20:06:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][100/1519] eta 0:23:57 lr 0.000024 time 0.9223 (1.0132) model_time 0.9222 (1.0056) loss 0.8252 (0.8748) grad_norm 6.9180 (9.0909/4.3383) mem 68106MB [2022-12-19 20:06:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][110/1519] eta 0:23:46 lr 0.000024 time 0.9259 (1.0122) model_time 0.9258 (1.0052) loss 1.0520 (0.8769) grad_norm 9.8853 (9.1107/4.1529) mem 68106MB [2022-12-19 20:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][120/1519] eta 0:23:36 lr 0.000024 time 0.9244 (1.0127) model_time 0.9243 (1.0063) loss 0.9213 (0.8820) grad_norm 9.0370 (9.0682/3.9894) mem 68106MB [2022-12-19 20:06:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][130/1519] eta 0:23:25 lr 0.000024 time 0.9290 (1.0122) model_time 0.9289 (1.0062) loss 1.0018 (0.8845) grad_norm 9.8624 (9.0697/3.8566) mem 68106MB [2022-12-19 20:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][140/1519] eta 0:23:14 lr 0.000024 time 0.9349 (1.0113) model_time 0.9347 (1.0057) loss 0.9312 (0.8872) grad_norm 7.5360 (9.0577/3.7238) mem 68106MB [2022-12-19 20:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][150/1519] eta 0:23:03 lr 0.000024 time 0.9273 (1.0104) model_time 0.9272 (1.0052) loss 1.1254 (0.8851) grad_norm 10.0592 (9.0332/3.6144) mem 68106MB [2022-12-19 20:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][160/1519] eta 0:22:52 lr 0.000024 time 0.9344 (1.0098) model_time 0.9342 (1.0048) loss 0.6817 (0.8848) grad_norm 8.3891 (9.0409/3.5396) mem 68106MB [2022-12-19 20:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][170/1519] eta 0:22:42 lr 0.000024 time 1.0065 (1.0098) model_time 1.0063 (1.0051) loss 0.9384 (0.8847) grad_norm 7.8199 (9.0141/3.4483) mem 68106MB [2022-12-19 20:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][180/1519] eta 0:22:31 lr 0.000024 time 0.9322 (1.0092) model_time 0.9320 (1.0048) loss 0.9106 (0.8849) grad_norm 10.4294 (9.0153/3.3628) mem 68106MB [2022-12-19 20:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][190/1519] eta 0:22:20 lr 0.000024 time 0.9291 (1.0089) model_time 0.9290 (1.0046) loss 0.8767 (0.8853) grad_norm 8.0583 (8.9643/3.3063) mem 68106MB [2022-12-19 20:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][200/1519] eta 0:22:10 lr 0.000024 time 0.9523 (1.0087) model_time 0.9521 (1.0046) loss 0.9222 (0.8879) grad_norm 8.3661 (8.9053/3.2358) mem 68106MB [2022-12-19 20:08:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][210/1519] eta 0:21:59 lr 0.000024 time 0.9301 (1.0082) model_time 0.9299 (1.0043) loss 0.8209 (0.8862) grad_norm 7.3831 (8.8639/3.1658) mem 68106MB [2022-12-19 20:08:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][220/1519] eta 0:21:49 lr 0.000024 time 0.9208 (1.0081) model_time 0.9206 (1.0043) loss 0.7464 (0.8853) grad_norm 8.1576 (8.8104/3.1167) mem 68106MB [2022-12-19 20:08:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][230/1519] eta 0:21:39 lr 0.000024 time 0.9199 (1.0078) model_time 0.9198 (1.0043) loss 0.8506 (0.8847) grad_norm 7.2790 (8.8334/3.0584) mem 68106MB [2022-12-19 20:08:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][240/1519] eta 0:21:28 lr 0.000024 time 0.9354 (1.0077) model_time 0.9353 (1.0042) loss 0.9897 (0.8870) grad_norm 9.0346 (8.8167/2.9970) mem 68106MB [2022-12-19 20:08:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][250/1519] eta 0:21:18 lr 0.000024 time 0.9329 (1.0075) model_time 0.9327 (1.0042) loss 0.8800 (0.8871) grad_norm 7.9929 (8.8271/2.9454) mem 68106MB [2022-12-19 20:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][260/1519] eta 0:21:09 lr 0.000024 time 0.9280 (1.0083) model_time 0.9278 (1.0051) loss 0.9814 (0.8859) grad_norm 7.5552 (8.7758/2.9033) mem 68106MB [2022-12-19 20:09:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][270/1519] eta 0:20:59 lr 0.000024 time 0.9270 (1.0085) model_time 0.9268 (1.0054) loss 1.0120 (0.8875) grad_norm 7.4826 (8.7665/2.8705) mem 68106MB [2022-12-19 20:09:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][280/1519] eta 0:20:49 lr 0.000024 time 0.9204 (1.0084) model_time 0.9202 (1.0054) loss 0.7069 (0.8883) grad_norm 11.2558 (8.7828/2.8316) mem 68106MB [2022-12-19 20:09:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][290/1519] eta 0:20:39 lr 0.000024 time 0.9384 (1.0081) model_time 0.9383 (1.0052) loss 0.6899 (0.8876) grad_norm 10.0491 (8.7742/2.7989) mem 68106MB [2022-12-19 20:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][300/1519] eta 0:20:28 lr 0.000024 time 0.9360 (1.0080) model_time 0.9358 (1.0052) loss 0.9606 (0.8878) grad_norm 12.8498 (8.8238/2.7810) mem 68106MB [2022-12-19 20:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][310/1519] eta 0:20:18 lr 0.000024 time 0.9201 (1.0077) model_time 0.9200 (1.0050) loss 0.9444 (0.8866) grad_norm 9.1576 (8.8725/2.8111) mem 68106MB [2022-12-19 20:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][320/1519] eta 0:20:08 lr 0.000024 time 0.9305 (1.0076) model_time 0.9304 (1.0049) loss 0.9074 (0.8893) grad_norm 9.3381 (8.8616/2.7808) mem 68106MB [2022-12-19 20:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][330/1519] eta 0:19:57 lr 0.000024 time 0.9449 (1.0074) model_time 0.9447 (1.0048) loss 0.9270 (0.8911) grad_norm 7.1699 (8.9115/2.8547) mem 68106MB [2022-12-19 20:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][340/1519] eta 0:19:47 lr 0.000024 time 0.9194 (1.0071) model_time 0.9192 (1.0046) loss 1.0966 (0.8906) grad_norm 9.1090 (8.8966/2.8244) mem 68106MB [2022-12-19 20:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][350/1519] eta 0:19:37 lr 0.000024 time 1.0143 (1.0073) model_time 1.0142 (1.0048) loss 1.0841 (0.8905) grad_norm 8.6639 (8.9090/2.7908) mem 68106MB [2022-12-19 20:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][360/1519] eta 0:19:27 lr 0.000024 time 0.9228 (1.0070) model_time 0.9227 (1.0046) loss 0.8545 (0.8909) grad_norm 12.5272 (8.9226/2.7868) mem 68106MB [2022-12-19 20:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][370/1519] eta 0:19:16 lr 0.000024 time 0.9171 (1.0068) model_time 0.9169 (1.0045) loss 0.8108 (0.8885) grad_norm 7.5811 (8.8950/2.7574) mem 68106MB [2022-12-19 20:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][380/1519] eta 0:19:06 lr 0.000024 time 0.9294 (1.0068) model_time 0.9293 (1.0045) loss 0.9205 (0.8897) grad_norm 8.2925 (8.8757/2.7240) mem 68106MB [2022-12-19 20:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][390/1519] eta 0:18:56 lr 0.000024 time 0.9255 (1.0069) model_time 0.9254 (1.0046) loss 0.8731 (0.8884) grad_norm 6.4635 (8.8625/2.7084) mem 68106MB [2022-12-19 20:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][400/1519] eta 0:18:46 lr 0.000024 time 0.9286 (1.0069) model_time 0.9285 (1.0046) loss 1.3214 (0.8896) grad_norm 7.7741 (8.8389/2.6872) mem 68106MB [2022-12-19 20:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][410/1519] eta 0:18:36 lr 0.000024 time 1.0280 (1.0070) model_time 1.0278 (1.0048) loss 0.7857 (0.8898) grad_norm 7.2212 (8.8442/2.7055) mem 68106MB [2022-12-19 20:11:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][420/1519] eta 0:18:26 lr 0.000024 time 0.9315 (1.0069) model_time 0.9314 (1.0047) loss 0.7340 (0.8890) grad_norm 12.3211 (8.8611/2.6987) mem 68106MB [2022-12-19 20:11:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][430/1519] eta 0:18:16 lr 0.000024 time 0.9319 (1.0070) model_time 0.9317 (1.0049) loss 1.0827 (0.8885) grad_norm 11.7231 (8.8866/2.6934) mem 68106MB [2022-12-19 20:12:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][440/1519] eta 0:18:06 lr 0.000024 time 0.9289 (1.0069) model_time 0.9287 (1.0048) loss 0.9037 (0.8880) grad_norm 8.1126 (8.8695/2.6656) mem 68106MB [2022-12-19 20:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][450/1519] eta 0:17:56 lr 0.000024 time 0.9212 (1.0073) model_time 0.9210 (1.0053) loss 1.1753 (0.8902) grad_norm 9.7474 (8.8892/2.6443) mem 68106MB [2022-12-19 20:12:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][460/1519] eta 0:17:46 lr 0.000024 time 0.9268 (1.0072) model_time 0.9266 (1.0052) loss 0.9101 (0.8908) grad_norm 5.1070 (8.8653/2.6334) mem 68106MB [2022-12-19 20:12:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][470/1519] eta 0:17:36 lr 0.000024 time 0.9286 (1.0070) model_time 0.9285 (1.0051) loss 0.7685 (0.8918) grad_norm 7.2510 (8.8400/2.6187) mem 68106MB [2022-12-19 20:12:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][480/1519] eta 0:17:26 lr 0.000024 time 0.9404 (1.0071) model_time 0.9403 (1.0052) loss 0.8583 (0.8923) grad_norm 6.4581 (8.7953/2.6135) mem 68106MB [2022-12-19 20:12:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][490/1519] eta 0:17:16 lr 0.000024 time 0.9259 (1.0071) model_time 0.9258 (1.0052) loss 0.8576 (0.8912) grad_norm 8.3952 (8.7886/2.5975) mem 68106MB [2022-12-19 20:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][500/1519] eta 0:17:06 lr 0.000024 time 0.9198 (1.0072) model_time 0.9197 (1.0053) loss 0.7523 (0.8899) grad_norm 7.9194 (8.7941/2.5789) mem 68106MB [2022-12-19 20:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][510/1519] eta 0:16:56 lr 0.000024 time 0.9341 (1.0070) model_time 0.9340 (1.0052) loss 0.9221 (0.8902) grad_norm 9.9952 (8.7983/2.5606) mem 68106MB [2022-12-19 20:13:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][520/1519] eta 0:16:45 lr 0.000024 time 0.9273 (1.0069) model_time 0.9272 (1.0051) loss 0.9547 (0.8910) grad_norm 6.5685 (8.7949/2.5504) mem 68106MB [2022-12-19 20:13:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][530/1519] eta 0:16:35 lr 0.000024 time 0.9206 (1.0067) model_time 0.9204 (1.0050) loss 0.7191 (0.8899) grad_norm 7.2151 (8.7967/2.5293) mem 68106MB [2022-12-19 20:13:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][540/1519] eta 0:16:25 lr 0.000024 time 0.9241 (1.0066) model_time 0.9240 (1.0048) loss 0.7609 (0.8895) grad_norm 8.5925 (8.7730/2.5172) mem 68106MB [2022-12-19 20:13:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][550/1519] eta 0:16:15 lr 0.000024 time 0.9169 (1.0065) model_time 0.9167 (1.0048) loss 1.1262 (0.8904) grad_norm 9.8697 (8.7652/2.5021) mem 68106MB [2022-12-19 20:14:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][560/1519] eta 0:16:05 lr 0.000024 time 0.9367 (1.0065) model_time 0.9366 (1.0048) loss 0.6944 (0.8895) grad_norm 9.4913 (8.7554/2.4825) mem 68106MB [2022-12-19 20:14:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][570/1519] eta 0:15:55 lr 0.000024 time 0.9296 (1.0065) model_time 0.9294 (1.0048) loss 0.9332 (0.8894) grad_norm 8.7997 (8.7450/2.4639) mem 68106MB [2022-12-19 20:14:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][580/1519] eta 0:15:45 lr 0.000024 time 0.9955 (1.0068) model_time 0.9953 (1.0051) loss 0.7636 (0.8891) grad_norm 17.1647 (8.7575/2.4993) mem 68106MB [2022-12-19 20:14:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][590/1519] eta 0:15:35 lr 0.000024 time 0.9711 (1.0067) model_time 0.9709 (1.0051) loss 0.7697 (0.8897) grad_norm 9.8787 (8.7499/2.4864) mem 68106MB [2022-12-19 20:14:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][600/1519] eta 0:15:25 lr 0.000024 time 0.9276 (1.0067) model_time 0.9274 (1.0051) loss 0.9088 (0.8885) grad_norm 8.3121 (8.7355/2.4744) mem 68106MB [2022-12-19 20:14:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][610/1519] eta 0:15:14 lr 0.000024 time 0.9308 (1.0066) model_time 0.9307 (1.0050) loss 0.9097 (0.8878) grad_norm 8.7540 (8.7447/2.4731) mem 68106MB [2022-12-19 20:15:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][620/1519] eta 0:15:04 lr 0.000024 time 0.9290 (1.0064) model_time 0.9288 (1.0049) loss 0.7968 (0.8859) grad_norm 8.0200 (8.7511/2.4639) mem 68106MB [2022-12-19 20:15:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][630/1519] eta 0:14:54 lr 0.000024 time 0.9268 (1.0063) model_time 0.9266 (1.0048) loss 0.8294 (0.8854) grad_norm 7.9591 (8.7394/2.4513) mem 68106MB [2022-12-19 20:15:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][640/1519] eta 0:14:44 lr 0.000024 time 0.9210 (1.0062) model_time 0.9208 (1.0047) loss 0.7175 (0.8847) grad_norm 10.0399 (8.7229/2.4626) mem 68106MB [2022-12-19 20:15:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][650/1519] eta 0:14:34 lr 0.000024 time 0.9246 (1.0061) model_time 0.9245 (1.0046) loss 0.8780 (0.8839) grad_norm 7.6727 (8.7316/2.4623) mem 68106MB [2022-12-19 20:15:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][660/1519] eta 0:14:24 lr 0.000024 time 0.8898 (1.0062) model_time 0.8897 (1.0047) loss 0.8461 (0.8848) grad_norm 14.2077 (8.6852/1.9219) mem 68106MB [2022-12-19 20:15:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][670/1519] eta 0:14:14 lr 0.000024 time 0.9241 (1.0062) model_time 0.9240 (1.0047) loss 0.8777 (0.8849) grad_norm 7.4555 (8.6764/1.9292) mem 68106MB [2022-12-19 20:16:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][680/1519] eta 0:14:04 lr 0.000024 time 0.9275 (1.0062) model_time 0.9273 (1.0047) loss 0.7757 (0.8850) grad_norm 10.4834 (8.6940/1.9273) mem 68106MB [2022-12-19 20:16:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][690/1519] eta 0:13:54 lr 0.000024 time 1.0587 (1.0063) model_time 1.0584 (1.0048) loss 0.8468 (0.8844) grad_norm 7.3980 (8.6935/1.9377) mem 68106MB [2022-12-19 20:16:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][700/1519] eta 0:13:44 lr 0.000024 time 0.9196 (1.0062) model_time 0.9194 (1.0047) loss 0.9344 (0.8845) grad_norm 8.2705 (8.7185/1.9605) mem 68106MB [2022-12-19 20:16:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][710/1519] eta 0:13:34 lr 0.000024 time 0.9243 (1.0062) model_time 0.9242 (1.0048) loss 0.7548 (0.8834) grad_norm 9.1520 (8.6869/1.9646) mem 68106MB [2022-12-19 20:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][720/1519] eta 0:13:23 lr 0.000024 time 0.9308 (1.0062) model_time 0.9306 (1.0048) loss 0.9755 (0.8835) grad_norm 7.8754 (8.6718/1.9664) mem 68106MB [2022-12-19 20:16:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][730/1519] eta 0:13:13 lr 0.000024 time 0.9325 (1.0062) model_time 0.9323 (1.0048) loss 0.9793 (0.8845) grad_norm 6.3990 (8.6413/1.9649) mem 68106MB [2022-12-19 20:17:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][740/1519] eta 0:13:03 lr 0.000024 time 1.0460 (1.0063) model_time 1.0458 (1.0049) loss 1.0183 (0.8848) grad_norm 7.7777 (8.6522/1.9929) mem 68106MB [2022-12-19 20:17:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][750/1519] eta 0:12:53 lr 0.000024 time 0.9231 (1.0064) model_time 0.9230 (1.0050) loss 1.2048 (0.8859) grad_norm 14.2210 (8.6898/2.0574) mem 68106MB [2022-12-19 20:17:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][760/1519] eta 0:12:43 lr 0.000024 time 0.9318 (1.0064) model_time 0.9316 (1.0051) loss 0.9478 (0.8858) grad_norm 6.4632 (8.6764/2.0554) mem 68106MB [2022-12-19 20:17:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][770/1519] eta 0:12:33 lr 0.000024 time 0.9371 (1.0064) model_time 0.9369 (1.0050) loss 0.6806 (0.8860) grad_norm 10.1794 (8.6976/2.0783) mem 68106MB [2022-12-19 20:17:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][780/1519] eta 0:12:23 lr 0.000024 time 0.9354 (1.0063) model_time 0.9352 (1.0050) loss 0.8220 (0.8850) grad_norm 12.5368 (8.6929/2.0921) mem 68106MB [2022-12-19 20:17:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][790/1519] eta 0:12:13 lr 0.000024 time 0.9387 (1.0062) model_time 0.9384 (1.0049) loss 0.9566 (0.8848) grad_norm 9.0395 (8.6968/2.0823) mem 68106MB [2022-12-19 20:18:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][800/1519] eta 0:12:03 lr 0.000024 time 0.9314 (1.0062) model_time 0.9312 (1.0049) loss 0.7552 (0.8842) grad_norm 8.8926 (8.7383/2.1153) mem 68106MB [2022-12-19 20:18:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][810/1519] eta 0:11:53 lr 0.000024 time 0.9346 (1.0062) model_time 0.9344 (1.0049) loss 0.7807 (0.8842) grad_norm 9.5207 (8.7687/2.1357) mem 68106MB [2022-12-19 20:18:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][820/1519] eta 0:11:43 lr 0.000024 time 0.9304 (1.0062) model_time 0.9302 (1.0049) loss 0.9557 (0.8848) grad_norm 9.9969 (8.7833/2.1313) mem 68106MB [2022-12-19 20:18:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][830/1519] eta 0:11:33 lr 0.000024 time 0.9245 (1.0062) model_time 0.9244 (1.0049) loss 0.8121 (0.8854) grad_norm 10.0519 (8.7965/2.1471) mem 68106MB [2022-12-19 20:18:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][840/1519] eta 0:11:23 lr 0.000024 time 1.1428 (1.0064) model_time 1.1426 (1.0051) loss 0.9990 (0.8852) grad_norm 7.8350 (8.7792/2.1626) mem 68106MB [2022-12-19 20:18:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][850/1519] eta 0:11:13 lr 0.000024 time 0.9400 (1.0063) model_time 0.9399 (1.0050) loss 1.1491 (0.8853) grad_norm 7.3488 (8.7471/2.1691) mem 68106MB [2022-12-19 20:19:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][860/1519] eta 0:11:03 lr 0.000024 time 0.9395 (1.0062) model_time 0.9393 (1.0050) loss 0.7084 (0.8844) grad_norm 6.6039 (8.7541/2.1681) mem 68106MB [2022-12-19 20:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][870/1519] eta 0:10:53 lr 0.000024 time 1.0082 (1.0062) model_time 1.0081 (1.0050) loss 1.2011 (0.8855) grad_norm 9.3986 (8.7491/2.1603) mem 68106MB [2022-12-19 20:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][880/1519] eta 0:10:42 lr 0.000024 time 0.9272 (1.0061) model_time 0.9269 (1.0049) loss 0.8253 (0.8855) grad_norm 7.7082 (8.7467/2.1804) mem 68106MB [2022-12-19 20:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][890/1519] eta 0:10:33 lr 0.000024 time 0.9238 (1.0065) model_time 0.9236 (1.0052) loss 0.7582 (0.8856) grad_norm 14.1600 (8.7653/2.1934) mem 68106MB [2022-12-19 20:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][900/1519] eta 0:10:23 lr 0.000024 time 0.9274 (1.0065) model_time 0.9272 (1.0053) loss 0.9087 (0.8847) grad_norm 10.9710 (8.7420/2.1877) mem 68106MB [2022-12-19 20:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][910/1519] eta 0:10:13 lr 0.000024 time 0.9168 (1.0066) model_time 0.9166 (1.0054) loss 0.6724 (0.8834) grad_norm 11.2927 (8.7328/2.1457) mem 68106MB [2022-12-19 20:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][920/1519] eta 0:10:02 lr 0.000024 time 0.9767 (1.0066) model_time 0.9766 (1.0054) loss 0.7332 (0.8826) grad_norm 5.4912 (8.7312/2.1674) mem 68106MB [2022-12-19 20:20:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][930/1519] eta 0:09:52 lr 0.000024 time 0.9236 (1.0066) model_time 0.9234 (1.0054) loss 0.7497 (0.8829) grad_norm 8.6001 (8.6865/2.0917) mem 68106MB [2022-12-19 20:20:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][940/1519] eta 0:09:42 lr 0.000024 time 0.9335 (1.0065) model_time 0.9334 (1.0053) loss 0.7481 (0.8839) grad_norm 10.9074 (8.7018/2.0976) mem 68106MB [2022-12-19 20:20:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][950/1519] eta 0:09:32 lr 0.000024 time 0.9240 (1.0064) model_time 0.9238 (1.0053) loss 0.7899 (0.8835) grad_norm 7.5463 (8.7015/2.1175) mem 68106MB [2022-12-19 20:20:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][960/1519] eta 0:09:22 lr 0.000024 time 0.9323 (1.0064) model_time 0.9321 (1.0052) loss 0.8971 (0.8847) grad_norm 12.5816 (8.7134/2.1039) mem 68106MB [2022-12-19 20:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][970/1519] eta 0:09:12 lr 0.000024 time 0.9238 (1.0063) model_time 0.9237 (1.0051) loss 0.6911 (0.8847) grad_norm 6.6876 (8.7270/2.1132) mem 68106MB [2022-12-19 20:21:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][980/1519] eta 0:09:02 lr 0.000024 time 0.9384 (1.0063) model_time 0.9383 (1.0051) loss 0.7277 (0.8845) grad_norm 7.5042 (8.7218/2.1226) mem 68106MB [2022-12-19 20:21:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][990/1519] eta 0:08:52 lr 0.000024 time 0.9278 (1.0062) model_time 0.9276 (1.0051) loss 0.6977 (0.8840) grad_norm 8.1098 (8.7226/2.1077) mem 68106MB [2022-12-19 20:21:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1000/1519] eta 0:08:42 lr 0.000024 time 0.9283 (1.0061) model_time 0.9281 (1.0050) loss 1.0886 (0.8839) grad_norm 7.7463 (8.7169/2.1077) mem 68106MB [2022-12-19 20:21:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1010/1519] eta 0:08:32 lr 0.000024 time 0.9870 (1.0062) model_time 0.9868 (1.0051) loss 0.6990 (0.8836) grad_norm 5.7009 (8.6829/2.0762) mem 68106MB [2022-12-19 20:21:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1020/1519] eta 0:08:22 lr 0.000024 time 0.9245 (1.0062) model_time 0.9244 (1.0051) loss 0.9218 (0.8839) grad_norm 8.7248 (8.6740/2.0574) mem 68106MB [2022-12-19 20:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1030/1519] eta 0:08:11 lr 0.000024 time 0.9581 (1.0061) model_time 0.9579 (1.0050) loss 0.8828 (0.8847) grad_norm 8.6615 (8.6468/2.0384) mem 68106MB [2022-12-19 20:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1040/1519] eta 0:08:01 lr 0.000024 time 0.9254 (1.0061) model_time 0.9252 (1.0050) loss 0.7282 (0.8842) grad_norm 10.5138 (8.6650/2.0551) mem 68106MB [2022-12-19 20:22:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1050/1519] eta 0:07:51 lr 0.000024 time 0.9196 (1.0060) model_time 0.9194 (1.0049) loss 0.7202 (0.8839) grad_norm 8.6704 (8.6348/2.0484) mem 68106MB [2022-12-19 20:22:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1060/1519] eta 0:07:41 lr 0.000024 time 0.9341 (1.0060) model_time 0.9340 (1.0050) loss 0.7787 (0.8836) grad_norm 8.8349 (8.6467/2.0386) mem 68106MB [2022-12-19 20:22:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1070/1519] eta 0:07:31 lr 0.000024 time 0.9280 (1.0062) model_time 0.9279 (1.0051) loss 0.8711 (0.8836) grad_norm 8.4053 (8.6899/2.0543) mem 68106MB [2022-12-19 20:22:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1080/1519] eta 0:07:21 lr 0.000024 time 0.9257 (1.0062) model_time 0.9256 (1.0051) loss 0.7352 (0.8833) grad_norm 12.9698 (8.7481/2.0738) mem 68106MB [2022-12-19 20:23:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1090/1519] eta 0:07:11 lr 0.000024 time 0.9203 (1.0061) model_time 0.9202 (1.0051) loss 1.0948 (0.8833) grad_norm 11.4294 (8.7541/2.0745) mem 68106MB [2022-12-19 20:23:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1100/1519] eta 0:07:01 lr 0.000024 time 0.9311 (1.0061) model_time 0.9310 (1.0050) loss 1.1084 (0.8842) grad_norm 7.3193 (8.7440/2.0824) mem 68106MB [2022-12-19 20:23:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1110/1519] eta 0:06:51 lr 0.000024 time 0.9222 (1.0061) model_time 0.9221 (1.0050) loss 0.9652 (0.8845) grad_norm 12.6325 (8.7685/2.1161) mem 68106MB [2022-12-19 20:23:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1120/1519] eta 0:06:41 lr 0.000024 time 0.9518 (1.0061) model_time 0.9517 (1.0050) loss 0.6991 (0.8843) grad_norm 10.1649 (8.7641/2.1104) mem 68106MB [2022-12-19 20:23:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1130/1519] eta 0:06:31 lr 0.000024 time 0.9334 (1.0060) model_time 0.9332 (1.0050) loss 0.7602 (0.8847) grad_norm 8.1513 (8.7637/2.1184) mem 68106MB [2022-12-19 20:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1140/1519] eta 0:06:21 lr 0.000024 time 0.9213 (1.0060) model_time 0.9212 (1.0050) loss 0.7479 (0.8845) grad_norm 12.0234 (8.8054/2.1224) mem 68106MB [2022-12-19 20:24:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1150/1519] eta 0:06:11 lr 0.000024 time 0.9264 (1.0059) model_time 0.9263 (1.0049) loss 0.8427 (0.8846) grad_norm 6.8232 (8.8092/2.1312) mem 68106MB [2022-12-19 20:24:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1160/1519] eta 0:06:01 lr 0.000024 time 0.9274 (1.0059) model_time 0.9273 (1.0049) loss 0.6972 (0.8846) grad_norm 9.3890 (8.8271/2.1383) mem 68106MB [2022-12-19 20:24:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1170/1519] eta 0:05:51 lr 0.000024 time 0.9255 (1.0059) model_time 0.9254 (1.0049) loss 0.9422 (0.8848) grad_norm 7.9032 (8.8256/2.1410) mem 68106MB [2022-12-19 20:24:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1180/1519] eta 0:05:40 lr 0.000024 time 0.9300 (1.0059) model_time 0.9299 (1.0049) loss 0.9078 (0.8849) grad_norm 7.1179 (8.8095/2.0817) mem 68106MB [2022-12-19 20:24:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1190/1519] eta 0:05:30 lr 0.000024 time 0.9780 (1.0060) model_time 0.9779 (1.0050) loss 0.8086 (0.8843) grad_norm 7.9508 (8.8104/2.0749) mem 68106MB [2022-12-19 20:24:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1200/1519] eta 0:05:20 lr 0.000024 time 0.9924 (1.0060) model_time 0.9923 (1.0050) loss 0.6988 (0.8842) grad_norm 9.3745 (8.8153/2.0779) mem 68106MB [2022-12-19 20:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1210/1519] eta 0:05:10 lr 0.000024 time 0.9365 (1.0060) model_time 0.9363 (1.0050) loss 0.9543 (0.8848) grad_norm 8.9594 (8.7832/2.0947) mem 68106MB [2022-12-19 20:25:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1220/1519] eta 0:05:00 lr 0.000024 time 0.9240 (1.0060) model_time 0.9239 (1.0050) loss 0.7584 (0.8839) grad_norm 10.7576 (8.7956/2.1010) mem 68106MB [2022-12-19 20:25:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1230/1519] eta 0:04:50 lr 0.000024 time 0.9252 (1.0060) model_time 0.9250 (1.0050) loss 0.8287 (0.8833) grad_norm 13.9338 (8.8421/2.1248) mem 68106MB [2022-12-19 20:25:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1240/1519] eta 0:04:40 lr 0.000024 time 0.9278 (1.0060) model_time 0.9276 (1.0051) loss 0.7813 (0.8829) grad_norm 9.0523 (8.8513/2.1131) mem 68106MB [2022-12-19 20:25:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1250/1519] eta 0:04:30 lr 0.000024 time 0.9305 (1.0060) model_time 0.9304 (1.0051) loss 0.8311 (0.8831) grad_norm 8.4805 (8.8471/2.1089) mem 68106MB [2022-12-19 20:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1260/1519] eta 0:04:20 lr 0.000024 time 0.9196 (1.0059) model_time 0.9194 (1.0050) loss 0.9385 (0.8833) grad_norm 10.1094 (8.7864/2.0206) mem 68106MB [2022-12-19 20:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1270/1519] eta 0:04:10 lr 0.000024 time 0.9364 (1.0059) model_time 0.9363 (1.0049) loss 0.9907 (0.8828) grad_norm 7.3642 (8.7913/2.0122) mem 68106MB [2022-12-19 20:26:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1280/1519] eta 0:04:00 lr 0.000024 time 0.9278 (1.0059) model_time 0.9277 (1.0049) loss 0.9452 (0.8825) grad_norm 9.1622 (8.7637/1.9968) mem 68106MB [2022-12-19 20:26:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1290/1519] eta 0:03:50 lr 0.000024 time 0.9024 (1.0059) model_time 0.9021 (1.0050) loss 0.7443 (0.8822) grad_norm 6.4199 (8.7525/1.9917) mem 68106MB [2022-12-19 20:26:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1300/1519] eta 0:03:40 lr 0.000024 time 0.9325 (1.0059) model_time 0.9324 (1.0049) loss 0.7949 (0.8818) grad_norm 8.2243 (8.7258/1.9679) mem 68106MB [2022-12-19 20:26:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1310/1519] eta 0:03:30 lr 0.000024 time 0.9649 (1.0059) model_time 0.9647 (1.0049) loss 1.1053 (0.8815) grad_norm 8.7046 (8.7531/1.9601) mem 68106MB [2022-12-19 20:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1320/1519] eta 0:03:20 lr 0.000024 time 0.9310 (1.0058) model_time 0.9308 (1.0049) loss 0.6928 (0.8812) grad_norm 7.6548 (8.7607/1.9600) mem 68106MB [2022-12-19 20:27:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1330/1519] eta 0:03:10 lr 0.000024 time 1.0139 (1.0058) model_time 1.0137 (1.0049) loss 0.7915 (0.8810) grad_norm 9.0835 (8.7858/1.9588) mem 68106MB [2022-12-19 20:27:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1340/1519] eta 0:03:00 lr 0.000024 time 0.9279 (1.0058) model_time 0.9278 (1.0049) loss 0.8977 (0.8812) grad_norm 10.8778 (8.7790/1.9349) mem 68106MB [2022-12-19 20:27:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1350/1519] eta 0:02:49 lr 0.000024 time 0.9247 (1.0058) model_time 0.9245 (1.0049) loss 0.7025 (0.8817) grad_norm 12.5500 (8.7559/1.8998) mem 68106MB [2022-12-19 20:27:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1360/1519] eta 0:02:39 lr 0.000024 time 0.9315 (1.0058) model_time 0.9313 (1.0049) loss 0.6902 (0.8814) grad_norm 9.8229 (8.7714/1.9050) mem 68106MB [2022-12-19 20:27:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1370/1519] eta 0:02:29 lr 0.000024 time 0.9287 (1.0057) model_time 0.9286 (1.0048) loss 0.7071 (0.8814) grad_norm 8.0215 (8.7346/1.8830) mem 68106MB [2022-12-19 20:27:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1380/1519] eta 0:02:19 lr 0.000024 time 1.0070 (1.0058) model_time 1.0068 (1.0049) loss 0.7806 (0.8811) grad_norm 8.9661 (8.7268/1.8645) mem 68106MB [2022-12-19 20:28:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1390/1519] eta 0:02:09 lr 0.000024 time 0.9305 (1.0058) model_time 0.9303 (1.0049) loss 1.0957 (0.8806) grad_norm 8.9782 (8.7244/1.8628) mem 68106MB [2022-12-19 20:28:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1400/1519] eta 0:01:59 lr 0.000024 time 0.9331 (1.0058) model_time 0.9330 (1.0049) loss 0.7552 (0.8804) grad_norm 7.5162 (8.6911/1.8296) mem 68106MB [2022-12-19 20:28:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1410/1519] eta 0:01:49 lr 0.000024 time 0.9261 (1.0058) model_time 0.9259 (1.0049) loss 0.7995 (0.8800) grad_norm 7.4831 (8.6565/1.8064) mem 68106MB [2022-12-19 20:28:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1420/1519] eta 0:01:39 lr 0.000024 time 0.9311 (1.0058) model_time 0.9309 (1.0049) loss 0.7007 (0.8800) grad_norm 5.9914 (8.6577/1.8101) mem 68106MB [2022-12-19 20:28:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1430/1519] eta 0:01:29 lr 0.000024 time 0.9326 (1.0057) model_time 0.9325 (1.0048) loss 0.6813 (0.8795) grad_norm 9.6440 (8.6370/1.7957) mem 68106MB [2022-12-19 20:28:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1440/1519] eta 0:01:19 lr 0.000024 time 0.9469 (1.0057) model_time 0.9468 (1.0049) loss 0.7985 (0.8792) grad_norm 8.5379 (8.6465/1.7810) mem 68106MB [2022-12-19 20:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1450/1519] eta 0:01:09 lr 0.000024 time 0.9227 (1.0057) model_time 0.9226 (1.0048) loss 0.8954 (0.8788) grad_norm 8.1965 (8.6594/1.7710) mem 68106MB [2022-12-19 20:29:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1460/1519] eta 0:00:59 lr 0.000024 time 0.9199 (1.0057) model_time 0.9198 (1.0048) loss 0.9117 (0.8795) grad_norm 10.3646 (8.6878/1.7975) mem 68106MB [2022-12-19 20:29:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1470/1519] eta 0:00:49 lr 0.000024 time 0.9241 (1.0058) model_time 0.9239 (1.0049) loss 0.6998 (0.8790) grad_norm 9.2467 (8.6922/1.7939) mem 68106MB [2022-12-19 20:29:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1480/1519] eta 0:00:39 lr 0.000024 time 0.9212 (1.0057) model_time 0.9210 (1.0048) loss 0.9712 (0.8788) grad_norm 12.2679 (8.6942/1.7931) mem 68106MB [2022-12-19 20:29:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1490/1519] eta 0:00:29 lr 0.000024 time 0.9786 (1.0057) model_time 0.9784 (1.0048) loss 0.7083 (0.8787) grad_norm 6.1852 (8.6486/1.7812) mem 68106MB [2022-12-19 20:29:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1500/1519] eta 0:00:19 lr 0.000024 time 0.9288 (1.0057) model_time 0.9286 (1.0048) loss 0.7955 (0.8791) grad_norm 10.3275 (8.6667/1.7818) mem 68106MB [2022-12-19 20:30:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [36/100][1510/1519] eta 0:00:09 lr 0.000024 time 0.9932 (1.0057) model_time 0.9931 (1.0048) loss 0.9654 (0.8792) grad_norm 13.1111 (8.6720/1.8011) mem 68106MB [2022-12-19 20:30:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 36 training takes 0:25:27 [2022-12-19 20:30:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_36.pth saving...... [2022-12-19 20:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_36.pth saved !!! [2022-12-19 20:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.692 (0.692) Loss 0.5101 (0.5101) Acc@1 90.278 (90.278) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 20:30:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.300 (0.334) Loss 0.5095 (0.4902) Acc@1 92.708 (91.793) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-19 20:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.316) Loss 0.4317 (0.4904) Acc@1 93.056 (91.849) Acc@5 98.958 (98.313) Mem 68106MB [2022-12-19 20:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.5927 (0.4943) Acc@1 89.236 (91.577) Acc@5 98.264 (98.275) Mem 68106MB [2022-12-19 20:30:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.308) Loss 0.4597 (0.4877) Acc@1 92.014 (91.701) Acc@5 99.306 (98.382) Mem 68106MB [2022-12-19 20:30:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.306 (0.306) Loss 0.4852 (0.4844) Acc@1 91.319 (91.789) Acc@5 98.958 (98.427) Mem 68106MB [2022-12-19 20:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.305) Loss 0.4905 (0.4843) Acc@1 89.236 (91.763) Acc@5 97.917 (98.418) Mem 68106MB [2022-12-19 20:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5182 (0.4864) Acc@1 91.667 (91.657) Acc@5 98.611 (98.396) Mem 68106MB [2022-12-19 20:30:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.298 (0.303) Loss 0.4015 (0.4845) Acc@1 93.403 (91.675) Acc@5 98.611 (98.431) Mem 68106MB [2022-12-19 20:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:36] * Acc@1 91.663 Acc@5 98.428 [2022-12-19 20:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.7% [2022-12-19 20:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 20:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 20:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.66% [2022-12-19 20:31:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][0/1519] eta 0:36:14 lr 0.000024 time 1.4315 (1.4315) model_time 0.9925 (0.9925) loss 1.0651 (1.0651) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 20:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][10/1519] eta 0:26:30 lr 0.000024 time 0.9264 (1.0540) model_time 0.9263 (1.0137) loss 0.6897 (0.8313) grad_norm 8.3240 (8.2746/1.4032) mem 68106MB [2022-12-19 20:31:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][20/1519] eta 0:25:38 lr 0.000024 time 0.9178 (1.0266) model_time 0.9177 (1.0054) loss 1.0978 (0.8524) grad_norm 10.3071 (8.7414/1.3302) mem 68106MB [2022-12-19 20:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][30/1519] eta 0:25:19 lr 0.000024 time 0.9336 (1.0203) model_time 0.9335 (1.0058) loss 0.8614 (0.8506) grad_norm 7.7290 (8.4288/1.2988) mem 68106MB [2022-12-19 20:32:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][40/1519] eta 0:25:04 lr 0.000024 time 0.9215 (1.0174) model_time 0.9213 (1.0064) loss 0.7367 (0.8520) grad_norm 7.2602 (8.1048/1.2883) mem 68106MB [2022-12-19 20:32:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][50/1519] eta 0:24:48 lr 0.000024 time 0.9230 (1.0136) model_time 0.9225 (1.0046) loss 0.6724 (0.8659) grad_norm 8.9317 (7.9908/1.2833) mem 68106MB [2022-12-19 20:32:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][60/1519] eta 0:24:35 lr 0.000024 time 0.9260 (1.0110) model_time 0.9259 (1.0034) loss 0.6993 (0.8719) grad_norm 7.1787 (8.0015/1.2768) mem 68106MB [2022-12-19 20:32:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][70/1519] eta 0:24:22 lr 0.000024 time 0.9194 (1.0094) model_time 0.9192 (1.0028) loss 1.1653 (0.8739) grad_norm 10.0603 (8.1209/1.2658) mem 68106MB [2022-12-19 20:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][80/1519] eta 0:24:12 lr 0.000024 time 0.9648 (1.0097) model_time 0.9646 (1.0039) loss 0.8272 (0.8714) grad_norm 7.5208 (8.1388/1.2350) mem 68106MB [2022-12-19 20:32:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][90/1519] eta 0:24:01 lr 0.000024 time 0.9292 (1.0086) model_time 0.9290 (1.0034) loss 0.7959 (0.8735) grad_norm 8.8761 (8.1512/1.2868) mem 68106MB [2022-12-19 20:33:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][100/1519] eta 0:23:49 lr 0.000024 time 0.9309 (1.0076) model_time 0.9307 (1.0029) loss 0.8910 (0.8772) grad_norm 8.9479 (8.2025/1.3304) mem 68106MB [2022-12-19 20:33:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][110/1519] eta 0:23:38 lr 0.000024 time 0.9330 (1.0067) model_time 0.9325 (1.0024) loss 0.9300 (0.8801) grad_norm 9.1597 (8.4477/1.6577) mem 68106MB [2022-12-19 20:33:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][120/1519] eta 0:23:27 lr 0.000024 time 0.9226 (1.0061) model_time 0.9224 (1.0021) loss 0.9109 (0.8864) grad_norm 6.6397 (8.4904/1.7169) mem 68106MB [2022-12-19 20:33:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][130/1519] eta 0:23:16 lr 0.000024 time 0.9197 (1.0054) model_time 0.9195 (1.0017) loss 0.8610 (0.8802) grad_norm 9.4460 (8.5197/1.6706) mem 68106MB [2022-12-19 20:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][140/1519] eta 0:23:06 lr 0.000024 time 0.9333 (1.0053) model_time 0.9330 (1.0018) loss 0.9867 (0.8785) grad_norm 6.5556 (8.6209/1.8642) mem 68106MB [2022-12-19 20:33:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][150/1519] eta 0:22:55 lr 0.000024 time 0.9321 (1.0050) model_time 0.9319 (1.0017) loss 0.9240 (0.8836) grad_norm 10.4208 (8.5865/1.8457) mem 68106MB [2022-12-19 20:34:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][160/1519] eta 0:22:45 lr 0.000024 time 0.9129 (1.0050) model_time 0.9128 (1.0019) loss 0.7661 (0.8797) grad_norm 9.3617 (8.5769/1.8217) mem 68106MB [2022-12-19 20:34:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][170/1519] eta 0:22:36 lr 0.000024 time 0.9742 (1.0053) model_time 0.9740 (1.0024) loss 1.3345 (0.8844) grad_norm 6.9112 (8.5271/1.8091) mem 68106MB [2022-12-19 20:34:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][180/1519] eta 0:22:26 lr 0.000024 time 0.9322 (1.0055) model_time 0.9320 (1.0027) loss 0.6898 (0.8840) grad_norm 8.1599 (8.5120/1.7876) mem 68106MB [2022-12-19 20:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][190/1519] eta 0:22:16 lr 0.000024 time 0.9368 (1.0058) model_time 0.9367 (1.0031) loss 0.8349 (0.8852) grad_norm 8.4768 (8.5073/1.7629) mem 68106MB [2022-12-19 20:34:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][200/1519] eta 0:22:06 lr 0.000024 time 0.9311 (1.0054) model_time 0.9310 (1.0028) loss 0.9563 (0.8827) grad_norm 7.4212 (8.5003/1.7420) mem 68106MB [2022-12-19 20:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][210/1519] eta 0:21:55 lr 0.000024 time 0.9288 (1.0053) model_time 0.9286 (1.0028) loss 0.8558 (0.8801) grad_norm 7.5837 (8.5156/1.7385) mem 68106MB [2022-12-19 20:35:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][220/1519] eta 0:21:47 lr 0.000024 time 0.9261 (1.0063) model_time 0.9259 (1.0039) loss 1.0391 (0.8806) grad_norm 6.4710 (8.4929/1.7253) mem 68106MB [2022-12-19 20:35:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][230/1519] eta 0:21:36 lr 0.000024 time 0.9263 (1.0059) model_time 0.9262 (1.0036) loss 1.1608 (0.8823) grad_norm 8.8073 (8.5269/1.7182) mem 68106MB [2022-12-19 20:35:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][240/1519] eta 0:21:26 lr 0.000024 time 0.9405 (1.0057) model_time 0.9403 (1.0035) loss 1.0945 (0.8822) grad_norm 7.7555 (8.5709/1.7875) mem 68106MB [2022-12-19 20:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][250/1519] eta 0:21:15 lr 0.000024 time 0.9360 (1.0055) model_time 0.9359 (1.0033) loss 0.6902 (0.8803) grad_norm 7.3033 (8.5620/1.7947) mem 68106MB [2022-12-19 20:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][260/1519] eta 0:21:05 lr 0.000024 time 0.9270 (1.0052) model_time 0.9269 (1.0032) loss 1.0791 (0.8797) grad_norm 9.2037 (8.5645/1.7700) mem 68106MB [2022-12-19 20:35:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][270/1519] eta 0:20:55 lr 0.000024 time 0.9303 (1.0054) model_time 0.9302 (1.0034) loss 1.0831 (0.8795) grad_norm 6.3174 (8.5214/1.7694) mem 68106MB [2022-12-19 20:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][280/1519] eta 0:20:45 lr 0.000024 time 0.9294 (1.0052) model_time 0.9293 (1.0033) loss 1.0655 (0.8849) grad_norm 7.8232 (8.4763/1.7563) mem 68106MB [2022-12-19 20:36:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][290/1519] eta 0:20:35 lr 0.000024 time 0.9312 (1.0053) model_time 0.9309 (1.0034) loss 1.0654 (0.8866) grad_norm 11.4364 (8.5174/1.7802) mem 68106MB [2022-12-19 20:36:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][300/1519] eta 0:20:25 lr 0.000024 time 0.9332 (1.0052) model_time 0.9330 (1.0033) loss 0.7458 (0.8873) grad_norm 8.7169 (8.5084/1.7716) mem 68106MB [2022-12-19 20:36:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][310/1519] eta 0:20:16 lr 0.000024 time 0.9132 (1.0066) model_time 0.9130 (1.0048) loss 1.0037 (0.8863) grad_norm 8.8148 (8.5284/1.7743) mem 68106MB [2022-12-19 20:36:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][320/1519] eta 0:20:06 lr 0.000024 time 0.9274 (1.0064) model_time 0.9273 (1.0046) loss 0.7647 (0.8868) grad_norm 12.4142 (8.5327/1.7925) mem 68106MB [2022-12-19 20:36:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][330/1519] eta 0:19:56 lr 0.000024 time 0.9423 (1.0063) model_time 0.9422 (1.0046) loss 1.1474 (0.8865) grad_norm 11.6004 (8.5317/1.7923) mem 68106MB [2022-12-19 20:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][340/1519] eta 0:19:46 lr 0.000024 time 0.9295 (1.0064) model_time 0.9294 (1.0047) loss 0.6992 (0.8850) grad_norm 7.2206 (8.5286/1.7804) mem 68106MB [2022-12-19 20:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][350/1519] eta 0:19:36 lr 0.000024 time 0.9342 (1.0061) model_time 0.9341 (1.0045) loss 0.9662 (0.8867) grad_norm 7.7262 (8.5169/1.7638) mem 68106MB [2022-12-19 20:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][360/1519] eta 0:19:26 lr 0.000024 time 0.9310 (1.0061) model_time 0.9309 (1.0045) loss 1.0683 (0.8849) grad_norm 7.7591 (8.5328/1.7996) mem 68106MB [2022-12-19 20:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][370/1519] eta 0:19:16 lr 0.000024 time 0.9319 (1.0061) model_time 0.9318 (1.0046) loss 0.8278 (0.8843) grad_norm 9.3623 (8.5778/1.8201) mem 68106MB [2022-12-19 20:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][380/1519] eta 0:19:05 lr 0.000024 time 0.9246 (1.0059) model_time 0.9244 (1.0043) loss 0.6867 (0.8831) grad_norm 7.2165 (8.5892/1.8214) mem 68106MB [2022-12-19 20:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][390/1519] eta 0:18:55 lr 0.000024 time 0.9279 (1.0057) model_time 0.9277 (1.0042) loss 0.7837 (0.8805) grad_norm 9.0748 (8.6011/1.8036) mem 68106MB [2022-12-19 20:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][400/1519] eta 0:18:45 lr 0.000024 time 0.9388 (1.0057) model_time 0.9387 (1.0043) loss 0.8507 (0.8802) grad_norm 6.4682 (8.6203/1.8131) mem 68106MB [2022-12-19 20:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][410/1519] eta 0:18:35 lr 0.000024 time 0.9665 (1.0057) model_time 0.9664 (1.0042) loss 1.0180 (0.8791) grad_norm 6.8318 (8.6134/1.8071) mem 68106MB [2022-12-19 20:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][420/1519] eta 0:18:25 lr 0.000024 time 0.9248 (1.0056) model_time 0.9246 (1.0042) loss 0.8666 (0.8779) grad_norm 9.8601 (8.6211/1.7925) mem 68106MB [2022-12-19 20:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][430/1519] eta 0:18:14 lr 0.000024 time 0.9257 (1.0055) model_time 0.9255 (1.0041) loss 0.7400 (0.8764) grad_norm 9.1999 (8.6460/1.8125) mem 68106MB [2022-12-19 20:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][440/1519] eta 0:18:04 lr 0.000024 time 0.9257 (1.0053) model_time 0.9256 (1.0040) loss 1.1200 (0.8766) grad_norm 7.3430 (8.6284/1.8027) mem 68106MB [2022-12-19 20:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][450/1519] eta 0:17:54 lr 0.000024 time 0.9240 (1.0053) model_time 0.9238 (1.0040) loss 0.9577 (0.8784) grad_norm 9.5286 (8.6132/1.7946) mem 68106MB [2022-12-19 20:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][460/1519] eta 0:17:44 lr 0.000024 time 0.9729 (1.0053) model_time 0.9728 (1.0040) loss 0.7752 (0.8786) grad_norm 11.2398 (8.6132/1.8011) mem 68106MB [2022-12-19 20:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][470/1519] eta 0:17:34 lr 0.000024 time 0.9175 (1.0051) model_time 0.9173 (1.0038) loss 0.9994 (0.8778) grad_norm 10.0175 (8.6088/1.7917) mem 68106MB [2022-12-19 20:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][480/1519] eta 0:17:24 lr 0.000024 time 0.9328 (1.0052) model_time 0.9326 (1.0039) loss 0.9232 (0.8779) grad_norm 10.2151 (8.6460/1.7988) mem 68106MB [2022-12-19 20:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][490/1519] eta 0:17:14 lr 0.000024 time 0.9236 (1.0054) model_time 0.9235 (1.0042) loss 0.9612 (0.8781) grad_norm 6.0214 (8.6527/1.8067) mem 68106MB [2022-12-19 20:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][500/1519] eta 0:17:04 lr 0.000024 time 0.9241 (1.0053) model_time 0.9240 (1.0041) loss 1.0132 (0.8789) grad_norm 7.0278 (8.6281/1.8030) mem 68106MB [2022-12-19 20:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][510/1519] eta 0:16:54 lr 0.000024 time 0.9277 (1.0055) model_time 0.9275 (1.0043) loss 0.7880 (0.8770) grad_norm 7.3782 (8.6311/1.7941) mem 68106MB [2022-12-19 20:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][520/1519] eta 0:16:44 lr 0.000024 time 0.9064 (1.0057) model_time 0.9063 (1.0045) loss 1.0716 (0.8772) grad_norm 10.2037 (8.6408/1.7933) mem 68106MB [2022-12-19 20:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][530/1519] eta 0:16:34 lr 0.000024 time 0.9304 (1.0058) model_time 0.9303 (1.0046) loss 0.7069 (0.8771) grad_norm 9.1345 (8.6397/1.7860) mem 68106MB [2022-12-19 20:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][540/1519] eta 0:16:24 lr 0.000024 time 0.9258 (1.0057) model_time 0.9257 (1.0046) loss 1.0319 (0.8773) grad_norm 12.2836 (8.6428/1.8006) mem 68106MB [2022-12-19 20:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][550/1519] eta 0:16:14 lr 0.000024 time 0.9207 (1.0056) model_time 0.9206 (1.0045) loss 1.0356 (0.8773) grad_norm 8.0366 (8.6824/1.8220) mem 68106MB [2022-12-19 20:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][560/1519] eta 0:16:04 lr 0.000024 time 0.9285 (1.0055) model_time 0.9283 (1.0044) loss 0.7340 (0.8757) grad_norm 7.4435 (8.6962/1.8174) mem 68106MB [2022-12-19 20:40:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][570/1519] eta 0:15:54 lr 0.000023 time 0.9286 (1.0054) model_time 0.9285 (1.0043) loss 0.8346 (0.8756) grad_norm 7.9793 (8.7111/1.8098) mem 68106MB [2022-12-19 20:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][580/1519] eta 0:15:44 lr 0.000023 time 0.9337 (1.0055) model_time 0.9322 (1.0044) loss 0.7710 (0.8745) grad_norm 7.2588 (8.7179/1.8068) mem 68106MB [2022-12-19 20:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][590/1519] eta 0:15:34 lr 0.000023 time 0.9692 (1.0055) model_time 0.9690 (1.0044) loss 1.0542 (0.8756) grad_norm 6.9607 (8.7452/1.8698) mem 68106MB [2022-12-19 20:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][600/1519] eta 0:15:24 lr 0.000023 time 0.9656 (1.0056) model_time 0.9655 (1.0045) loss 0.8366 (0.8768) grad_norm 10.1626 (8.7547/1.8935) mem 68106MB [2022-12-19 20:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][610/1519] eta 0:15:13 lr 0.000023 time 0.9227 (1.0054) model_time 0.9226 (1.0044) loss 0.6897 (0.8745) grad_norm 7.4718 (8.7583/1.8873) mem 68106MB [2022-12-19 20:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][620/1519] eta 0:15:03 lr 0.000023 time 0.9269 (1.0054) model_time 0.9268 (1.0043) loss 0.6885 (0.8739) grad_norm 8.4126 (8.7592/1.8845) mem 68106MB [2022-12-19 20:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][630/1519] eta 0:14:53 lr 0.000023 time 0.9261 (1.0055) model_time 0.9260 (1.0045) loss 0.8207 (0.8737) grad_norm 8.8140 (8.7790/1.8773) mem 68106MB [2022-12-19 20:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][640/1519] eta 0:14:43 lr 0.000023 time 0.9791 (1.0055) model_time 0.9789 (1.0045) loss 0.7143 (0.8741) grad_norm 8.7806 (8.8133/1.8744) mem 68106MB [2022-12-19 20:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][650/1519] eta 0:14:33 lr 0.000023 time 0.9258 (1.0054) model_time 0.9257 (1.0044) loss 0.9477 (0.8747) grad_norm 8.3071 (8.8245/1.8649) mem 68106MB [2022-12-19 20:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][660/1519] eta 0:14:23 lr 0.000023 time 0.9310 (1.0053) model_time 0.9309 (1.0043) loss 0.7691 (0.8737) grad_norm 9.8143 (8.8393/1.8683) mem 68106MB [2022-12-19 20:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][670/1519] eta 0:14:13 lr 0.000023 time 0.9466 (1.0055) model_time 0.9464 (1.0045) loss 0.9679 (0.8730) grad_norm 11.1178 (8.8318/1.8795) mem 68106MB [2022-12-19 20:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][680/1519] eta 0:14:03 lr 0.000023 time 0.9239 (1.0054) model_time 0.9238 (1.0044) loss 0.8631 (0.8722) grad_norm 8.7882 (8.8590/1.8872) mem 68106MB [2022-12-19 20:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][690/1519] eta 0:13:53 lr 0.000023 time 0.9230 (1.0054) model_time 0.9228 (1.0044) loss 1.1986 (0.8722) grad_norm 9.0308 (8.8594/1.8807) mem 68106MB [2022-12-19 20:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][700/1519] eta 0:13:43 lr 0.000023 time 0.9315 (1.0054) model_time 0.9313 (1.0044) loss 0.7772 (0.8722) grad_norm 6.3714 (8.8502/1.8865) mem 68106MB [2022-12-19 20:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][710/1519] eta 0:13:33 lr 0.000023 time 0.9271 (1.0053) model_time 0.9270 (1.0043) loss 1.0732 (0.8727) grad_norm 15.5020 (8.8497/1.9210) mem 68106MB [2022-12-19 20:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][720/1519] eta 0:13:23 lr 0.000023 time 0.9293 (1.0052) model_time 0.9292 (1.0042) loss 0.9083 (0.8731) grad_norm 11.0285 (8.8673/1.9183) mem 68106MB [2022-12-19 20:43:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][730/1519] eta 0:13:13 lr 0.000023 time 0.9266 (1.0051) model_time 0.9265 (1.0042) loss 0.7440 (0.8724) grad_norm 7.5121 (8.8912/1.9524) mem 68106MB [2022-12-19 20:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][740/1519] eta 0:13:02 lr 0.000023 time 0.9208 (1.0050) model_time 0.9206 (1.0041) loss 0.8643 (0.8725) grad_norm 6.8898 (8.8658/1.9067) mem 68106MB [2022-12-19 20:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][750/1519] eta 0:12:52 lr 0.000023 time 0.9167 (1.0050) model_time 0.9165 (1.0040) loss 0.8992 (0.8720) grad_norm 7.4411 (8.8827/1.9124) mem 68106MB [2022-12-19 20:44:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][760/1519] eta 0:12:42 lr 0.000023 time 0.9201 (1.0051) model_time 0.9200 (1.0041) loss 0.7126 (0.8729) grad_norm 7.4515 (8.8640/1.9138) mem 68106MB [2022-12-19 20:44:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][770/1519] eta 0:12:32 lr 0.000023 time 0.9292 (1.0050) model_time 0.9290 (1.0041) loss 0.6835 (0.8730) grad_norm 13.9032 (8.9014/1.9504) mem 68106MB [2022-12-19 20:44:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][780/1519] eta 0:12:22 lr 0.000023 time 0.9316 (1.0052) model_time 0.9314 (1.0043) loss 0.9056 (0.8725) grad_norm 10.0220 (8.9243/1.9492) mem 68106MB [2022-12-19 20:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][790/1519] eta 0:12:13 lr 0.000023 time 0.9089 (1.0056) model_time 0.9088 (1.0047) loss 0.8542 (0.8734) grad_norm 7.5315 (8.9295/1.9520) mem 68106MB [2022-12-19 20:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][800/1519] eta 0:12:03 lr 0.000023 time 0.9259 (1.0057) model_time 0.9258 (1.0048) loss 0.7140 (0.8726) grad_norm 9.1803 (8.9348/1.9525) mem 68106MB [2022-12-19 20:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][810/1519] eta 0:11:53 lr 0.000023 time 0.9324 (1.0057) model_time 0.9322 (1.0048) loss 0.7908 (0.8729) grad_norm 7.0472 (8.9217/1.9486) mem 68106MB [2022-12-19 20:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][820/1519] eta 0:11:43 lr 0.000023 time 0.9292 (1.0058) model_time 0.9290 (1.0049) loss 0.7567 (0.8723) grad_norm 6.6861 (8.9122/1.9528) mem 68106MB [2022-12-19 20:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][830/1519] eta 0:11:32 lr 0.000023 time 0.9212 (1.0057) model_time 0.9211 (1.0048) loss 0.7186 (0.8722) grad_norm 7.2477 (8.8770/1.9579) mem 68106MB [2022-12-19 20:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][840/1519] eta 0:11:23 lr 0.000023 time 1.1469 (1.0060) model_time 1.1468 (1.0052) loss 1.0941 (0.8724) grad_norm 9.7328 (8.8531/1.9318) mem 68106MB [2022-12-19 20:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][850/1519] eta 0:11:12 lr 0.000023 time 0.9294 (1.0060) model_time 0.9292 (1.0051) loss 0.8039 (0.8719) grad_norm 8.8164 (8.8653/1.9177) mem 68106MB [2022-12-19 20:45:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][860/1519] eta 0:11:02 lr 0.000023 time 0.9248 (1.0058) model_time 0.9246 (1.0050) loss 0.6917 (0.8731) grad_norm 10.1550 (8.8732/1.9178) mem 68106MB [2022-12-19 20:46:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][870/1519] eta 0:10:52 lr 0.000023 time 0.9263 (1.0058) model_time 0.9262 (1.0050) loss 0.8831 (0.8730) grad_norm 8.2171 (8.9000/1.9039) mem 68106MB [2022-12-19 20:46:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][880/1519] eta 0:10:42 lr 0.000023 time 0.9256 (1.0058) model_time 0.9255 (1.0049) loss 0.8419 (0.8726) grad_norm 9.4863 (8.9330/1.9045) mem 68106MB [2022-12-19 20:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][890/1519] eta 0:10:32 lr 0.000023 time 0.9317 (1.0057) model_time 0.9314 (1.0049) loss 0.8702 (0.8729) grad_norm 10.1023 (8.9061/1.8983) mem 68106MB [2022-12-19 20:46:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][900/1519] eta 0:10:22 lr 0.000023 time 0.9257 (1.0056) model_time 0.9255 (1.0048) loss 1.0210 (0.8734) grad_norm 11.0709 (8.9009/1.9019) mem 68106MB [2022-12-19 20:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][910/1519] eta 0:10:12 lr 0.000023 time 0.9317 (1.0056) model_time 0.9314 (1.0047) loss 0.8804 (0.8734) grad_norm 6.1917 (8.8610/1.9110) mem 68106MB [2022-12-19 20:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][920/1519] eta 0:10:02 lr 0.000023 time 0.9261 (1.0055) model_time 0.9260 (1.0047) loss 0.6919 (0.8731) grad_norm 14.3408 (8.8612/1.9254) mem 68106MB [2022-12-19 20:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][930/1519] eta 0:09:52 lr 0.000023 time 0.9334 (1.0054) model_time 0.9332 (1.0046) loss 0.9772 (0.8732) grad_norm 8.3139 (8.8648/1.9135) mem 68106MB [2022-12-19 20:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][940/1519] eta 0:09:42 lr 0.000023 time 0.9309 (1.0055) model_time 0.9307 (1.0047) loss 1.1729 (0.8740) grad_norm 10.2485 (8.8853/1.9167) mem 68106MB [2022-12-19 20:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][950/1519] eta 0:09:32 lr 0.000023 time 0.9280 (1.0054) model_time 0.9279 (1.0046) loss 0.8799 (0.8747) grad_norm 8.0718 (8.8773/1.9211) mem 68106MB [2022-12-19 20:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][960/1519] eta 0:09:22 lr 0.000023 time 0.9313 (1.0054) model_time 0.9312 (1.0046) loss 0.8225 (0.8738) grad_norm 10.6927 (8.8811/1.8993) mem 68106MB [2022-12-19 20:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][970/1519] eta 0:09:11 lr 0.000023 time 0.9322 (1.0054) model_time 0.9321 (1.0046) loss 0.7859 (0.8736) grad_norm 7.6950 (8.8594/1.8958) mem 68106MB [2022-12-19 20:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][980/1519] eta 0:09:01 lr 0.000023 time 0.9120 (1.0055) model_time 0.9118 (1.0047) loss 0.9234 (0.8732) grad_norm 6.1674 (8.8509/1.9188) mem 68106MB [2022-12-19 20:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][990/1519] eta 0:08:51 lr 0.000023 time 0.9296 (1.0055) model_time 0.9295 (1.0047) loss 0.9385 (0.8727) grad_norm 6.9574 (8.8143/1.9369) mem 68106MB [2022-12-19 20:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1000/1519] eta 0:08:41 lr 0.000023 time 0.9331 (1.0054) model_time 0.9330 (1.0047) loss 0.9764 (0.8731) grad_norm 7.7399 (8.8043/1.9303) mem 68106MB [2022-12-19 20:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1010/1519] eta 0:08:31 lr 0.000023 time 0.9267 (1.0054) model_time 0.9266 (1.0046) loss 0.8919 (0.8740) grad_norm 14.1055 (8.8087/1.9573) mem 68106MB [2022-12-19 20:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1020/1519] eta 0:08:21 lr 0.000023 time 0.9284 (1.0053) model_time 0.9280 (1.0045) loss 1.0336 (0.8739) grad_norm 11.7751 (8.7981/1.9706) mem 68106MB [2022-12-19 20:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1030/1519] eta 0:08:11 lr 0.000023 time 0.9325 (1.0052) model_time 0.9323 (1.0045) loss 0.9058 (0.8745) grad_norm 8.6912 (8.7730/1.9498) mem 68106MB [2022-12-19 20:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1040/1519] eta 0:08:01 lr 0.000023 time 0.9341 (1.0052) model_time 0.9340 (1.0044) loss 0.7751 (0.8741) grad_norm 7.2461 (8.7718/1.9486) mem 68106MB [2022-12-19 20:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1050/1519] eta 0:07:51 lr 0.000023 time 0.9326 (1.0051) model_time 0.9325 (1.0043) loss 0.9030 (0.8737) grad_norm 6.7573 (8.7611/1.9532) mem 68106MB [2022-12-19 20:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1060/1519] eta 0:07:41 lr 0.000023 time 0.9391 (1.0051) model_time 0.9389 (1.0043) loss 0.9942 (0.8736) grad_norm 8.2740 (8.7724/1.9426) mem 68106MB [2022-12-19 20:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1070/1519] eta 0:07:31 lr 0.000023 time 0.9315 (1.0051) model_time 0.9314 (1.0043) loss 1.0159 (0.8741) grad_norm 7.5348 (8.7603/1.9413) mem 68106MB [2022-12-19 20:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1080/1519] eta 0:07:21 lr 0.000023 time 0.9302 (1.0050) model_time 0.9301 (1.0042) loss 1.0184 (0.8745) grad_norm 7.4721 (8.7501/1.9798) mem 68106MB [2022-12-19 20:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1090/1519] eta 0:07:11 lr 0.000023 time 0.9194 (1.0050) model_time 0.9193 (1.0042) loss 0.9453 (0.8756) grad_norm 7.8871 (8.7445/1.9653) mem 68106MB [2022-12-19 20:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1100/1519] eta 0:07:01 lr 0.000023 time 0.9366 (1.0050) model_time 0.9365 (1.0042) loss 0.7677 (0.8752) grad_norm 12.9892 (8.7694/1.9753) mem 68106MB [2022-12-19 20:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1110/1519] eta 0:06:51 lr 0.000023 time 0.9146 (1.0053) model_time 0.9144 (1.0045) loss 0.8197 (0.8757) grad_norm 8.1240 (8.7591/1.9735) mem 68106MB [2022-12-19 20:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1120/1519] eta 0:06:41 lr 0.000023 time 0.9343 (1.0053) model_time 0.9342 (1.0045) loss 0.7951 (0.8755) grad_norm 10.8609 (8.7576/1.9654) mem 68106MB [2022-12-19 20:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1130/1519] eta 0:06:31 lr 0.000023 time 0.9367 (1.0054) model_time 0.9365 (1.0046) loss 0.8572 (0.8753) grad_norm 7.7860 (8.7451/1.9714) mem 68106MB [2022-12-19 20:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1140/1519] eta 0:06:21 lr 0.000023 time 0.9183 (1.0053) model_time 0.9181 (1.0046) loss 0.8763 (0.8746) grad_norm 7.7005 (8.7248/1.9555) mem 68106MB [2022-12-19 20:50:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1150/1519] eta 0:06:10 lr 0.000023 time 0.9379 (1.0054) model_time 0.9377 (1.0047) loss 0.8910 (0.8744) grad_norm 7.9594 (8.6815/1.9302) mem 68106MB [2022-12-19 20:50:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1160/1519] eta 0:06:00 lr 0.000023 time 0.9367 (1.0054) model_time 0.9364 (1.0047) loss 0.7154 (0.8742) grad_norm 9.8718 (8.6797/1.9259) mem 68106MB [2022-12-19 20:51:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1170/1519] eta 0:05:50 lr 0.000023 time 0.9382 (1.0054) model_time 0.9380 (1.0046) loss 0.8272 (0.8739) grad_norm 9.8389 (8.6720/1.9239) mem 68106MB [2022-12-19 20:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1180/1519] eta 0:05:40 lr 0.000023 time 0.9270 (1.0055) model_time 0.9269 (1.0048) loss 0.9032 (0.8735) grad_norm 14.2790 (8.7474/2.0685) mem 68106MB [2022-12-19 20:51:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1190/1519] eta 0:05:30 lr 0.000023 time 0.9278 (1.0056) model_time 0.9276 (1.0048) loss 1.2891 (0.8741) grad_norm 8.3353 (8.7401/2.0224) mem 68106MB [2022-12-19 20:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1200/1519] eta 0:05:20 lr 0.000023 time 0.9254 (1.0058) model_time 0.9252 (1.0051) loss 1.0172 (0.8743) grad_norm 7.5260 (8.7669/2.0631) mem 68106MB [2022-12-19 20:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1210/1519] eta 0:05:10 lr 0.000023 time 0.9297 (1.0059) model_time 0.9295 (1.0052) loss 1.0068 (0.8738) grad_norm 7.6195 (8.7462/2.0711) mem 68106MB [2022-12-19 20:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1220/1519] eta 0:05:00 lr 0.000023 time 0.9352 (1.0059) model_time 0.9350 (1.0052) loss 0.9201 (0.8741) grad_norm 7.9589 (8.7351/2.0720) mem 68106MB [2022-12-19 20:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1230/1519] eta 0:04:50 lr 0.000023 time 0.9302 (1.0059) model_time 0.9300 (1.0051) loss 0.9844 (0.8743) grad_norm 10.0852 (8.7363/2.0861) mem 68106MB [2022-12-19 20:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1240/1519] eta 0:04:40 lr 0.000023 time 0.9330 (1.0058) model_time 0.9328 (1.0051) loss 1.1648 (0.8744) grad_norm 10.3494 (8.7380/2.0814) mem 68106MB [2022-12-19 20:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1250/1519] eta 0:04:30 lr 0.000023 time 0.9164 (1.0059) model_time 0.9162 (1.0052) loss 0.7607 (0.8742) grad_norm 10.7509 (8.7369/2.0901) mem 68106MB [2022-12-19 20:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1260/1519] eta 0:04:20 lr 0.000023 time 0.9231 (1.0059) model_time 0.9229 (1.0052) loss 1.0738 (0.8742) grad_norm 12.8743 (8.7572/2.0999) mem 68106MB [2022-12-19 20:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1270/1519] eta 0:04:10 lr 0.000023 time 0.9241 (1.0059) model_time 0.9236 (1.0052) loss 0.8851 (0.8739) grad_norm 11.5257 (8.7589/2.1053) mem 68106MB [2022-12-19 20:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1280/1519] eta 0:04:00 lr 0.000023 time 0.9274 (1.0059) model_time 0.9272 (1.0052) loss 0.7261 (0.8742) grad_norm 12.3853 (8.7662/2.1079) mem 68106MB [2022-12-19 20:53:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1290/1519] eta 0:03:50 lr 0.000023 time 0.9342 (1.0060) model_time 0.9341 (1.0053) loss 0.7325 (0.8739) grad_norm 9.6631 (8.7871/2.1118) mem 68106MB [2022-12-19 20:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1300/1519] eta 0:03:40 lr 0.000023 time 0.9271 (1.0060) model_time 0.9269 (1.0053) loss 0.8674 (0.8741) grad_norm 5.9457 (8.7762/2.1081) mem 68106MB [2022-12-19 20:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1310/1519] eta 0:03:30 lr 0.000023 time 0.9291 (1.0060) model_time 0.9289 (1.0053) loss 0.7065 (0.8741) grad_norm 10.2801 (8.7333/2.0443) mem 68106MB [2022-12-19 20:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1320/1519] eta 0:03:20 lr 0.000023 time 0.9356 (1.0061) model_time 0.9354 (1.0054) loss 0.8632 (0.8738) grad_norm 15.7926 (8.7523/2.0825) mem 68106MB [2022-12-19 20:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1330/1519] eta 0:03:10 lr 0.000023 time 0.9353 (1.0061) model_time 0.9351 (1.0054) loss 0.6977 (0.8739) grad_norm 8.8744 (8.7869/2.2293) mem 68106MB [2022-12-19 20:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1340/1519] eta 0:03:00 lr 0.000023 time 0.9279 (1.0060) model_time 0.9276 (1.0053) loss 0.9526 (0.8741) grad_norm 7.3559 (8.7988/2.2452) mem 68106MB [2022-12-19 20:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1350/1519] eta 0:02:50 lr 0.000023 time 0.9305 (1.0059) model_time 0.9303 (1.0052) loss 1.0500 (0.8743) grad_norm 9.2111 (8.7816/2.2336) mem 68106MB [2022-12-19 20:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1360/1519] eta 0:02:39 lr 0.000023 time 0.9342 (1.0059) model_time 0.9340 (1.0052) loss 0.7196 (0.8740) grad_norm 7.2982 (8.7962/2.2286) mem 68106MB [2022-12-19 20:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1370/1519] eta 0:02:29 lr 0.000023 time 0.9337 (1.0059) model_time 0.9336 (1.0052) loss 0.7560 (0.8739) grad_norm 8.1290 (8.7952/2.1995) mem 68106MB [2022-12-19 20:54:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1380/1519] eta 0:02:19 lr 0.000023 time 1.0228 (1.0059) model_time 1.0226 (1.0053) loss 0.8581 (0.8741) grad_norm 13.8609 (8.8170/2.2202) mem 68106MB [2022-12-19 20:54:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1390/1519] eta 0:02:09 lr 0.000023 time 0.9422 (1.0059) model_time 0.9420 (1.0052) loss 1.1389 (0.8740) grad_norm 10.0044 (8.8071/2.2185) mem 68106MB [2022-12-19 20:54:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1400/1519] eta 0:01:59 lr 0.000023 time 0.9332 (1.0058) model_time 0.9330 (1.0052) loss 1.0405 (0.8749) grad_norm 7.9252 (8.8143/2.2289) mem 68106MB [2022-12-19 20:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1410/1519] eta 0:01:49 lr 0.000023 time 0.9275 (1.0059) model_time 0.9273 (1.0052) loss 0.8396 (0.8752) grad_norm 8.8295 (8.8112/2.2333) mem 68106MB [2022-12-19 20:55:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1420/1519] eta 0:01:39 lr 0.000023 time 0.8782 (1.0060) model_time 0.8781 (1.0054) loss 0.8942 (0.8747) grad_norm 6.0798 (8.8131/2.2334) mem 68106MB [2022-12-19 20:55:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1430/1519] eta 0:01:29 lr 0.000023 time 0.9337 (1.0061) model_time 0.9336 (1.0054) loss 1.2562 (0.8754) grad_norm 10.0165 (8.8637/2.2392) mem 68106MB [2022-12-19 20:55:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1440/1519] eta 0:01:19 lr 0.000023 time 0.9412 (1.0063) model_time 0.9411 (1.0056) loss 0.9917 (0.8755) grad_norm 8.0169 (8.8518/2.2401) mem 68106MB [2022-12-19 20:55:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1450/1519] eta 0:01:09 lr 0.000023 time 0.9283 (1.0062) model_time 0.9281 (1.0056) loss 0.9724 (0.8757) grad_norm 6.1075 (8.8449/2.2568) mem 68106MB [2022-12-19 20:55:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1460/1519] eta 0:00:59 lr 0.000023 time 0.9423 (1.0063) model_time 0.9421 (1.0056) loss 0.8263 (0.8760) grad_norm 7.6770 (8.8604/2.2711) mem 68106MB [2022-12-19 20:56:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1470/1519] eta 0:00:49 lr 0.000023 time 0.9361 (1.0062) model_time 0.9338 (1.0056) loss 0.6932 (0.8761) grad_norm 8.5084 (8.8448/2.2728) mem 68106MB [2022-12-19 20:56:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1480/1519] eta 0:00:39 lr 0.000023 time 0.9298 (1.0062) model_time 0.9297 (1.0055) loss 0.6796 (0.8758) grad_norm 5.6999 (8.8176/2.2803) mem 68106MB [2022-12-19 20:56:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1490/1519] eta 0:00:29 lr 0.000023 time 0.9377 (1.0062) model_time 0.9376 (1.0055) loss 0.8948 (0.8758) grad_norm 10.6370 (8.8423/2.3054) mem 68106MB [2022-12-19 20:56:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1500/1519] eta 0:00:19 lr 0.000023 time 0.9380 (1.0061) model_time 0.9377 (1.0055) loss 0.6822 (0.8754) grad_norm 7.7957 (8.8509/2.2985) mem 68106MB [2022-12-19 20:56:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [37/100][1510/1519] eta 0:00:09 lr 0.000023 time 0.9229 (1.0061) model_time 0.9228 (1.0055) loss 0.7066 (0.8756) grad_norm 10.5074 (8.8920/2.2831) mem 68106MB [2022-12-19 20:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 37 training takes 0:25:28 [2022-12-19 20:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_37.pth saving...... [2022-12-19 20:57:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_37.pth saved !!! [2022-12-19 20:57:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.665 (0.665) Loss 0.5173 (0.5173) Acc@1 90.972 (90.972) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 20:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.330) Loss 0.5067 (0.4901) Acc@1 91.319 (91.856) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-19 20:57:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.314) Loss 0.4553 (0.4928) Acc@1 91.667 (91.733) Acc@5 98.958 (98.380) Mem 68106MB [2022-12-19 20:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.304 (0.309) Loss 0.5749 (0.4960) Acc@1 90.278 (91.599) Acc@5 97.222 (98.297) Mem 68106MB [2022-12-19 20:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.307) Loss 0.4779 (0.4879) Acc@1 91.667 (91.701) Acc@5 98.958 (98.357) Mem 68106MB [2022-12-19 20:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.306) Loss 0.4999 (0.4854) Acc@1 90.625 (91.653) Acc@5 98.958 (98.407) Mem 68106MB [2022-12-19 20:57:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.306 (0.305) Loss 0.4807 (0.4849) Acc@1 91.667 (91.650) Acc@5 98.264 (98.412) Mem 68106MB [2022-12-19 20:57:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5320 (0.4864) Acc@1 92.014 (91.623) Acc@5 97.917 (98.406) Mem 68106MB [2022-12-19 20:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.306 (0.303) Loss 0.3980 (0.4841) Acc@1 92.361 (91.662) Acc@5 98.958 (98.453) Mem 68106MB [2022-12-19 20:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:37] * Acc@1 91.630 Acc@5 98.449 [2022-12-19 20:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.6% [2022-12-19 20:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.66% [2022-12-19 20:57:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][0/1519] eta 0:45:09 lr 0.000023 time 1.7840 (1.7840) model_time 0.9707 (0.9707) loss 0.7012 (0.7012) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 20:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][10/1519] eta 0:26:54 lr 0.000023 time 0.9329 (1.0702) model_time 0.9328 (0.9959) loss 0.8951 (0.8210) grad_norm 7.3096 (7.7412/0.5832) mem 68106MB [2022-12-19 20:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][20/1519] eta 0:25:54 lr 0.000023 time 0.9397 (1.0368) model_time 0.9395 (0.9977) loss 0.8007 (0.8090) grad_norm 7.7827 (7.6708/0.5741) mem 68106MB [2022-12-19 20:58:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][30/1519] eta 0:25:26 lr 0.000023 time 0.9377 (1.0249) model_time 0.9375 (0.9983) loss 0.9843 (0.8217) grad_norm 8.5436 (7.8390/0.8829) mem 68106MB [2022-12-19 20:58:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][40/1519] eta 0:25:08 lr 0.000023 time 0.9291 (1.0199) model_time 0.9289 (0.9996) loss 0.7321 (0.8282) grad_norm 7.6951 (7.9286/0.9210) mem 68106MB [2022-12-19 20:58:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][50/1519] eta 0:24:57 lr 0.000023 time 0.9378 (1.0191) model_time 0.9377 (1.0028) loss 0.7871 (0.8279) grad_norm 8.7539 (8.0318/0.9291) mem 68106MB [2022-12-19 20:58:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][60/1519] eta 0:24:45 lr 0.000023 time 0.9816 (1.0179) model_time 0.9814 (1.0041) loss 0.8435 (0.8357) grad_norm 9.1883 (8.1896/1.0949) mem 68106MB [2022-12-19 20:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][70/1519] eta 0:24:31 lr 0.000023 time 0.9905 (1.0158) model_time 0.9903 (1.0040) loss 1.0417 (0.8486) grad_norm 8.7766 (8.2886/1.0867) mem 68106MB [2022-12-19 20:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][80/1519] eta 0:24:19 lr 0.000023 time 0.9326 (1.0140) model_time 0.9324 (1.0036) loss 0.8687 (0.8454) grad_norm 8.2103 (8.3114/1.1217) mem 68106MB [2022-12-19 20:59:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][90/1519] eta 0:24:07 lr 0.000023 time 0.9357 (1.0127) model_time 0.9355 (1.0034) loss 0.7372 (0.8507) grad_norm 8.2975 (8.1952/1.1459) mem 68106MB [2022-12-19 20:59:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][100/1519] eta 0:23:57 lr 0.000023 time 1.0263 (1.0133) model_time 1.0262 (1.0048) loss 0.7254 (0.8469) grad_norm 5.9515 (8.3156/1.3136) mem 68106MB [2022-12-19 20:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][110/1519] eta 0:23:48 lr 0.000023 time 0.9399 (1.0140) model_time 0.9398 (1.0063) loss 0.7880 (0.8505) grad_norm 7.6519 (8.2423/1.2998) mem 68106MB [2022-12-19 20:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][120/1519] eta 0:23:37 lr 0.000023 time 0.9323 (1.0135) model_time 0.9322 (1.0064) loss 0.9065 (0.8517) grad_norm 11.9237 (8.2918/1.3617) mem 68106MB [2022-12-19 20:59:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][130/1519] eta 0:23:26 lr 0.000023 time 0.9506 (1.0128) model_time 0.9504 (1.0062) loss 0.8259 (0.8562) grad_norm 8.9463 (8.2950/1.3808) mem 68106MB [2022-12-19 21:00:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][140/1519] eta 0:23:15 lr 0.000023 time 0.9328 (1.0120) model_time 0.9324 (1.0059) loss 0.7571 (0.8557) grad_norm 8.1806 (8.3829/1.4152) mem 68106MB [2022-12-19 21:00:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][150/1519] eta 0:23:07 lr 0.000023 time 0.9308 (1.0132) model_time 0.9306 (1.0074) loss 0.8559 (0.8588) grad_norm 9.5898 (8.4108/1.4047) mem 68106MB [2022-12-19 21:00:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][160/1519] eta 0:22:55 lr 0.000023 time 0.9296 (1.0122) model_time 0.9294 (1.0068) loss 0.8108 (0.8582) grad_norm 11.9097 (8.4062/1.4349) mem 68106MB [2022-12-19 21:00:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][170/1519] eta 0:22:45 lr 0.000023 time 0.9301 (1.0121) model_time 0.9299 (1.0070) loss 1.0373 (0.8538) grad_norm 11.8277 (8.4044/1.4911) mem 68106MB [2022-12-19 21:00:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][180/1519] eta 0:22:34 lr 0.000023 time 0.9341 (1.0114) model_time 0.9339 (1.0066) loss 1.1501 (0.8534) grad_norm 6.8173 (8.4447/1.5544) mem 68106MB [2022-12-19 21:00:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][190/1519] eta 0:22:23 lr 0.000023 time 0.9317 (1.0109) model_time 0.9315 (1.0062) loss 0.8400 (0.8552) grad_norm 8.0269 (8.4156/1.5393) mem 68106MB [2022-12-19 21:01:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][200/1519] eta 0:22:12 lr 0.000023 time 0.9312 (1.0102) model_time 0.9310 (1.0058) loss 0.7157 (0.8511) grad_norm 7.3175 (8.4035/1.5173) mem 68106MB [2022-12-19 21:01:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][210/1519] eta 0:22:02 lr 0.000023 time 0.9361 (1.0107) model_time 0.9359 (1.0065) loss 0.6994 (0.8514) grad_norm 10.3721 (8.4196/1.4970) mem 68106MB [2022-12-19 21:01:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][220/1519] eta 0:21:52 lr 0.000023 time 1.0180 (1.0106) model_time 1.0178 (1.0065) loss 0.7125 (0.8512) grad_norm 9.9257 (8.4295/1.5956) mem 68106MB [2022-12-19 21:01:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][230/1519] eta 0:21:42 lr 0.000023 time 0.9706 (1.0106) model_time 0.9704 (1.0067) loss 0.7332 (0.8504) grad_norm 9.1007 (8.4233/1.5972) mem 68106MB [2022-12-19 21:01:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][240/1519] eta 0:21:32 lr 0.000023 time 0.9350 (1.0103) model_time 0.9348 (1.0066) loss 0.7148 (0.8545) grad_norm 8.9335 (8.5023/1.6587) mem 68106MB [2022-12-19 21:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][250/1519] eta 0:21:22 lr 0.000023 time 1.0351 (1.0103) model_time 1.0350 (1.0067) loss 0.7764 (0.8555) grad_norm 7.8145 (8.4877/1.6643) mem 68106MB [2022-12-19 21:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][260/1519] eta 0:21:11 lr 0.000023 time 0.9323 (1.0098) model_time 0.9321 (1.0063) loss 0.9985 (0.8570) grad_norm 7.7233 (8.5147/1.6797) mem 68106MB [2022-12-19 21:02:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][270/1519] eta 0:21:00 lr 0.000023 time 0.9316 (1.0093) model_time 0.9315 (1.0060) loss 0.9618 (0.8560) grad_norm 7.2911 (8.4912/1.6574) mem 68106MB [2022-12-19 21:02:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][280/1519] eta 0:20:50 lr 0.000023 time 0.9447 (1.0091) model_time 0.9444 (1.0058) loss 1.0043 (0.8553) grad_norm 8.3518 (8.4772/1.6320) mem 68106MB [2022-12-19 21:02:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][290/1519] eta 0:20:39 lr 0.000023 time 0.9350 (1.0087) model_time 0.9349 (1.0055) loss 0.8202 (0.8548) grad_norm 7.8579 (8.4793/1.6599) mem 68106MB [2022-12-19 21:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][300/1519] eta 0:20:29 lr 0.000023 time 0.9343 (1.0083) model_time 0.9340 (1.0052) loss 0.8361 (0.8548) grad_norm 9.5652 (8.4619/1.6433) mem 68106MB [2022-12-19 21:02:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][310/1519] eta 0:20:18 lr 0.000023 time 0.9498 (1.0080) model_time 0.9496 (1.0050) loss 0.9156 (0.8573) grad_norm 8.2844 (8.4631/1.6699) mem 68106MB [2022-12-19 21:03:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][320/1519] eta 0:20:08 lr 0.000023 time 0.9297 (1.0079) model_time 0.9295 (1.0050) loss 0.7357 (0.8594) grad_norm 13.6335 (8.4960/1.7059) mem 68106MB [2022-12-19 21:03:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][330/1519] eta 0:19:59 lr 0.000023 time 0.9333 (1.0084) model_time 0.9331 (1.0056) loss 0.8173 (0.8582) grad_norm 5.8884 (8.5076/1.7699) mem 68106MB [2022-12-19 21:03:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][340/1519] eta 0:19:48 lr 0.000023 time 0.9476 (1.0082) model_time 0.9475 (1.0054) loss 0.8410 (0.8608) grad_norm 7.9473 (8.5171/1.7728) mem 68106MB [2022-12-19 21:03:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][350/1519] eta 0:19:38 lr 0.000023 time 0.9345 (1.0080) model_time 0.9342 (1.0053) loss 1.1885 (0.8639) grad_norm 11.0581 (8.5269/1.7927) mem 68106MB [2022-12-19 21:03:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][360/1519] eta 0:19:28 lr 0.000023 time 0.9344 (1.0078) model_time 0.9342 (1.0052) loss 0.8663 (0.8628) grad_norm 9.1939 (8.5494/1.7811) mem 68106MB [2022-12-19 21:03:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][370/1519] eta 0:19:17 lr 0.000023 time 0.9312 (1.0078) model_time 0.9310 (1.0052) loss 0.8819 (0.8631) grad_norm 10.8749 (8.5543/1.7745) mem 68106MB [2022-12-19 21:04:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][380/1519] eta 0:19:07 lr 0.000023 time 0.9371 (1.0077) model_time 0.9370 (1.0052) loss 0.6920 (0.8627) grad_norm 6.7296 (8.5443/1.7592) mem 68106MB [2022-12-19 21:04:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][390/1519] eta 0:18:57 lr 0.000023 time 0.9353 (1.0076) model_time 0.9352 (1.0051) loss 1.2638 (0.8649) grad_norm 11.4398 (8.5617/1.7761) mem 68106MB [2022-12-19 21:04:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][400/1519] eta 0:18:47 lr 0.000023 time 1.0051 (1.0075) model_time 1.0050 (1.0051) loss 0.7447 (0.8636) grad_norm 7.1884 (8.5717/1.7693) mem 68106MB [2022-12-19 21:04:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][410/1519] eta 0:18:37 lr 0.000023 time 0.9294 (1.0075) model_time 0.9287 (1.0052) loss 0.8032 (0.8650) grad_norm 7.7158 (8.5784/1.7652) mem 68106MB [2022-12-19 21:04:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][420/1519] eta 0:18:27 lr 0.000023 time 0.9287 (1.0078) model_time 0.9285 (1.0055) loss 1.0453 (0.8654) grad_norm 11.9681 (8.5782/1.7653) mem 68106MB [2022-12-19 21:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][430/1519] eta 0:18:17 lr 0.000023 time 0.9307 (1.0081) model_time 0.9305 (1.0059) loss 0.6676 (0.8636) grad_norm 6.9235 (8.5637/1.7546) mem 68106MB [2022-12-19 21:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][440/1519] eta 0:18:07 lr 0.000023 time 0.9307 (1.0080) model_time 0.9306 (1.0057) loss 0.6881 (0.8638) grad_norm 9.1298 (8.6176/1.7971) mem 68106MB [2022-12-19 21:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][450/1519] eta 0:17:57 lr 0.000023 time 0.9156 (1.0078) model_time 0.9155 (1.0056) loss 0.8124 (0.8652) grad_norm 11.6355 (8.6242/1.8039) mem 68106MB [2022-12-19 21:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][460/1519] eta 0:17:47 lr 0.000023 time 0.9305 (1.0077) model_time 0.9303 (1.0056) loss 0.7112 (0.8641) grad_norm 7.9554 (8.6424/1.8181) mem 68106MB [2022-12-19 21:05:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][470/1519] eta 0:17:37 lr 0.000023 time 0.9386 (1.0076) model_time 0.9385 (1.0055) loss 1.2715 (0.8654) grad_norm 6.2701 (8.6550/1.8318) mem 68106MB [2022-12-19 21:05:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][480/1519] eta 0:17:26 lr 0.000023 time 0.9440 (1.0076) model_time 0.9439 (1.0056) loss 0.7896 (0.8633) grad_norm 6.1241 (8.6404/1.8208) mem 68106MB [2022-12-19 21:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][490/1519] eta 0:17:16 lr 0.000023 time 0.9390 (1.0077) model_time 0.9389 (1.0057) loss 1.0005 (0.8660) grad_norm 7.4493 (8.6382/1.8102) mem 68106MB [2022-12-19 21:06:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][500/1519] eta 0:17:06 lr 0.000023 time 0.9507 (1.0077) model_time 0.9506 (1.0057) loss 0.7367 (0.8655) grad_norm 5.5878 (8.5993/1.8212) mem 68106MB [2022-12-19 21:06:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][510/1519] eta 0:16:56 lr 0.000023 time 0.9356 (1.0075) model_time 0.9355 (1.0055) loss 0.9551 (0.8661) grad_norm 8.2724 (8.5921/1.8279) mem 68106MB [2022-12-19 21:06:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][520/1519] eta 0:16:46 lr 0.000023 time 0.9293 (1.0077) model_time 0.9292 (1.0057) loss 0.6939 (0.8673) grad_norm 13.9783 (8.5915/1.8570) mem 68106MB [2022-12-19 21:06:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][530/1519] eta 0:16:36 lr 0.000023 time 0.9281 (1.0077) model_time 0.9280 (1.0057) loss 0.8109 (0.8670) grad_norm 9.4853 (8.5779/1.8474) mem 68106MB [2022-12-19 21:06:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][540/1519] eta 0:16:26 lr 0.000023 time 0.9342 (1.0076) model_time 0.9340 (1.0057) loss 0.9552 (0.8665) grad_norm 8.2709 (8.5801/1.8586) mem 68106MB [2022-12-19 21:07:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][550/1519] eta 0:16:16 lr 0.000023 time 0.9340 (1.0074) model_time 0.9339 (1.0055) loss 0.9711 (0.8667) grad_norm 8.8092 (8.6108/1.9270) mem 68106MB [2022-12-19 21:07:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][560/1519] eta 0:16:06 lr 0.000023 time 0.9267 (1.0073) model_time 0.9266 (1.0055) loss 1.0145 (0.8663) grad_norm 6.1249 (8.6104/1.9297) mem 68106MB [2022-12-19 21:07:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][570/1519] eta 0:15:55 lr 0.000023 time 0.9404 (1.0073) model_time 0.9402 (1.0055) loss 0.9464 (0.8678) grad_norm 7.7485 (8.6085/1.9218) mem 68106MB [2022-12-19 21:07:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][580/1519] eta 0:15:45 lr 0.000023 time 0.9361 (1.0072) model_time 0.9358 (1.0054) loss 0.9341 (0.8667) grad_norm 10.4388 (8.6120/1.9101) mem 68106MB [2022-12-19 21:07:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][590/1519] eta 0:15:35 lr 0.000023 time 0.9282 (1.0072) model_time 0.9280 (1.0054) loss 0.9428 (0.8666) grad_norm 12.6511 (8.6217/1.9104) mem 68106MB [2022-12-19 21:07:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][600/1519] eta 0:15:25 lr 0.000023 time 0.9332 (1.0071) model_time 0.9330 (1.0053) loss 0.7790 (0.8656) grad_norm 10.0851 (8.6170/1.9018) mem 68106MB [2022-12-19 21:08:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][610/1519] eta 0:15:15 lr 0.000023 time 0.9363 (1.0071) model_time 0.9362 (1.0054) loss 1.1034 (0.8666) grad_norm 7.6785 (8.6254/1.8996) mem 68106MB [2022-12-19 21:08:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][620/1519] eta 0:15:05 lr 0.000023 time 0.9374 (1.0070) model_time 0.9373 (1.0053) loss 1.1771 (0.8664) grad_norm 10.4648 (8.6400/1.8991) mem 68106MB [2022-12-19 21:08:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][630/1519] eta 0:14:55 lr 0.000023 time 0.9329 (1.0069) model_time 0.9328 (1.0052) loss 1.1098 (0.8666) grad_norm 6.4675 (8.6473/1.9042) mem 68106MB [2022-12-19 21:08:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][640/1519] eta 0:14:45 lr 0.000023 time 0.9184 (1.0068) model_time 0.9183 (1.0052) loss 0.6841 (0.8669) grad_norm 10.5541 (8.6440/1.9208) mem 68106MB [2022-12-19 21:08:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][650/1519] eta 0:14:34 lr 0.000023 time 0.9453 (1.0068) model_time 0.9451 (1.0052) loss 0.6818 (0.8657) grad_norm 9.0996 (8.6787/1.9530) mem 68106MB [2022-12-19 21:08:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][660/1519] eta 0:14:24 lr 0.000023 time 0.9803 (1.0068) model_time 0.9801 (1.0052) loss 0.6864 (0.8668) grad_norm 10.4080 (8.6796/1.9578) mem 68106MB [2022-12-19 21:09:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][670/1519] eta 0:14:14 lr 0.000023 time 0.9770 (1.0068) model_time 0.9769 (1.0052) loss 0.7090 (0.8658) grad_norm 9.1427 (8.7081/2.0186) mem 68106MB [2022-12-19 21:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][680/1519] eta 0:14:04 lr 0.000023 time 1.0062 (1.0069) model_time 1.0060 (1.0053) loss 0.7077 (0.8654) grad_norm 8.1684 (8.7028/2.0137) mem 68106MB [2022-12-19 21:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][690/1519] eta 0:13:54 lr 0.000023 time 0.9315 (1.0067) model_time 0.9314 (1.0052) loss 0.8336 (0.8663) grad_norm 8.2499 (8.7106/2.0090) mem 68106MB [2022-12-19 21:09:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][700/1519] eta 0:13:44 lr 0.000023 time 0.9377 (1.0067) model_time 0.9376 (1.0052) loss 1.0338 (0.8661) grad_norm 8.1996 (8.7109/1.9946) mem 68106MB [2022-12-19 21:09:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][710/1519] eta 0:13:34 lr 0.000023 time 0.9398 (1.0066) model_time 0.9397 (1.0051) loss 0.6973 (0.8660) grad_norm 9.1555 (8.7522/2.0289) mem 68106MB [2022-12-19 21:09:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][720/1519] eta 0:13:24 lr 0.000023 time 0.9343 (1.0067) model_time 0.9342 (1.0052) loss 0.8059 (0.8665) grad_norm 7.1456 (8.7647/2.0309) mem 68106MB [2022-12-19 21:10:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][730/1519] eta 0:13:14 lr 0.000023 time 0.9681 (1.0071) model_time 0.9679 (1.0056) loss 0.8680 (0.8671) grad_norm 7.9418 (8.7584/2.0274) mem 68106MB [2022-12-19 21:10:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][740/1519] eta 0:13:04 lr 0.000023 time 0.9382 (1.0070) model_time 0.9380 (1.0055) loss 0.8954 (0.8680) grad_norm 9.3228 (8.7390/2.0203) mem 68106MB [2022-12-19 21:10:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][750/1519] eta 0:12:54 lr 0.000023 time 0.9409 (1.0069) model_time 0.9408 (1.0054) loss 0.9531 (0.8685) grad_norm 9.3464 (8.7283/2.0202) mem 68106MB [2022-12-19 21:10:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][760/1519] eta 0:12:44 lr 0.000023 time 0.9258 (1.0068) model_time 0.9257 (1.0053) loss 0.8100 (0.8690) grad_norm 11.3155 (8.7357/2.0191) mem 68106MB [2022-12-19 21:10:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][770/1519] eta 0:12:34 lr 0.000023 time 0.9340 (1.0071) model_time 0.9338 (1.0057) loss 0.8438 (0.8692) grad_norm 8.4579 (8.7538/2.0045) mem 68106MB [2022-12-19 21:10:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][780/1519] eta 0:12:24 lr 0.000023 time 0.9310 (1.0070) model_time 0.9309 (1.0056) loss 0.7656 (0.8697) grad_norm 7.6112 (8.7405/1.9858) mem 68106MB [2022-12-19 21:11:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][790/1519] eta 0:12:14 lr 0.000023 time 0.9511 (1.0070) model_time 0.9510 (1.0056) loss 0.8956 (0.8693) grad_norm 9.9732 (8.7470/1.9837) mem 68106MB [2022-12-19 21:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][800/1519] eta 0:12:03 lr 0.000023 time 0.9382 (1.0069) model_time 0.9381 (1.0055) loss 0.8157 (0.8690) grad_norm 6.1160 (8.7442/1.9865) mem 68106MB [2022-12-19 21:11:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][810/1519] eta 0:11:53 lr 0.000023 time 0.9332 (1.0068) model_time 0.9330 (1.0054) loss 0.9252 (0.8686) grad_norm 8.8946 (8.7559/1.9952) mem 68106MB [2022-12-19 21:11:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][820/1519] eta 0:11:43 lr 0.000023 time 0.9307 (1.0067) model_time 0.9305 (1.0053) loss 0.8128 (0.8688) grad_norm 9.9389 (8.7670/1.9808) mem 68106MB [2022-12-19 21:11:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][830/1519] eta 0:11:33 lr 0.000023 time 0.9271 (1.0068) model_time 0.9270 (1.0054) loss 0.9111 (0.8688) grad_norm 8.4240 (8.7698/1.9708) mem 68106MB [2022-12-19 21:11:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][840/1519] eta 0:11:23 lr 0.000023 time 0.9322 (1.0067) model_time 0.9321 (1.0054) loss 0.7351 (0.8678) grad_norm 8.0386 (8.7333/1.9460) mem 68106MB [2022-12-19 21:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][850/1519] eta 0:11:13 lr 0.000023 time 0.9174 (1.0068) model_time 0.9172 (1.0054) loss 1.0122 (0.8679) grad_norm 6.8546 (8.7444/1.9658) mem 68106MB [2022-12-19 21:12:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][860/1519] eta 0:11:03 lr 0.000023 time 0.9277 (1.0067) model_time 0.9275 (1.0053) loss 0.9812 (0.8677) grad_norm 13.2563 (8.7677/1.9776) mem 68106MB [2022-12-19 21:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][870/1519] eta 0:10:53 lr 0.000023 time 0.9371 (1.0066) model_time 0.9370 (1.0053) loss 1.0603 (0.8676) grad_norm 7.9276 (8.7928/1.9939) mem 68106MB [2022-12-19 21:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][880/1519] eta 0:10:43 lr 0.000023 time 0.9361 (1.0066) model_time 0.9360 (1.0053) loss 0.9225 (0.8676) grad_norm 7.6234 (8.7792/2.0022) mem 68106MB [2022-12-19 21:12:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][890/1519] eta 0:10:33 lr 0.000023 time 0.9328 (1.0065) model_time 0.9326 (1.0052) loss 0.8100 (0.8672) grad_norm 8.7978 (8.7701/1.9851) mem 68106MB [2022-12-19 21:12:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][900/1519] eta 0:10:23 lr 0.000023 time 0.9320 (1.0065) model_time 0.9319 (1.0053) loss 0.7179 (0.8668) grad_norm 8.5394 (8.7799/1.9812) mem 68106MB [2022-12-19 21:13:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][910/1519] eta 0:10:12 lr 0.000023 time 0.9780 (1.0065) model_time 0.9777 (1.0052) loss 0.8232 (0.8675) grad_norm 9.0646 (8.7725/1.9645) mem 68106MB [2022-12-19 21:13:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][920/1519] eta 0:10:02 lr 0.000023 time 0.9278 (1.0065) model_time 0.9277 (1.0052) loss 0.9602 (0.8679) grad_norm 8.8229 (8.7834/1.9783) mem 68106MB [2022-12-19 21:13:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][930/1519] eta 0:09:52 lr 0.000023 time 0.9254 (1.0064) model_time 0.9252 (1.0052) loss 0.9277 (0.8676) grad_norm 6.7224 (8.7551/1.9495) mem 68106MB [2022-12-19 21:13:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][940/1519] eta 0:09:42 lr 0.000023 time 0.9365 (1.0063) model_time 0.9364 (1.0051) loss 0.8432 (0.8682) grad_norm 9.9919 (8.8189/2.2184) mem 68106MB [2022-12-19 21:13:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][950/1519] eta 0:09:32 lr 0.000023 time 0.9289 (1.0063) model_time 0.9286 (1.0051) loss 0.7679 (0.8677) grad_norm 15.8377 (8.8342/2.2412) mem 68106MB [2022-12-19 21:13:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][960/1519] eta 0:09:22 lr 0.000023 time 0.9253 (1.0063) model_time 0.9252 (1.0051) loss 0.7089 (0.8672) grad_norm 10.5848 (8.8255/2.2396) mem 68106MB [2022-12-19 21:14:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][970/1519] eta 0:09:12 lr 0.000023 time 0.9289 (1.0062) model_time 0.9288 (1.0050) loss 0.6808 (0.8667) grad_norm 7.9831 (8.8427/2.2925) mem 68106MB [2022-12-19 21:14:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][980/1519] eta 0:09:02 lr 0.000023 time 0.9355 (1.0062) model_time 0.9354 (1.0050) loss 0.8647 (0.8667) grad_norm 5.9259 (8.8408/2.2986) mem 68106MB [2022-12-19 21:14:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][990/1519] eta 0:08:52 lr 0.000023 time 0.9272 (1.0062) model_time 0.9271 (1.0051) loss 0.8706 (0.8666) grad_norm 7.5105 (8.8320/2.2822) mem 68106MB [2022-12-19 21:14:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1000/1519] eta 0:08:42 lr 0.000023 time 0.9310 (1.0062) model_time 0.9309 (1.0050) loss 1.0580 (0.8666) grad_norm 8.4940 (8.8187/2.2846) mem 68106MB [2022-12-19 21:14:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1010/1519] eta 0:08:32 lr 0.000023 time 0.9687 (1.0063) model_time 0.9685 (1.0051) loss 1.3969 (0.8667) grad_norm 7.6673 (8.8070/2.2830) mem 68106MB [2022-12-19 21:14:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1020/1519] eta 0:08:22 lr 0.000023 time 0.9362 (1.0062) model_time 0.9361 (1.0050) loss 1.1896 (0.8674) grad_norm 11.0568 (8.8160/2.2820) mem 68106MB [2022-12-19 21:15:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1030/1519] eta 0:08:12 lr 0.000023 time 0.9341 (1.0061) model_time 0.9339 (1.0050) loss 1.0248 (0.8672) grad_norm 9.2937 (8.8070/2.2860) mem 68106MB [2022-12-19 21:15:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1040/1519] eta 0:08:01 lr 0.000023 time 0.9727 (1.0062) model_time 0.9726 (1.0051) loss 0.6778 (0.8670) grad_norm 10.2944 (8.7617/2.2592) mem 68106MB [2022-12-19 21:15:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1050/1519] eta 0:07:51 lr 0.000023 time 0.9352 (1.0061) model_time 0.9351 (1.0050) loss 0.9262 (0.8678) grad_norm 9.4038 (8.7477/2.2508) mem 68106MB [2022-12-19 21:15:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1060/1519] eta 0:07:41 lr 0.000023 time 0.9283 (1.0061) model_time 0.9282 (1.0049) loss 0.9040 (0.8676) grad_norm 9.5530 (8.7754/2.2833) mem 68106MB [2022-12-19 21:15:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1070/1519] eta 0:07:31 lr 0.000023 time 0.9439 (1.0060) model_time 0.9437 (1.0049) loss 0.7821 (0.8677) grad_norm 10.0497 (8.7719/2.2701) mem 68106MB [2022-12-19 21:15:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1080/1519] eta 0:07:21 lr 0.000023 time 0.9154 (1.0061) model_time 0.9153 (1.0049) loss 0.9739 (0.8674) grad_norm 6.3837 (8.7660/2.2715) mem 68106MB [2022-12-19 21:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1090/1519] eta 0:07:11 lr 0.000023 time 0.9484 (1.0060) model_time 0.9482 (1.0049) loss 0.8293 (0.8673) grad_norm 7.9812 (8.7684/2.2691) mem 68106MB [2022-12-19 21:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1100/1519] eta 0:07:01 lr 0.000023 time 0.9357 (1.0060) model_time 0.9356 (1.0049) loss 0.7091 (0.8672) grad_norm 8.2039 (8.8015/2.2518) mem 68106MB [2022-12-19 21:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1110/1519] eta 0:06:51 lr 0.000023 time 0.9317 (1.0060) model_time 0.9316 (1.0049) loss 0.8659 (0.8673) grad_norm 7.3539 (8.8050/2.2646) mem 68106MB [2022-12-19 21:16:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1120/1519] eta 0:06:41 lr 0.000023 time 0.9309 (1.0059) model_time 0.9307 (1.0048) loss 0.9496 (0.8675) grad_norm 10.4471 (8.8037/2.2437) mem 68106MB [2022-12-19 21:16:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1130/1519] eta 0:06:31 lr 0.000023 time 0.9306 (1.0059) model_time 0.9304 (1.0048) loss 0.9251 (0.8676) grad_norm 6.0167 (8.8082/2.2434) mem 68106MB [2022-12-19 21:16:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1140/1519] eta 0:06:21 lr 0.000023 time 0.8820 (1.0059) model_time 0.8819 (1.0048) loss 0.6887 (0.8675) grad_norm 7.9798 (8.8006/2.2252) mem 68106MB [2022-12-19 21:17:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1150/1519] eta 0:06:11 lr 0.000023 time 0.9382 (1.0059) model_time 0.9381 (1.0048) loss 0.8645 (0.8675) grad_norm 9.2070 (8.7615/2.1660) mem 68106MB [2022-12-19 21:17:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1160/1519] eta 0:06:01 lr 0.000023 time 1.0031 (1.0059) model_time 1.0029 (1.0048) loss 0.8868 (0.8669) grad_norm 8.9490 (8.7649/2.1497) mem 68106MB [2022-12-19 21:17:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1170/1519] eta 0:05:51 lr 0.000023 time 0.9409 (1.0059) model_time 0.9407 (1.0049) loss 1.0019 (0.8667) grad_norm 7.1473 (8.7774/2.1525) mem 68106MB [2022-12-19 21:17:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1180/1519] eta 0:05:41 lr 0.000023 time 0.9351 (1.0060) model_time 0.9349 (1.0049) loss 1.0551 (0.8670) grad_norm 7.3465 (8.7644/2.1536) mem 68106MB [2022-12-19 21:17:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1190/1519] eta 0:05:30 lr 0.000023 time 0.9327 (1.0061) model_time 0.9326 (1.0050) loss 1.2773 (0.8673) grad_norm 10.0254 (8.7484/2.1443) mem 68106MB [2022-12-19 21:17:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1200/1519] eta 0:05:20 lr 0.000023 time 0.9312 (1.0060) model_time 0.9311 (1.0050) loss 0.6857 (0.8669) grad_norm 8.7537 (8.7326/2.1495) mem 68106MB [2022-12-19 21:18:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1210/1519] eta 0:05:10 lr 0.000023 time 0.9322 (1.0060) model_time 0.9320 (1.0049) loss 0.9820 (0.8671) grad_norm 10.4370 (8.7344/2.1591) mem 68106MB [2022-12-19 21:18:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1220/1519] eta 0:05:00 lr 0.000023 time 0.9472 (1.0061) model_time 0.9470 (1.0050) loss 0.6815 (0.8664) grad_norm 9.0496 (8.7630/2.2023) mem 68106MB [2022-12-19 21:18:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1230/1519] eta 0:04:50 lr 0.000023 time 1.1725 (1.0063) model_time 1.1723 (1.0052) loss 0.9394 (0.8661) grad_norm 8.6142 (8.7500/2.1973) mem 68106MB [2022-12-19 21:18:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1240/1519] eta 0:04:40 lr 0.000023 time 0.9320 (1.0062) model_time 0.9318 (1.0052) loss 0.7897 (0.8657) grad_norm 11.1889 (8.7679/2.1878) mem 68106MB [2022-12-19 21:18:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1250/1519] eta 0:04:30 lr 0.000023 time 0.9255 (1.0061) model_time 0.9247 (1.0051) loss 0.7053 (0.8657) grad_norm 8.6191 (8.7387/2.1594) mem 68106MB [2022-12-19 21:18:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1260/1519] eta 0:04:20 lr 0.000023 time 0.9302 (1.0061) model_time 0.9301 (1.0051) loss 1.1877 (0.8657) grad_norm 8.5990 (8.7344/2.1498) mem 68106MB [2022-12-19 21:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1270/1519] eta 0:04:10 lr 0.000023 time 0.9312 (1.0061) model_time 0.9311 (1.0051) loss 0.6988 (0.8658) grad_norm 9.7598 (8.6900/2.1065) mem 68106MB [2022-12-19 21:19:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1280/1519] eta 0:04:00 lr 0.000023 time 0.9169 (1.0061) model_time 0.9167 (1.0051) loss 0.6789 (0.8662) grad_norm 8.0303 (8.6897/2.1144) mem 68106MB [2022-12-19 21:19:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1290/1519] eta 0:03:50 lr 0.000023 time 0.9571 (1.0061) model_time 0.9570 (1.0051) loss 0.8523 (0.8656) grad_norm 5.7438 (8.6977/2.1179) mem 68106MB [2022-12-19 21:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1300/1519] eta 0:03:40 lr 0.000023 time 0.9341 (1.0061) model_time 0.9340 (1.0051) loss 0.8484 (0.8658) grad_norm 9.0900 (8.6763/2.1196) mem 68106MB [2022-12-19 21:19:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1310/1519] eta 0:03:30 lr 0.000023 time 0.9282 (1.0060) model_time 0.9280 (1.0050) loss 0.7194 (0.8658) grad_norm 6.9129 (8.6767/2.1407) mem 68106MB [2022-12-19 21:19:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1320/1519] eta 0:03:20 lr 0.000023 time 0.9971 (1.0061) model_time 0.9969 (1.0051) loss 0.7498 (0.8654) grad_norm 6.1091 (8.6401/2.1343) mem 68106MB [2022-12-19 21:20:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1330/1519] eta 0:03:10 lr 0.000023 time 0.9346 (1.0061) model_time 0.9345 (1.0051) loss 0.9333 (0.8652) grad_norm 6.5542 (8.6446/2.1334) mem 68106MB [2022-12-19 21:20:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1340/1519] eta 0:03:00 lr 0.000023 time 1.0471 (1.0062) model_time 1.0470 (1.0052) loss 0.7966 (0.8650) grad_norm 7.4338 (8.6397/2.1334) mem 68106MB [2022-12-19 21:20:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1350/1519] eta 0:02:50 lr 0.000023 time 0.9385 (1.0061) model_time 0.9383 (1.0052) loss 0.9003 (0.8656) grad_norm 8.4320 (8.6606/2.1368) mem 68106MB [2022-12-19 21:20:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1360/1519] eta 0:02:39 lr 0.000023 time 0.9238 (1.0061) model_time 0.9237 (1.0052) loss 0.9982 (0.8653) grad_norm 7.3869 (8.6365/2.1352) mem 68106MB [2022-12-19 21:20:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1370/1519] eta 0:02:29 lr 0.000023 time 0.9373 (1.0061) model_time 0.9371 (1.0052) loss 0.6954 (0.8654) grad_norm 8.2722 (8.6446/2.1635) mem 68106MB [2022-12-19 21:20:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1380/1519] eta 0:02:19 lr 0.000023 time 0.9283 (1.0061) model_time 0.9281 (1.0051) loss 0.7523 (0.8660) grad_norm 7.5511 (8.6500/2.1647) mem 68106MB [2022-12-19 21:21:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1390/1519] eta 0:02:09 lr 0.000023 time 0.9438 (1.0060) model_time 0.9436 (1.0051) loss 1.0271 (0.8661) grad_norm 8.0890 (8.6497/2.1618) mem 68106MB [2022-12-19 21:21:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1400/1519] eta 0:01:59 lr 0.000023 time 0.9271 (1.0060) model_time 0.9269 (1.0051) loss 1.0174 (0.8661) grad_norm 12.2930 (8.6666/2.1749) mem 68106MB [2022-12-19 21:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1410/1519] eta 0:01:49 lr 0.000023 time 0.9296 (1.0060) model_time 0.9294 (1.0050) loss 0.7823 (0.8663) grad_norm 9.0012 (8.6532/2.1735) mem 68106MB [2022-12-19 21:21:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1420/1519] eta 0:01:39 lr 0.000023 time 0.9292 (1.0059) model_time 0.9290 (1.0050) loss 0.7958 (0.8664) grad_norm 8.4910 (8.6375/2.1552) mem 68106MB [2022-12-19 21:21:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1430/1519] eta 0:01:29 lr 0.000023 time 0.9364 (1.0059) model_time 0.9363 (1.0049) loss 1.0503 (0.8668) grad_norm 8.1007 (8.6544/2.1610) mem 68106MB [2022-12-19 21:21:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1440/1519] eta 0:01:19 lr 0.000023 time 0.9319 (1.0058) model_time 0.9317 (1.0049) loss 0.9476 (0.8665) grad_norm 8.7084 (8.6750/2.1698) mem 68106MB [2022-12-19 21:22:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1450/1519] eta 0:01:09 lr 0.000023 time 0.9278 (1.0058) model_time 0.9276 (1.0049) loss 1.0816 (0.8666) grad_norm 9.9349 (8.6675/2.1524) mem 68106MB [2022-12-19 21:22:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1460/1519] eta 0:00:59 lr 0.000023 time 0.9315 (1.0058) model_time 0.9313 (1.0048) loss 1.2777 (0.8672) grad_norm 9.0584 (8.6424/2.1269) mem 68106MB [2022-12-19 21:22:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1470/1519] eta 0:00:49 lr 0.000023 time 0.9329 (1.0058) model_time 0.9327 (1.0048) loss 0.9006 (0.8675) grad_norm 6.9335 (8.6052/2.1178) mem 68106MB [2022-12-19 21:22:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1480/1519] eta 0:00:39 lr 0.000023 time 0.9310 (1.0057) model_time 0.9308 (1.0048) loss 0.8137 (0.8677) grad_norm 9.3734 (8.6336/2.1112) mem 68106MB [2022-12-19 21:22:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1490/1519] eta 0:00:29 lr 0.000023 time 0.9979 (1.0058) model_time 0.9977 (1.0049) loss 0.6816 (0.8676) grad_norm 10.0833 (8.6961/2.1800) mem 68106MB [2022-12-19 21:22:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1500/1519] eta 0:00:19 lr 0.000023 time 0.9362 (1.0058) model_time 0.9360 (1.0049) loss 1.4967 (0.8684) grad_norm 7.0633 (8.7051/2.1828) mem 68106MB [2022-12-19 21:23:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [38/100][1510/1519] eta 0:00:09 lr 0.000023 time 0.9221 (1.0057) model_time 0.9219 (1.0048) loss 1.1449 (0.8687) grad_norm 8.3332 (8.7179/2.1827) mem 68106MB [2022-12-19 21:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 38 training takes 0:25:27 [2022-12-19 21:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_38.pth saving...... [2022-12-19 21:23:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_38.pth saved !!! [2022-12-19 21:23:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.657 (0.657) Loss 0.5156 (0.5156) Acc@1 90.625 (90.625) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 21:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.330) Loss 0.5048 (0.4790) Acc@1 92.708 (92.172) Acc@5 98.264 (98.611) Mem 68106MB [2022-12-19 21:23:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4309 (0.4800) Acc@1 91.667 (91.882) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-19 21:23:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.293 (0.308) Loss 0.5776 (0.4843) Acc@1 89.583 (91.756) Acc@5 97.917 (98.387) Mem 68106MB [2022-12-19 21:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.302 (0.306) Loss 0.4560 (0.4779) Acc@1 91.319 (91.751) Acc@5 98.958 (98.442) Mem 68106MB [2022-12-19 21:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 0.4801 (0.4771) Acc@1 89.931 (91.728) Acc@5 98.958 (98.448) Mem 68106MB [2022-12-19 21:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.304) Loss 0.4937 (0.4775) Acc@1 91.319 (91.689) Acc@5 97.917 (98.446) Mem 68106MB [2022-12-19 21:24:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5217 (0.4792) Acc@1 91.319 (91.632) Acc@5 97.917 (98.425) Mem 68106MB [2022-12-19 21:24:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.3989 (0.4767) Acc@1 92.361 (91.637) Acc@5 98.611 (98.461) Mem 68106MB [2022-12-19 21:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:38] * Acc@1 91.646 Acc@5 98.461 [2022-12-19 21:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.6% [2022-12-19 21:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.66% [2022-12-19 21:24:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][0/1519] eta 0:46:21 lr 0.000023 time 1.8314 (1.8314) model_time 1.1497 (1.1497) loss 0.8709 (0.8709) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 21:24:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][10/1519] eta 0:27:22 lr 0.000023 time 0.9890 (1.0883) model_time 0.9889 (1.0260) loss 0.9000 (0.8233) grad_norm 14.3320 (9.8587/2.3136) mem 68106MB [2022-12-19 21:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][20/1519] eta 0:26:27 lr 0.000023 time 0.9157 (1.0589) model_time 0.9155 (1.0260) loss 0.6769 (0.8145) grad_norm 10.5626 (9.1172/2.0824) mem 68106MB [2022-12-19 21:24:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][30/1519] eta 0:25:47 lr 0.000023 time 0.9217 (1.0393) model_time 0.9214 (1.0169) loss 0.6835 (0.8348) grad_norm 6.9880 (8.7392/2.0032) mem 68106MB [2022-12-19 21:24:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][40/1519] eta 0:25:22 lr 0.000023 time 0.9321 (1.0296) model_time 0.9318 (1.0125) loss 0.7998 (0.8323) grad_norm 5.4874 (8.3417/1.9638) mem 68106MB [2022-12-19 21:24:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][50/1519] eta 0:25:04 lr 0.000023 time 0.9180 (1.0239) model_time 0.9179 (1.0101) loss 0.6656 (0.8419) grad_norm 8.5511 (8.0658/1.9435) mem 68106MB [2022-12-19 21:25:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][60/1519] eta 0:24:49 lr 0.000023 time 0.9205 (1.0212) model_time 0.9203 (1.0096) loss 0.9401 (0.8387) grad_norm 8.3218 (8.0637/1.8266) mem 68106MB [2022-12-19 21:25:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][70/1519] eta 0:24:35 lr 0.000023 time 0.9324 (1.0180) model_time 0.9322 (1.0079) loss 0.7343 (0.8377) grad_norm 9.8771 (8.2720/1.9457) mem 68106MB [2022-12-19 21:25:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][80/1519] eta 0:24:23 lr 0.000023 time 0.9731 (1.0169) model_time 0.9730 (1.0080) loss 1.2163 (0.8443) grad_norm 9.7101 (8.3082/1.8531) mem 68106MB [2022-12-19 21:25:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][90/1519] eta 0:24:10 lr 0.000023 time 0.9345 (1.0153) model_time 0.9344 (1.0073) loss 1.0098 (0.8529) grad_norm 8.8193 (8.3509/1.8311) mem 68106MB [2022-12-19 21:25:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][100/1519] eta 0:23:58 lr 0.000023 time 0.9356 (1.0137) model_time 0.9355 (1.0065) loss 0.7221 (0.8548) grad_norm 8.0219 (8.2881/1.7650) mem 68106MB [2022-12-19 21:25:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][110/1519] eta 0:23:46 lr 0.000023 time 0.9401 (1.0125) model_time 0.9399 (1.0059) loss 1.0314 (0.8567) grad_norm 6.8457 (8.2917/1.7220) mem 68106MB [2022-12-19 21:26:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][120/1519] eta 0:23:35 lr 0.000023 time 0.9299 (1.0121) model_time 0.9297 (1.0060) loss 1.0797 (0.8550) grad_norm 9.8935 (8.4568/1.8790) mem 68106MB [2022-12-19 21:26:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][130/1519] eta 0:23:24 lr 0.000023 time 0.9320 (1.0110) model_time 0.9318 (1.0054) loss 0.7203 (0.8490) grad_norm 9.0464 (8.4724/1.8168) mem 68106MB [2022-12-19 21:26:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][140/1519] eta 0:23:13 lr 0.000023 time 0.9280 (1.0105) model_time 0.9277 (1.0053) loss 0.6820 (0.8502) grad_norm 11.9379 (8.5607/1.8076) mem 68106MB [2022-12-19 21:26:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][150/1519] eta 0:23:02 lr 0.000023 time 0.9402 (1.0099) model_time 0.9400 (1.0050) loss 0.9833 (0.8565) grad_norm 11.8433 (8.6748/1.8293) mem 68106MB [2022-12-19 21:26:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][160/1519] eta 0:22:51 lr 0.000023 time 0.9234 (1.0094) model_time 0.9232 (1.0047) loss 0.6726 (0.8549) grad_norm 7.7290 (8.5845/1.8095) mem 68106MB [2022-12-19 21:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][170/1519] eta 0:22:42 lr 0.000023 time 0.9286 (1.0099) model_time 0.9284 (1.0055) loss 0.7270 (0.8583) grad_norm 9.0822 (8.5225/1.7879) mem 68106MB [2022-12-19 21:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][180/1519] eta 0:22:31 lr 0.000023 time 0.9277 (1.0094) model_time 0.9275 (1.0052) loss 0.7568 (0.8573) grad_norm 7.5537 (8.6764/2.1200) mem 68106MB [2022-12-19 21:27:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][190/1519] eta 0:22:21 lr 0.000023 time 0.9389 (1.0092) model_time 0.9387 (1.0052) loss 0.9085 (0.8582) grad_norm 10.2397 (8.6216/2.1012) mem 68106MB [2022-12-19 21:27:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][200/1519] eta 0:22:10 lr 0.000023 time 0.9225 (1.0087) model_time 0.9224 (1.0049) loss 0.7771 (0.8567) grad_norm 7.3649 (8.6289/2.0878) mem 68106MB [2022-12-19 21:27:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][210/1519] eta 0:21:59 lr 0.000023 time 0.9330 (1.0084) model_time 0.9328 (1.0047) loss 0.6814 (0.8524) grad_norm 7.8021 (8.6443/2.0469) mem 68106MB [2022-12-19 21:27:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][220/1519] eta 0:21:49 lr 0.000023 time 0.9375 (1.0081) model_time 0.9374 (1.0046) loss 0.7390 (0.8519) grad_norm 8.2804 (8.6861/2.0265) mem 68106MB [2022-12-19 21:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][230/1519] eta 0:21:38 lr 0.000023 time 0.9362 (1.0077) model_time 0.9361 (1.0043) loss 1.1186 (0.8539) grad_norm 7.3362 (8.6445/2.0011) mem 68106MB [2022-12-19 21:28:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][240/1519] eta 0:21:29 lr 0.000023 time 0.9289 (1.0080) model_time 0.9288 (1.0048) loss 0.7799 (0.8546) grad_norm 7.0088 (8.5782/1.9849) mem 68106MB [2022-12-19 21:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][250/1519] eta 0:21:18 lr 0.000023 time 0.9331 (1.0077) model_time 0.9330 (1.0046) loss 0.7723 (0.8541) grad_norm 7.5293 (8.5502/1.9558) mem 68106MB [2022-12-19 21:28:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][260/1519] eta 0:21:08 lr 0.000023 time 0.9784 (1.0079) model_time 0.9783 (1.0049) loss 1.0034 (0.8529) grad_norm 6.7493 (8.4990/1.9385) mem 68106MB [2022-12-19 21:28:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][270/1519] eta 0:20:59 lr 0.000023 time 0.9292 (1.0081) model_time 0.9291 (1.0051) loss 0.7627 (0.8531) grad_norm 6.3917 (8.4501/1.9224) mem 68106MB [2022-12-19 21:28:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][280/1519] eta 0:20:48 lr 0.000023 time 0.9226 (1.0079) model_time 0.9224 (1.0051) loss 0.8027 (0.8529) grad_norm 8.3408 (8.4265/1.9090) mem 68106MB [2022-12-19 21:28:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][290/1519] eta 0:20:38 lr 0.000023 time 0.9282 (1.0078) model_time 0.9280 (1.0050) loss 1.3982 (0.8576) grad_norm 7.3538 (8.4451/1.9033) mem 68106MB [2022-12-19 21:29:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][300/1519] eta 0:20:28 lr 0.000023 time 0.9232 (1.0080) model_time 0.9230 (1.0053) loss 0.9195 (0.8566) grad_norm 6.0778 (8.4267/1.8917) mem 68106MB [2022-12-19 21:29:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][310/1519] eta 0:20:18 lr 0.000023 time 0.9213 (1.0077) model_time 0.9212 (1.0051) loss 0.7094 (0.8581) grad_norm 7.7453 (8.4650/1.8883) mem 68106MB [2022-12-19 21:29:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][320/1519] eta 0:20:09 lr 0.000023 time 0.9310 (1.0088) model_time 0.9309 (1.0063) loss 1.0449 (0.8575) grad_norm 7.3375 (8.4484/1.8675) mem 68106MB [2022-12-19 21:29:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][330/1519] eta 0:19:59 lr 0.000023 time 0.9033 (1.0092) model_time 0.9032 (1.0067) loss 0.7427 (0.8570) grad_norm 9.8875 (8.4700/1.8600) mem 68106MB [2022-12-19 21:29:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][340/1519] eta 0:19:50 lr 0.000023 time 0.9250 (1.0098) model_time 0.9249 (1.0074) loss 0.7137 (0.8571) grad_norm 9.7930 (8.4455/1.8476) mem 68106MB [2022-12-19 21:29:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][350/1519] eta 0:19:39 lr 0.000023 time 0.9258 (1.0093) model_time 0.9257 (1.0069) loss 0.6743 (0.8566) grad_norm 7.5703 (8.4606/1.8354) mem 68106MB [2022-12-19 21:30:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][360/1519] eta 0:19:29 lr 0.000023 time 0.9688 (1.0091) model_time 0.9687 (1.0068) loss 0.7622 (0.8544) grad_norm 8.7961 (8.4624/1.8218) mem 68106MB [2022-12-19 21:30:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][370/1519] eta 0:19:19 lr 0.000023 time 0.9232 (1.0089) model_time 0.9230 (1.0067) loss 0.7176 (0.8535) grad_norm 8.3694 (8.4191/1.8278) mem 68106MB [2022-12-19 21:30:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][380/1519] eta 0:19:08 lr 0.000023 time 0.9301 (1.0088) model_time 0.9300 (1.0066) loss 0.9750 (0.8539) grad_norm 8.3936 (8.3938/1.8165) mem 68106MB [2022-12-19 21:30:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][390/1519] eta 0:18:58 lr 0.000023 time 0.9254 (1.0085) model_time 0.9253 (1.0064) loss 0.7232 (0.8551) grad_norm 7.6523 (8.3888/1.7993) mem 68106MB [2022-12-19 21:30:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][400/1519] eta 0:18:48 lr 0.000023 time 0.9201 (1.0084) model_time 0.9200 (1.0063) loss 0.9384 (0.8547) grad_norm 7.0701 (8.4108/1.8046) mem 68106MB [2022-12-19 21:30:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][410/1519] eta 0:18:37 lr 0.000023 time 0.9289 (1.0081) model_time 0.9287 (1.0060) loss 0.9828 (0.8581) grad_norm 7.7750 (8.4050/1.7890) mem 68106MB [2022-12-19 21:31:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][420/1519] eta 0:18:27 lr 0.000023 time 0.9348 (1.0079) model_time 0.9347 (1.0059) loss 0.7053 (0.8553) grad_norm 6.6018 (8.3747/1.7797) mem 68106MB [2022-12-19 21:31:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][430/1519] eta 0:18:17 lr 0.000023 time 0.9298 (1.0077) model_time 0.9297 (1.0057) loss 0.7132 (0.8536) grad_norm 10.7545 (8.3993/1.7939) mem 68106MB [2022-12-19 21:31:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][440/1519] eta 0:18:07 lr 0.000023 time 0.9192 (1.0075) model_time 0.9191 (1.0055) loss 0.9393 (0.8537) grad_norm 10.1751 (8.4094/1.7790) mem 68106MB [2022-12-19 21:31:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][450/1519] eta 0:17:56 lr 0.000023 time 0.9241 (1.0074) model_time 0.9240 (1.0055) loss 0.8500 (0.8545) grad_norm 7.0123 (8.3998/1.7738) mem 68106MB [2022-12-19 21:31:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][460/1519] eta 0:17:46 lr 0.000023 time 0.9308 (1.0072) model_time 0.9306 (1.0054) loss 0.8670 (0.8534) grad_norm 7.2475 (8.4069/1.7692) mem 68106MB [2022-12-19 21:31:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][470/1519] eta 0:17:36 lr 0.000023 time 0.9308 (1.0072) model_time 0.9306 (1.0054) loss 1.0085 (0.8548) grad_norm 6.4938 (8.3905/1.7624) mem 68106MB [2022-12-19 21:32:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][480/1519] eta 0:17:26 lr 0.000023 time 0.9326 (1.0074) model_time 0.9324 (1.0056) loss 0.8182 (0.8554) grad_norm 9.7033 (8.3995/1.7516) mem 68106MB [2022-12-19 21:32:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][490/1519] eta 0:17:16 lr 0.000023 time 0.9315 (1.0072) model_time 0.9313 (1.0054) loss 1.1257 (0.8566) grad_norm 9.0530 (8.4116/1.7649) mem 68106MB [2022-12-19 21:32:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][500/1519] eta 0:17:06 lr 0.000023 time 0.9305 (1.0071) model_time 0.9303 (1.0054) loss 0.9348 (0.8557) grad_norm 6.5109 (8.4169/1.7602) mem 68106MB [2022-12-19 21:32:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][510/1519] eta 0:16:56 lr 0.000023 time 0.9235 (1.0071) model_time 0.9233 (1.0054) loss 0.8481 (0.8569) grad_norm 7.1348 (8.4329/1.7650) mem 68106MB [2022-12-19 21:32:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][520/1519] eta 0:16:45 lr 0.000023 time 0.9277 (1.0069) model_time 0.9276 (1.0052) loss 0.7586 (0.8565) grad_norm 7.8892 (8.4225/1.7633) mem 68106MB [2022-12-19 21:32:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][530/1519] eta 0:16:35 lr 0.000023 time 0.9384 (1.0068) model_time 0.9383 (1.0051) loss 0.8213 (0.8564) grad_norm 13.5160 (8.4711/1.7936) mem 68106MB [2022-12-19 21:33:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][540/1519] eta 0:16:25 lr 0.000023 time 0.9764 (1.0067) model_time 0.9761 (1.0051) loss 1.0703 (0.8577) grad_norm 9.1608 (8.4716/1.7883) mem 68106MB [2022-12-19 21:33:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][550/1519] eta 0:16:15 lr 0.000023 time 1.1021 (1.0070) model_time 1.1020 (1.0053) loss 0.9097 (0.8586) grad_norm 8.5144 (8.4644/1.7760) mem 68106MB [2022-12-19 21:33:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][560/1519] eta 0:16:05 lr 0.000023 time 0.9374 (1.0068) model_time 0.9372 (1.0052) loss 0.9747 (0.8588) grad_norm 8.0066 (8.4427/1.7694) mem 68106MB [2022-12-19 21:33:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][570/1519] eta 0:15:55 lr 0.000023 time 0.9243 (1.0068) model_time 0.9242 (1.0052) loss 0.9053 (0.8591) grad_norm 6.0124 (8.4317/1.7722) mem 68106MB [2022-12-19 21:33:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][580/1519] eta 0:15:45 lr 0.000023 time 0.9419 (1.0070) model_time 0.9418 (1.0054) loss 0.6895 (0.8600) grad_norm 6.5119 (8.4183/1.7639) mem 68106MB [2022-12-19 21:33:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][590/1519] eta 0:15:35 lr 0.000023 time 0.9348 (1.0068) model_time 0.9346 (1.0053) loss 1.0266 (0.8601) grad_norm 9.0582 (8.4281/1.7612) mem 68106MB [2022-12-19 21:34:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][600/1519] eta 0:15:25 lr 0.000023 time 0.9320 (1.0067) model_time 0.9319 (1.0052) loss 0.7089 (0.8608) grad_norm 8.9586 (8.4425/1.7506) mem 68106MB [2022-12-19 21:34:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][610/1519] eta 0:15:15 lr 0.000023 time 0.9230 (1.0068) model_time 0.9229 (1.0053) loss 1.0498 (0.8606) grad_norm 9.5079 (8.4314/1.7688) mem 68106MB [2022-12-19 21:34:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][620/1519] eta 0:15:05 lr 0.000023 time 0.9408 (1.0067) model_time 0.9407 (1.0052) loss 0.9887 (0.8617) grad_norm 6.2213 (8.3962/1.7403) mem 68106MB [2022-12-19 21:34:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][630/1519] eta 0:14:55 lr 0.000022 time 1.0177 (1.0070) model_time 1.0173 (1.0056) loss 1.1244 (0.8611) grad_norm 7.2965 (8.3723/1.7328) mem 68106MB [2022-12-19 21:34:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][640/1519] eta 0:14:45 lr 0.000022 time 0.9345 (1.0070) model_time 0.9344 (1.0055) loss 1.0482 (0.8609) grad_norm 12.6144 (8.3930/1.7447) mem 68106MB [2022-12-19 21:35:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][650/1519] eta 0:14:35 lr 0.000022 time 0.8887 (1.0071) model_time 0.8884 (1.0057) loss 0.8833 (0.8614) grad_norm 13.7617 (8.4424/1.7540) mem 68106MB [2022-12-19 21:35:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][660/1519] eta 0:14:25 lr 0.000022 time 0.9347 (1.0071) model_time 0.9346 (1.0056) loss 0.9850 (0.8618) grad_norm 7.1059 (8.4379/1.7555) mem 68106MB [2022-12-19 21:35:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][670/1519] eta 0:14:14 lr 0.000022 time 0.9310 (1.0069) model_time 0.9308 (1.0055) loss 0.7747 (0.8610) grad_norm 6.3649 (8.4223/1.7458) mem 68106MB [2022-12-19 21:35:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][680/1519] eta 0:14:04 lr 0.000022 time 0.9287 (1.0068) model_time 0.9285 (1.0055) loss 0.7692 (0.8609) grad_norm 11.4184 (8.4232/1.7522) mem 68106MB [2022-12-19 21:35:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][690/1519] eta 0:13:54 lr 0.000022 time 0.9335 (1.0067) model_time 0.9334 (1.0054) loss 0.7607 (0.8612) grad_norm 12.4089 (8.4294/1.7690) mem 68106MB [2022-12-19 21:35:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][700/1519] eta 0:13:44 lr 0.000022 time 0.9286 (1.0067) model_time 0.9284 (1.0053) loss 0.8159 (0.8622) grad_norm 5.8087 (8.4335/1.7786) mem 68106MB [2022-12-19 21:36:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][710/1519] eta 0:13:34 lr 0.000022 time 0.9300 (1.0067) model_time 0.9298 (1.0053) loss 0.8823 (0.8619) grad_norm 10.0775 (8.4528/1.7849) mem 68106MB [2022-12-19 21:36:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][720/1519] eta 0:13:24 lr 0.000022 time 0.9247 (1.0066) model_time 0.9245 (1.0053) loss 0.7003 (0.8624) grad_norm 7.5583 (8.4426/1.7487) mem 68106MB [2022-12-19 21:36:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][730/1519] eta 0:13:14 lr 0.000022 time 1.0675 (1.0067) model_time 1.0674 (1.0054) loss 1.1517 (0.8628) grad_norm 8.3210 (8.4321/1.7504) mem 68106MB [2022-12-19 21:36:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][740/1519] eta 0:13:04 lr 0.000022 time 0.9245 (1.0066) model_time 0.9243 (1.0053) loss 0.7928 (0.8624) grad_norm 5.8808 (8.3929/1.7708) mem 68106MB [2022-12-19 21:36:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][750/1519] eta 0:12:54 lr 0.000022 time 0.9320 (1.0065) model_time 0.9319 (1.0052) loss 1.2925 (0.8641) grad_norm 7.0060 (8.3553/1.7560) mem 68106MB [2022-12-19 21:36:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][760/1519] eta 0:12:43 lr 0.000022 time 0.9138 (1.0065) model_time 0.9136 (1.0052) loss 0.9795 (0.8639) grad_norm 7.3081 (8.3563/1.7421) mem 68106MB [2022-12-19 21:37:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][770/1519] eta 0:12:33 lr 0.000022 time 0.9313 (1.0065) model_time 0.9312 (1.0052) loss 0.9594 (0.8637) grad_norm 9.3817 (8.3783/1.7413) mem 68106MB [2022-12-19 21:37:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][780/1519] eta 0:12:23 lr 0.000022 time 0.9310 (1.0065) model_time 0.9309 (1.0052) loss 0.7439 (0.8632) grad_norm 10.3280 (8.3429/1.6176) mem 68106MB [2022-12-19 21:37:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][790/1519] eta 0:12:13 lr 0.000022 time 0.9220 (1.0065) model_time 0.9219 (1.0052) loss 0.7304 (0.8636) grad_norm 8.7256 (8.3608/1.6239) mem 68106MB [2022-12-19 21:37:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][800/1519] eta 0:12:03 lr 0.000022 time 0.9295 (1.0064) model_time 0.9293 (1.0052) loss 0.8349 (0.8629) grad_norm 7.6219 (8.3277/1.6094) mem 68106MB [2022-12-19 21:37:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][810/1519] eta 0:11:53 lr 0.000022 time 0.9651 (1.0068) model_time 0.9650 (1.0055) loss 0.8270 (0.8634) grad_norm 7.9284 (8.3169/1.6163) mem 68106MB [2022-12-19 21:37:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][820/1519] eta 0:11:43 lr 0.000022 time 0.9206 (1.0068) model_time 0.9205 (1.0055) loss 0.8146 (0.8629) grad_norm 9.1882 (8.2867/1.6309) mem 68106MB [2022-12-19 21:38:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][830/1519] eta 0:11:33 lr 0.000022 time 0.9236 (1.0068) model_time 0.9234 (1.0055) loss 0.9354 (0.8629) grad_norm 9.6368 (8.2819/1.6391) mem 68106MB [2022-12-19 21:38:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][840/1519] eta 0:11:23 lr 0.000022 time 0.9288 (1.0066) model_time 0.9287 (1.0054) loss 0.7907 (0.8632) grad_norm 7.0006 (8.2896/1.6385) mem 68106MB [2022-12-19 21:38:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][850/1519] eta 0:11:13 lr 0.000022 time 0.9287 (1.0066) model_time 0.9285 (1.0054) loss 0.8300 (0.8636) grad_norm 8.4032 (8.3154/1.6457) mem 68106MB [2022-12-19 21:38:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][860/1519] eta 0:11:03 lr 0.000022 time 0.9696 (1.0066) model_time 0.9695 (1.0054) loss 1.1142 (0.8641) grad_norm 5.9280 (8.3170/1.6468) mem 68106MB [2022-12-19 21:38:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][870/1519] eta 0:10:53 lr 0.000022 time 0.9220 (1.0065) model_time 0.9219 (1.0053) loss 0.8856 (0.8638) grad_norm 5.5372 (8.3259/1.6518) mem 68106MB [2022-12-19 21:38:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][880/1519] eta 0:10:43 lr 0.000022 time 0.9164 (1.0064) model_time 0.9161 (1.0052) loss 0.6988 (0.8630) grad_norm 11.7339 (8.3546/1.6523) mem 68106MB [2022-12-19 21:39:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][890/1519] eta 0:10:33 lr 0.000022 time 0.9872 (1.0068) model_time 0.9870 (1.0056) loss 0.8435 (0.8631) grad_norm 7.3947 (8.3444/1.6519) mem 68106MB [2022-12-19 21:39:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][900/1519] eta 0:10:23 lr 0.000022 time 0.9311 (1.0067) model_time 0.9309 (1.0055) loss 0.8241 (0.8634) grad_norm 8.8230 (8.3518/1.6462) mem 68106MB [2022-12-19 21:39:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][910/1519] eta 0:10:13 lr 0.000022 time 0.9189 (1.0066) model_time 0.9188 (1.0055) loss 0.8755 (0.8636) grad_norm 8.0679 (8.3206/1.6466) mem 68106MB [2022-12-19 21:39:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][920/1519] eta 0:10:02 lr 0.000022 time 0.9330 (1.0067) model_time 0.9329 (1.0055) loss 0.8948 (0.8631) grad_norm 7.4955 (8.3162/1.6477) mem 68106MB [2022-12-19 21:39:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][930/1519] eta 0:09:52 lr 0.000022 time 0.9258 (1.0066) model_time 0.9257 (1.0055) loss 1.2672 (0.8645) grad_norm 9.2332 (8.3105/1.6433) mem 68106MB [2022-12-19 21:39:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][940/1519] eta 0:09:42 lr 0.000022 time 1.2141 (1.0068) model_time 1.2140 (1.0057) loss 0.9569 (0.8645) grad_norm 6.7662 (8.3396/1.6784) mem 68106MB [2022-12-19 21:40:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][950/1519] eta 0:09:32 lr 0.000022 time 0.9333 (1.0068) model_time 0.9332 (1.0057) loss 1.1662 (0.8647) grad_norm 7.4885 (8.3185/1.6825) mem 68106MB [2022-12-19 21:40:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][960/1519] eta 0:09:22 lr 0.000022 time 0.9297 (1.0068) model_time 0.9295 (1.0057) loss 0.8902 (0.8647) grad_norm 10.2643 (8.3243/1.6820) mem 68106MB [2022-12-19 21:40:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][970/1519] eta 0:09:12 lr 0.000022 time 0.9631 (1.0069) model_time 0.9630 (1.0058) loss 0.7639 (0.8640) grad_norm 7.8216 (8.3827/1.7819) mem 68106MB [2022-12-19 21:40:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][980/1519] eta 0:09:02 lr 0.000022 time 0.9199 (1.0069) model_time 0.9197 (1.0058) loss 0.7453 (0.8640) grad_norm 9.4026 (8.4009/1.7781) mem 68106MB [2022-12-19 21:40:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][990/1519] eta 0:08:52 lr 0.000022 time 0.9934 (1.0068) model_time 0.9931 (1.0058) loss 0.7274 (0.8645) grad_norm 7.3644 (8.4064/1.8016) mem 68106MB [2022-12-19 21:40:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1000/1519] eta 0:08:42 lr 0.000022 time 0.9223 (1.0068) model_time 0.9222 (1.0057) loss 1.3434 (0.8650) grad_norm 10.7441 (8.4166/1.8041) mem 68106MB [2022-12-19 21:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1010/1519] eta 0:08:32 lr 0.000022 time 0.9211 (1.0067) model_time 0.9209 (1.0057) loss 0.9864 (0.8655) grad_norm 7.7964 (8.4261/1.8076) mem 68106MB [2022-12-19 21:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1020/1519] eta 0:08:22 lr 0.000022 time 0.9225 (1.0067) model_time 0.9222 (1.0056) loss 0.6922 (0.8656) grad_norm 13.3937 (8.4600/1.8252) mem 68106MB [2022-12-19 21:41:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1030/1519] eta 0:08:12 lr 0.000022 time 0.9197 (1.0066) model_time 0.9196 (1.0056) loss 0.6823 (0.8660) grad_norm 12.1064 (8.4650/1.8379) mem 68106MB [2022-12-19 21:41:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1040/1519] eta 0:08:02 lr 0.000022 time 0.9305 (1.0066) model_time 0.9304 (1.0055) loss 0.8107 (0.8658) grad_norm 9.6640 (8.4979/1.9577) mem 68106MB [2022-12-19 21:41:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1050/1519] eta 0:07:52 lr 0.000022 time 0.9366 (1.0066) model_time 0.9365 (1.0056) loss 0.7442 (0.8652) grad_norm 7.1687 (8.5151/1.9616) mem 68106MB [2022-12-19 21:41:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1060/1519] eta 0:07:42 lr 0.000022 time 0.9212 (1.0066) model_time 0.9211 (1.0056) loss 1.1865 (0.8654) grad_norm 7.3158 (8.4972/1.9574) mem 68106MB [2022-12-19 21:42:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1070/1519] eta 0:07:31 lr 0.000022 time 0.9808 (1.0066) model_time 0.9806 (1.0056) loss 0.9823 (0.8647) grad_norm 14.9014 (8.5209/1.9896) mem 68106MB [2022-12-19 21:42:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1080/1519] eta 0:07:21 lr 0.000022 time 0.9686 (1.0065) model_time 0.9684 (1.0055) loss 0.8062 (0.8645) grad_norm 7.7354 (8.5328/2.0030) mem 68106MB [2022-12-19 21:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1090/1519] eta 0:07:11 lr 0.000022 time 0.9837 (1.0065) model_time 0.9836 (1.0055) loss 0.9426 (0.8649) grad_norm 6.0073 (8.5187/2.0033) mem 68106MB [2022-12-19 21:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1100/1519] eta 0:07:01 lr 0.000022 time 0.9159 (1.0065) model_time 0.9158 (1.0055) loss 0.8719 (0.8648) grad_norm 7.8361 (8.5145/2.0035) mem 68106MB [2022-12-19 21:42:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1110/1519] eta 0:06:51 lr 0.000022 time 0.9194 (1.0065) model_time 0.9191 (1.0055) loss 0.7114 (0.8644) grad_norm 15.1331 (8.5332/2.0287) mem 68106MB [2022-12-19 21:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1120/1519] eta 0:06:41 lr 0.000022 time 0.9381 (1.0065) model_time 0.9380 (1.0055) loss 0.8521 (0.8646) grad_norm 10.7598 (8.5624/2.0348) mem 68106MB [2022-12-19 21:43:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1130/1519] eta 0:06:31 lr 0.000022 time 1.0097 (1.0067) model_time 1.0096 (1.0057) loss 1.2121 (0.8649) grad_norm 12.1566 (8.5397/2.0183) mem 68106MB [2022-12-19 21:43:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1140/1519] eta 0:06:21 lr 0.000022 time 0.9255 (1.0067) model_time 0.9253 (1.0057) loss 0.9498 (0.8648) grad_norm 8.9277 (8.5374/2.0123) mem 68106MB [2022-12-19 21:43:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1150/1519] eta 0:06:11 lr 0.000022 time 0.9230 (1.0067) model_time 0.9228 (1.0057) loss 0.9038 (0.8650) grad_norm 8.7995 (8.5668/2.0235) mem 68106MB [2022-12-19 21:43:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1160/1519] eta 0:06:01 lr 0.000022 time 0.9261 (1.0066) model_time 0.9260 (1.0057) loss 0.6948 (0.8655) grad_norm 8.3624 (8.5822/2.0224) mem 68106MB [2022-12-19 21:43:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1170/1519] eta 0:05:51 lr 0.000022 time 0.9764 (1.0066) model_time 0.9763 (1.0057) loss 0.8228 (0.8651) grad_norm 6.5340 (8.5918/2.0200) mem 68106MB [2022-12-19 21:43:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1180/1519] eta 0:05:41 lr 0.000022 time 0.9460 (1.0066) model_time 0.9459 (1.0057) loss 1.1233 (0.8655) grad_norm 5.8336 (8.5877/2.0277) mem 68106MB [2022-12-19 21:44:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1190/1519] eta 0:05:31 lr 0.000022 time 0.9643 (1.0066) model_time 0.9642 (1.0056) loss 1.0189 (0.8658) grad_norm 10.8091 (8.5856/2.0289) mem 68106MB [2022-12-19 21:44:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1200/1519] eta 0:05:21 lr 0.000022 time 0.9232 (1.0066) model_time 0.9230 (1.0056) loss 0.8238 (0.8664) grad_norm 11.0542 (8.5961/2.0485) mem 68106MB [2022-12-19 21:44:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1210/1519] eta 0:05:11 lr 0.000022 time 0.9276 (1.0065) model_time 0.9275 (1.0056) loss 0.7486 (0.8662) grad_norm 7.1061 (8.5946/2.0365) mem 68106MB [2022-12-19 21:44:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1220/1519] eta 0:05:00 lr 0.000022 time 0.9359 (1.0065) model_time 0.9358 (1.0056) loss 0.8048 (0.8656) grad_norm 7.1251 (8.6206/2.0511) mem 68106MB [2022-12-19 21:44:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1230/1519] eta 0:04:50 lr 0.000022 time 1.0159 (1.0065) model_time 1.0158 (1.0056) loss 0.9094 (0.8658) grad_norm 8.4273 (8.6391/2.0519) mem 68106MB [2022-12-19 21:44:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1240/1519] eta 0:04:40 lr 0.000022 time 0.9212 (1.0064) model_time 0.9211 (1.0055) loss 0.6744 (0.8658) grad_norm 6.6618 (8.6320/2.0425) mem 68106MB [2022-12-19 21:45:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1250/1519] eta 0:04:30 lr 0.000022 time 0.9207 (1.0064) model_time 0.9206 (1.0055) loss 0.7200 (0.8656) grad_norm 6.7031 (8.6248/2.0264) mem 68106MB [2022-12-19 21:45:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1260/1519] eta 0:04:20 lr 0.000022 time 0.9315 (1.0064) model_time 0.9313 (1.0055) loss 0.9376 (0.8656) grad_norm 7.1579 (8.6099/2.0317) mem 68106MB [2022-12-19 21:45:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1270/1519] eta 0:04:10 lr 0.000022 time 0.9290 (1.0063) model_time 0.9289 (1.0054) loss 1.1089 (0.8655) grad_norm 7.6560 (8.6287/2.0364) mem 68106MB [2022-12-19 21:45:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1280/1519] eta 0:04:00 lr 0.000022 time 0.9814 (1.0063) model_time 0.9813 (1.0054) loss 0.7446 (0.8652) grad_norm 6.9340 (8.6217/2.0337) mem 68106MB [2022-12-19 21:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1290/1519] eta 0:03:50 lr 0.000022 time 0.9366 (1.0063) model_time 0.9365 (1.0054) loss 1.1417 (0.8658) grad_norm 6.0508 (8.5889/2.0236) mem 68106MB [2022-12-19 21:45:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1300/1519] eta 0:03:40 lr 0.000022 time 0.9179 (1.0062) model_time 0.9178 (1.0053) loss 0.8373 (0.8654) grad_norm 9.5097 (8.5918/2.0202) mem 68106MB [2022-12-19 21:46:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1310/1519] eta 0:03:30 lr 0.000022 time 0.9343 (1.0062) model_time 0.9341 (1.0053) loss 0.8489 (0.8654) grad_norm 12.1576 (8.5787/2.0258) mem 68106MB [2022-12-19 21:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1320/1519] eta 0:03:20 lr 0.000022 time 0.9220 (1.0061) model_time 0.9219 (1.0052) loss 0.7321 (0.8653) grad_norm 7.5253 (8.5763/2.0242) mem 68106MB [2022-12-19 21:46:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1330/1519] eta 0:03:10 lr 0.000022 time 0.9255 (1.0061) model_time 0.9254 (1.0052) loss 0.6951 (0.8654) grad_norm 7.3224 (8.5543/2.0312) mem 68106MB [2022-12-19 21:46:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1340/1519] eta 0:03:00 lr 0.000022 time 0.9228 (1.0060) model_time 0.9226 (1.0051) loss 1.3158 (0.8660) grad_norm 8.0246 (8.5798/2.0121) mem 68106MB [2022-12-19 21:46:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1350/1519] eta 0:02:50 lr 0.000022 time 0.9261 (1.0060) model_time 0.9260 (1.0051) loss 0.7541 (0.8664) grad_norm 6.8471 (8.5855/2.0076) mem 68106MB [2022-12-19 21:46:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1360/1519] eta 0:02:39 lr 0.000022 time 0.9318 (1.0060) model_time 0.9316 (1.0051) loss 0.6972 (0.8663) grad_norm 6.6282 (8.5990/2.0265) mem 68106MB [2022-12-19 21:47:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1370/1519] eta 0:02:29 lr 0.000022 time 0.9757 (1.0060) model_time 0.9756 (1.0051) loss 0.7819 (0.8663) grad_norm 8.0928 (8.5937/2.0226) mem 68106MB [2022-12-19 21:47:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1380/1519] eta 0:02:19 lr 0.000022 time 0.9298 (1.0060) model_time 0.9297 (1.0052) loss 1.0572 (0.8665) grad_norm 7.4495 (8.5696/2.0176) mem 68106MB [2022-12-19 21:47:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1390/1519] eta 0:02:09 lr 0.000022 time 0.9224 (1.0061) model_time 0.9222 (1.0052) loss 0.7461 (0.8664) grad_norm 7.3695 (8.5623/2.0104) mem 68106MB [2022-12-19 21:47:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1400/1519] eta 0:01:59 lr 0.000022 time 0.9309 (1.0061) model_time 0.9307 (1.0052) loss 0.8474 (0.8665) grad_norm 6.5979 (8.6028/2.0302) mem 68106MB [2022-12-19 21:47:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1410/1519] eta 0:01:49 lr 0.000022 time 1.0514 (1.0062) model_time 1.0513 (1.0053) loss 0.7319 (0.8665) grad_norm 6.6190 (8.5945/2.0254) mem 68106MB [2022-12-19 21:47:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1420/1519] eta 0:01:39 lr 0.000022 time 0.9194 (1.0061) model_time 0.9192 (1.0053) loss 0.8298 (0.8669) grad_norm 6.2767 (8.5969/2.0080) mem 68106MB [2022-12-19 21:48:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1430/1519] eta 0:01:29 lr 0.000022 time 0.9746 (1.0061) model_time 0.9744 (1.0053) loss 0.9719 (0.8666) grad_norm 7.4078 (8.6241/2.0069) mem 68106MB [2022-12-19 21:48:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1440/1519] eta 0:01:19 lr 0.000022 time 0.9318 (1.0063) model_time 0.9317 (1.0055) loss 1.0005 (0.8669) grad_norm 10.6652 (8.6399/2.0103) mem 68106MB [2022-12-19 21:48:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1450/1519] eta 0:01:09 lr 0.000022 time 0.9960 (1.0066) model_time 0.9959 (1.0058) loss 0.7679 (0.8668) grad_norm 11.7627 (8.6454/2.0186) mem 68106MB [2022-12-19 21:48:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1460/1519] eta 0:00:59 lr 0.000022 time 0.9264 (1.0066) model_time 0.9263 (1.0058) loss 0.8260 (0.8668) grad_norm 10.6383 (8.6871/2.0180) mem 68106MB [2022-12-19 21:48:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1470/1519] eta 0:00:49 lr 0.000022 time 0.9296 (1.0066) model_time 0.9294 (1.0057) loss 0.8779 (0.8667) grad_norm 8.8848 (8.7103/2.0109) mem 68106MB [2022-12-19 21:48:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1480/1519] eta 0:00:39 lr 0.000022 time 0.9212 (1.0065) model_time 0.9210 (1.0057) loss 0.7827 (0.8660) grad_norm 7.4834 (8.7084/2.0156) mem 68106MB [2022-12-19 21:49:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1490/1519] eta 0:00:29 lr 0.000022 time 0.9242 (1.0065) model_time 0.9241 (1.0057) loss 0.8481 (0.8660) grad_norm 8.8676 (8.7084/2.0105) mem 68106MB [2022-12-19 21:49:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1500/1519] eta 0:00:19 lr 0.000022 time 0.9244 (1.0065) model_time 0.9242 (1.0057) loss 0.7835 (0.8661) grad_norm 7.6034 (8.7125/2.0177) mem 68106MB [2022-12-19 21:49:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [39/100][1510/1519] eta 0:00:09 lr 0.000022 time 0.9206 (1.0064) model_time 0.9205 (1.0056) loss 0.8873 (0.8665) grad_norm 6.7271 (8.7194/2.0086) mem 68106MB [2022-12-19 21:49:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 39 training takes 0:25:28 [2022-12-19 21:49:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_39.pth saving...... [2022-12-19 21:49:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_39.pth saved !!! [2022-12-19 21:49:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.5309 (0.5309) Acc@1 91.319 (91.319) Acc@5 97.917 (97.917) Mem 68106MB [2022-12-19 21:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.332) Loss 0.4930 (0.4870) Acc@1 92.014 (91.888) Acc@5 98.264 (98.580) Mem 68106MB [2022-12-19 21:50:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4515 (0.4805) Acc@1 91.319 (91.948) Acc@5 98.958 (98.578) Mem 68106MB [2022-12-19 21:50:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 0.5889 (0.4855) Acc@1 89.931 (91.767) Acc@5 97.917 (98.477) Mem 68106MB [2022-12-19 21:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.303 (0.308) Loss 0.4340 (0.4769) Acc@1 93.403 (91.828) Acc@5 98.958 (98.535) Mem 68106MB [2022-12-19 21:50:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.305 (0.307) Loss 0.4916 (0.4754) Acc@1 89.583 (91.857) Acc@5 99.306 (98.584) Mem 68106MB [2022-12-19 21:50:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.306) Loss 0.4914 (0.4755) Acc@1 92.361 (91.946) Acc@5 98.264 (98.554) Mem 68106MB [2022-12-19 21:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.301 (0.305) Loss 0.5288 (0.4776) Acc@1 91.319 (91.818) Acc@5 98.264 (98.538) Mem 68106MB [2022-12-19 21:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.304) Loss 0.3895 (0.4756) Acc@1 92.361 (91.830) Acc@5 98.958 (98.555) Mem 68106MB [2022-12-19 21:50:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:39] * Acc@1 91.802 Acc@5 98.551 [2022-12-19 21:50:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.8% [2022-12-19 21:50:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 21:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 21:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.80% [2022-12-19 21:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][0/1519] eta 0:36:05 lr 0.000022 time 1.4255 (1.4255) model_time 0.9775 (0.9775) loss 1.3700 (1.3700) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 21:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][10/1519] eta 0:26:11 lr 0.000022 time 0.9319 (1.0415) model_time 0.9318 (1.0004) loss 0.7482 (0.9495) grad_norm 7.2598 (8.4847/1.6579) mem 68106MB [2022-12-19 21:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][20/1519] eta 0:25:33 lr 0.000022 time 0.9254 (1.0230) model_time 0.9253 (1.0014) loss 0.9327 (0.9141) grad_norm 6.4367 (7.7288/1.5343) mem 68106MB [2022-12-19 21:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][30/1519] eta 0:25:12 lr 0.000022 time 0.9237 (1.0157) model_time 0.9236 (1.0010) loss 0.8645 (0.9037) grad_norm 10.1166 (8.1409/1.5499) mem 68106MB [2022-12-19 21:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][40/1519] eta 0:24:55 lr 0.000022 time 0.9222 (1.0110) model_time 0.9221 (0.9998) loss 0.8952 (0.8938) grad_norm 6.5528 (8.3467/1.7432) mem 68106MB [2022-12-19 21:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][50/1519] eta 0:24:43 lr 0.000022 time 0.9247 (1.0097) model_time 0.9245 (1.0006) loss 1.0775 (0.8985) grad_norm 8.2663 (8.3229/1.6893) mem 68106MB [2022-12-19 21:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][60/1519] eta 0:24:30 lr 0.000022 time 0.9280 (1.0078) model_time 0.9280 (1.0002) loss 0.7097 (0.8949) grad_norm 7.3289 (8.1536/1.6245) mem 68106MB [2022-12-19 21:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][70/1519] eta 0:24:17 lr 0.000022 time 0.9216 (1.0062) model_time 0.9214 (0.9996) loss 1.1359 (0.9016) grad_norm 7.8270 (8.2453/1.5319) mem 68106MB [2022-12-19 21:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][80/1519] eta 0:24:06 lr 0.000022 time 0.9293 (1.0056) model_time 0.9292 (0.9997) loss 0.7137 (0.9031) grad_norm 9.3585 (8.3585/1.5689) mem 68106MB [2022-12-19 21:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][90/1519] eta 0:23:59 lr 0.000022 time 0.9118 (1.0074) model_time 0.9116 (1.0022) loss 0.8050 (0.8921) grad_norm 6.5410 (8.3884/1.6692) mem 68106MB [2022-12-19 21:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][100/1519] eta 0:23:49 lr 0.000022 time 0.9310 (1.0071) model_time 0.9309 (1.0024) loss 0.7051 (0.8907) grad_norm 6.6352 (8.2601/1.6449) mem 68106MB [2022-12-19 21:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][110/1519] eta 0:23:38 lr 0.000022 time 0.9344 (1.0064) model_time 0.9343 (1.0021) loss 0.8216 (0.8915) grad_norm 8.4601 (8.2780/1.5843) mem 68106MB [2022-12-19 21:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][120/1519] eta 0:23:28 lr 0.000022 time 0.9305 (1.0068) model_time 0.9303 (1.0028) loss 1.0470 (0.8906) grad_norm 7.3912 (8.2797/1.5860) mem 68106MB [2022-12-19 21:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][130/1519] eta 0:23:18 lr 0.000022 time 0.9089 (1.0065) model_time 0.9088 (1.0028) loss 0.6956 (0.8931) grad_norm 6.0412 (8.2917/1.6136) mem 68106MB [2022-12-19 21:53:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][140/1519] eta 0:23:08 lr 0.000022 time 0.9978 (1.0071) model_time 0.9976 (1.0036) loss 1.0083 (0.8886) grad_norm 11.0617 (8.2418/1.6453) mem 68106MB [2022-12-19 21:53:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][150/1519] eta 0:22:58 lr 0.000022 time 0.9256 (1.0069) model_time 0.9255 (1.0036) loss 0.9598 (0.8853) grad_norm 7.0225 (8.3082/1.6615) mem 68106MB [2022-12-19 21:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][160/1519] eta 0:22:47 lr 0.000022 time 0.9314 (1.0065) model_time 0.9310 (1.0034) loss 1.0254 (0.8846) grad_norm 9.0941 (8.3394/1.6521) mem 68106MB [2022-12-19 21:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][170/1519] eta 0:22:37 lr 0.000022 time 0.9290 (1.0061) model_time 0.9288 (1.0032) loss 0.6940 (0.8785) grad_norm 7.7485 (8.3394/1.6715) mem 68106MB [2022-12-19 21:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][180/1519] eta 0:22:27 lr 0.000022 time 0.9220 (1.0064) model_time 0.9218 (1.0036) loss 0.8399 (0.8773) grad_norm 8.1847 (8.3376/1.6479) mem 68106MB [2022-12-19 21:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][190/1519] eta 0:22:18 lr 0.000022 time 0.9246 (1.0068) model_time 0.9244 (1.0042) loss 0.8601 (0.8773) grad_norm 8.5511 (8.3573/1.6503) mem 68106MB [2022-12-19 21:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][200/1519] eta 0:22:07 lr 0.000022 time 0.9306 (1.0066) model_time 0.9304 (1.0040) loss 1.1676 (0.8753) grad_norm 10.1177 (8.3585/1.6464) mem 68106MB [2022-12-19 21:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][210/1519] eta 0:21:57 lr 0.000022 time 0.9254 (1.0066) model_time 0.9252 (1.0042) loss 0.8061 (0.8722) grad_norm 6.7773 (8.3852/1.6680) mem 68106MB [2022-12-19 21:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][220/1519] eta 0:21:47 lr 0.000022 time 0.9250 (1.0063) model_time 0.9248 (1.0039) loss 0.7113 (0.8729) grad_norm 8.9050 (8.3893/1.6364) mem 68106MB [2022-12-19 21:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][230/1519] eta 0:21:37 lr 0.000022 time 0.9230 (1.0066) model_time 0.9229 (1.0044) loss 0.8565 (0.8698) grad_norm 8.6660 (8.3555/1.6198) mem 68106MB [2022-12-19 21:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][240/1519] eta 0:21:27 lr 0.000022 time 0.9887 (1.0068) model_time 0.9885 (1.0046) loss 0.6924 (0.8710) grad_norm 6.7329 (8.4509/1.7155) mem 68106MB [2022-12-19 21:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][250/1519] eta 0:21:17 lr 0.000022 time 0.9155 (1.0066) model_time 0.9154 (1.0045) loss 0.7061 (0.8697) grad_norm 6.7382 (8.4202/1.7001) mem 68106MB [2022-12-19 21:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][260/1519] eta 0:21:06 lr 0.000022 time 0.9191 (1.0062) model_time 0.9189 (1.0042) loss 0.8826 (0.8685) grad_norm 7.6398 (8.4827/1.7888) mem 68106MB [2022-12-19 21:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][270/1519] eta 0:20:56 lr 0.000022 time 0.9262 (1.0060) model_time 0.9260 (1.0040) loss 0.7442 (0.8677) grad_norm 5.7895 (8.4863/1.7888) mem 68106MB [2022-12-19 21:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][280/1519] eta 0:20:46 lr 0.000022 time 0.9214 (1.0061) model_time 0.9212 (1.0042) loss 0.7387 (0.8674) grad_norm 7.9062 (8.5125/1.7691) mem 68106MB [2022-12-19 21:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][290/1519] eta 0:20:36 lr 0.000022 time 0.9358 (1.0060) model_time 0.9357 (1.0042) loss 0.7105 (0.8653) grad_norm 8.3029 (8.5161/1.7494) mem 68106MB [2022-12-19 21:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][300/1519] eta 0:20:26 lr 0.000022 time 0.9229 (1.0058) model_time 0.9227 (1.0040) loss 0.6778 (0.8649) grad_norm 8.0775 (8.4674/1.7654) mem 68106MB [2022-12-19 21:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][310/1519] eta 0:20:15 lr 0.000022 time 0.9259 (1.0057) model_time 0.9257 (1.0040) loss 0.9881 (0.8630) grad_norm 8.3506 (8.4658/1.7422) mem 68106MB [2022-12-19 21:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][320/1519] eta 0:20:05 lr 0.000022 time 0.9236 (1.0055) model_time 0.9235 (1.0038) loss 0.8071 (0.8626) grad_norm 9.6116 (8.4789/1.7367) mem 68106MB [2022-12-19 21:56:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][330/1519] eta 0:19:55 lr 0.000022 time 0.9235 (1.0053) model_time 0.9233 (1.0036) loss 0.7926 (0.8626) grad_norm 7.0778 (8.5058/1.7484) mem 68106MB [2022-12-19 21:56:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][340/1519] eta 0:19:44 lr 0.000022 time 0.9212 (1.0051) model_time 0.9211 (1.0034) loss 0.7374 (0.8636) grad_norm 6.5495 (8.5199/1.7395) mem 68106MB [2022-12-19 21:56:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][350/1519] eta 0:19:34 lr 0.000022 time 0.9388 (1.0050) model_time 0.9386 (1.0034) loss 0.7427 (0.8626) grad_norm 9.0781 (8.4907/1.7333) mem 68106MB [2022-12-19 21:56:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][360/1519] eta 0:19:24 lr 0.000022 time 0.9269 (1.0051) model_time 0.9268 (1.0036) loss 0.7162 (0.8601) grad_norm 6.7688 (8.4924/1.7337) mem 68106MB [2022-12-19 21:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][370/1519] eta 0:19:14 lr 0.000022 time 0.9248 (1.0050) model_time 0.9247 (1.0034) loss 0.7213 (0.8589) grad_norm 8.3582 (8.5063/1.7270) mem 68106MB [2022-12-19 21:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][380/1519] eta 0:19:04 lr 0.000022 time 0.9204 (1.0049) model_time 0.9203 (1.0034) loss 0.7859 (0.8603) grad_norm 9.5647 (8.4880/1.7201) mem 68106MB [2022-12-19 21:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][390/1519] eta 0:18:54 lr 0.000022 time 0.9480 (1.0049) model_time 0.9479 (1.0034) loss 0.7429 (0.8609) grad_norm 9.3565 (8.4990/1.7062) mem 68106MB [2022-12-19 21:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][400/1519] eta 0:18:44 lr 0.000022 time 0.9363 (1.0047) model_time 0.9361 (1.0033) loss 0.8465 (0.8624) grad_norm 9.1306 (8.4775/1.6955) mem 68106MB [2022-12-19 21:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][410/1519] eta 0:18:34 lr 0.000022 time 0.9271 (1.0051) model_time 0.9269 (1.0037) loss 0.7892 (0.8618) grad_norm 9.7984 (8.4815/1.6806) mem 68106MB [2022-12-19 21:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][420/1519] eta 0:18:24 lr 0.000022 time 0.9925 (1.0052) model_time 0.9923 (1.0038) loss 0.8462 (0.8623) grad_norm 8.0124 (8.4962/1.6801) mem 68106MB [2022-12-19 21:58:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][430/1519] eta 0:18:14 lr 0.000022 time 0.9267 (1.0052) model_time 0.9266 (1.0039) loss 0.8014 (0.8621) grad_norm 7.8110 (8.5530/1.8184) mem 68106MB [2022-12-19 21:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][440/1519] eta 0:18:04 lr 0.000022 time 0.9192 (1.0051) model_time 0.9190 (1.0037) loss 0.7404 (0.8620) grad_norm 6.7957 (8.5270/1.8082) mem 68106MB [2022-12-19 21:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][450/1519] eta 0:17:54 lr 0.000022 time 0.9401 (1.0052) model_time 0.9399 (1.0038) loss 1.0346 (0.8636) grad_norm 8.1895 (8.5409/1.8204) mem 68106MB [2022-12-19 21:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][460/1519] eta 0:17:44 lr 0.000022 time 0.9247 (1.0052) model_time 0.9245 (1.0039) loss 1.0644 (0.8640) grad_norm 10.7059 (8.5576/1.8087) mem 68106MB [2022-12-19 21:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][470/1519] eta 0:17:34 lr 0.000022 time 0.9330 (1.0053) model_time 0.9328 (1.0040) loss 0.6941 (0.8633) grad_norm 10.8643 (8.5553/1.7994) mem 68106MB [2022-12-19 21:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][480/1519] eta 0:17:24 lr 0.000022 time 0.9317 (1.0053) model_time 0.9316 (1.0041) loss 0.8094 (0.8629) grad_norm 6.6552 (8.5533/1.7994) mem 68106MB [2022-12-19 21:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][490/1519] eta 0:17:14 lr 0.000022 time 0.9319 (1.0057) model_time 0.9318 (1.0044) loss 1.1250 (0.8622) grad_norm 8.2287 (8.5518/1.7846) mem 68106MB [2022-12-19 21:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][500/1519] eta 0:17:04 lr 0.000022 time 0.9383 (1.0058) model_time 0.9382 (1.0046) loss 0.8657 (0.8621) grad_norm 8.3314 (8.5661/1.8093) mem 68106MB [2022-12-19 21:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][510/1519] eta 0:16:54 lr 0.000022 time 0.9253 (1.0057) model_time 0.9252 (1.0045) loss 0.8265 (0.8614) grad_norm 13.2640 (8.6077/1.9034) mem 68106MB [2022-12-19 21:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][520/1519] eta 0:16:44 lr 0.000022 time 0.9331 (1.0058) model_time 0.9330 (1.0046) loss 0.7378 (0.8616) grad_norm 9.1986 (8.6084/1.8986) mem 68106MB [2022-12-19 21:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][530/1519] eta 0:16:34 lr 0.000022 time 0.9215 (1.0057) model_time 0.9213 (1.0045) loss 0.8243 (0.8617) grad_norm 10.0109 (8.6484/1.9081) mem 68106MB [2022-12-19 21:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][540/1519] eta 0:16:24 lr 0.000022 time 0.9293 (1.0058) model_time 0.9292 (1.0046) loss 0.9090 (0.8621) grad_norm 6.7127 (8.6191/1.9077) mem 68106MB [2022-12-19 22:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][550/1519] eta 0:16:14 lr 0.000022 time 0.9409 (1.0057) model_time 0.9408 (1.0046) loss 1.1773 (0.8622) grad_norm 9.8016 (8.6236/1.9103) mem 68106MB [2022-12-19 22:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][560/1519] eta 0:16:04 lr 0.000022 time 0.9275 (1.0056) model_time 0.9274 (1.0045) loss 0.8015 (0.8607) grad_norm 11.8119 (8.6364/1.9063) mem 68106MB [2022-12-19 22:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][570/1519] eta 0:15:54 lr 0.000022 time 0.9307 (1.0056) model_time 0.9306 (1.0045) loss 0.7253 (0.8608) grad_norm 7.7345 (8.6271/1.8933) mem 68106MB [2022-12-19 22:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][580/1519] eta 0:15:44 lr 0.000022 time 0.9324 (1.0055) model_time 0.9323 (1.0044) loss 0.7166 (0.8613) grad_norm 7.1950 (8.6181/1.8802) mem 68106MB [2022-12-19 22:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][590/1519] eta 0:15:34 lr 0.000022 time 0.9085 (1.0055) model_time 0.9083 (1.0044) loss 0.9239 (0.8609) grad_norm 7.7331 (8.6031/1.8759) mem 68106MB [2022-12-19 22:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][600/1519] eta 0:15:24 lr 0.000022 time 0.9341 (1.0055) model_time 0.9340 (1.0044) loss 0.7522 (0.8610) grad_norm 11.1769 (8.6048/1.8763) mem 68106MB [2022-12-19 22:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][610/1519] eta 0:15:13 lr 0.000022 time 0.9213 (1.0053) model_time 0.9212 (1.0043) loss 1.1127 (0.8606) grad_norm 13.1452 (8.6443/1.9281) mem 68106MB [2022-12-19 22:01:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][620/1519] eta 0:15:03 lr 0.000022 time 0.9261 (1.0053) model_time 0.9253 (1.0042) loss 0.7067 (0.8611) grad_norm 8.0840 (8.6669/1.9129) mem 68106MB [2022-12-19 22:01:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][630/1519] eta 0:14:53 lr 0.000022 time 0.9318 (1.0052) model_time 0.9317 (1.0041) loss 0.8154 (0.8623) grad_norm 8.3146 (8.6647/1.9152) mem 68106MB [2022-12-19 22:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][640/1519] eta 0:14:43 lr 0.000022 time 0.9292 (1.0053) model_time 0.9289 (1.0042) loss 0.8233 (0.8625) grad_norm 6.4791 (8.6678/1.9322) mem 68106MB [2022-12-19 22:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][650/1519] eta 0:14:33 lr 0.000022 time 0.9341 (1.0052) model_time 0.9339 (1.0041) loss 0.8608 (0.8632) grad_norm 8.8875 (8.6777/1.9353) mem 68106MB [2022-12-19 22:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][660/1519] eta 0:14:23 lr 0.000022 time 0.9288 (1.0052) model_time 0.9286 (1.0042) loss 1.0633 (0.8640) grad_norm 7.6820 (8.7228/1.9435) mem 68106MB [2022-12-19 22:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][670/1519] eta 0:14:13 lr 0.000022 time 0.9119 (1.0054) model_time 0.9116 (1.0043) loss 0.8294 (0.8631) grad_norm 7.0732 (8.7171/1.9512) mem 68106MB [2022-12-19 22:02:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][680/1519] eta 0:14:03 lr 0.000022 time 0.9322 (1.0053) model_time 0.9320 (1.0042) loss 0.8858 (0.8639) grad_norm 6.8995 (8.6914/1.9552) mem 68106MB [2022-12-19 22:02:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][690/1519] eta 0:13:53 lr 0.000022 time 0.9438 (1.0053) model_time 0.9436 (1.0043) loss 0.9362 (0.8647) grad_norm 8.9508 (8.6951/1.9398) mem 68106MB [2022-12-19 22:02:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][700/1519] eta 0:13:43 lr 0.000022 time 0.9370 (1.0055) model_time 0.9368 (1.0045) loss 0.8232 (0.8645) grad_norm 6.5269 (8.7284/1.9486) mem 68106MB [2022-12-19 22:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][710/1519] eta 0:13:33 lr 0.000022 time 0.9298 (1.0054) model_time 0.9295 (1.0044) loss 0.7502 (0.8646) grad_norm 6.6437 (8.7181/1.9527) mem 68106MB [2022-12-19 22:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][720/1519] eta 0:13:23 lr 0.000022 time 0.9369 (1.0059) model_time 0.9368 (1.0049) loss 0.7382 (0.8636) grad_norm 7.2746 (8.6953/1.9566) mem 68106MB [2022-12-19 22:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][730/1519] eta 0:13:13 lr 0.000022 time 0.9397 (1.0060) model_time 0.9396 (1.0051) loss 0.9130 (0.8641) grad_norm 8.1744 (8.7029/1.9578) mem 68106MB [2022-12-19 22:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][740/1519] eta 0:13:03 lr 0.000022 time 0.9299 (1.0061) model_time 0.9298 (1.0051) loss 0.8666 (0.8650) grad_norm 9.0762 (8.7085/1.9439) mem 68106MB [2022-12-19 22:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][750/1519] eta 0:12:53 lr 0.000022 time 0.9280 (1.0061) model_time 0.9277 (1.0051) loss 0.7047 (0.8656) grad_norm 8.8685 (8.6927/1.9360) mem 68106MB [2022-12-19 22:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][760/1519] eta 0:12:43 lr 0.000022 time 0.9159 (1.0061) model_time 0.9155 (1.0051) loss 0.7942 (0.8659) grad_norm 9.1887 (8.6878/1.9359) mem 68106MB [2022-12-19 22:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][770/1519] eta 0:12:33 lr 0.000022 time 0.9368 (1.0061) model_time 0.9366 (1.0051) loss 0.8133 (0.8662) grad_norm 7.9121 (8.7117/1.9618) mem 68106MB [2022-12-19 22:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][780/1519] eta 0:12:23 lr 0.000022 time 0.9257 (1.0060) model_time 0.9255 (1.0050) loss 0.8942 (0.8662) grad_norm 7.9788 (8.7237/1.9607) mem 68106MB [2022-12-19 22:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][790/1519] eta 0:12:13 lr 0.000022 time 0.9193 (1.0059) model_time 0.9192 (1.0050) loss 0.8236 (0.8667) grad_norm 10.9692 (8.7247/1.9588) mem 68106MB [2022-12-19 22:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][800/1519] eta 0:12:03 lr 0.000022 time 0.9516 (1.0059) model_time 0.9514 (1.0050) loss 0.6820 (0.8669) grad_norm 8.4499 (8.7396/1.9536) mem 68106MB [2022-12-19 22:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][810/1519] eta 0:11:53 lr 0.000022 time 0.9333 (1.0061) model_time 0.9331 (1.0051) loss 0.7617 (0.8689) grad_norm 11.2200 (8.7454/1.9507) mem 68106MB [2022-12-19 22:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][820/1519] eta 0:11:43 lr 0.000022 time 0.9122 (1.0060) model_time 0.9120 (1.0051) loss 0.7085 (0.8687) grad_norm 5.8316 (8.7346/1.9590) mem 68106MB [2022-12-19 22:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][830/1519] eta 0:11:33 lr 0.000022 time 0.9391 (1.0060) model_time 0.9389 (1.0051) loss 0.6987 (0.8691) grad_norm 10.0179 (8.7698/2.0059) mem 68106MB [2022-12-19 22:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][840/1519] eta 0:11:23 lr 0.000022 time 0.9308 (1.0059) model_time 0.9306 (1.0050) loss 1.0076 (0.8694) grad_norm 9.7863 (8.7697/1.9980) mem 68106MB [2022-12-19 22:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][850/1519] eta 0:11:12 lr 0.000022 time 0.9304 (1.0059) model_time 0.9302 (1.0050) loss 0.9579 (0.8699) grad_norm 8.2871 (8.7912/1.9929) mem 68106MB [2022-12-19 22:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][860/1519] eta 0:11:02 lr 0.000022 time 0.9304 (1.0058) model_time 0.9302 (1.0049) loss 0.7143 (0.8694) grad_norm 12.7817 (8.7800/1.9687) mem 68106MB [2022-12-19 22:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][870/1519] eta 0:10:52 lr 0.000022 time 0.9295 (1.0058) model_time 0.9293 (1.0049) loss 1.3501 (0.8700) grad_norm 11.4212 (8.8005/1.9724) mem 68106MB [2022-12-19 22:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][880/1519] eta 0:10:42 lr 0.000022 time 0.9422 (1.0058) model_time 0.9420 (1.0049) loss 1.0060 (0.8701) grad_norm 7.6743 (8.7917/1.9840) mem 68106MB [2022-12-19 22:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][890/1519] eta 0:10:32 lr 0.000022 time 0.9358 (1.0058) model_time 0.9355 (1.0049) loss 1.1314 (0.8705) grad_norm 8.8777 (8.7858/1.9829) mem 68106MB [2022-12-19 22:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][900/1519] eta 0:10:22 lr 0.000022 time 0.9427 (1.0058) model_time 0.9425 (1.0049) loss 1.1353 (0.8704) grad_norm 9.5691 (8.8121/1.9602) mem 68106MB [2022-12-19 22:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][910/1519] eta 0:10:12 lr 0.000022 time 0.9960 (1.0058) model_time 0.9958 (1.0050) loss 0.7645 (0.8708) grad_norm 7.8343 (8.7997/1.9641) mem 68106MB [2022-12-19 22:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][920/1519] eta 0:10:02 lr 0.000022 time 0.9409 (1.0059) model_time 0.9407 (1.0050) loss 0.7054 (0.8703) grad_norm 8.6484 (8.7902/1.9563) mem 68106MB [2022-12-19 22:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][930/1519] eta 0:09:52 lr 0.000022 time 0.9288 (1.0058) model_time 0.9284 (1.0049) loss 0.8802 (0.8700) grad_norm 8.7448 (8.7678/1.9428) mem 68106MB [2022-12-19 22:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][940/1519] eta 0:09:42 lr 0.000022 time 0.9303 (1.0058) model_time 0.9301 (1.0049) loss 0.7814 (0.8698) grad_norm 11.5858 (8.7742/1.9484) mem 68106MB [2022-12-19 22:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][950/1519] eta 0:09:32 lr 0.000022 time 0.9362 (1.0057) model_time 0.9360 (1.0048) loss 0.7729 (0.8689) grad_norm 6.7687 (8.8003/1.9844) mem 68106MB [2022-12-19 22:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][960/1519] eta 0:09:22 lr 0.000022 time 0.9375 (1.0057) model_time 0.9373 (1.0048) loss 0.8910 (0.8691) grad_norm 7.1588 (8.7814/1.9807) mem 68106MB [2022-12-19 22:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][970/1519] eta 0:09:12 lr 0.000022 time 0.9191 (1.0057) model_time 0.9189 (1.0048) loss 0.7473 (0.8689) grad_norm 8.8415 (8.7618/1.9821) mem 68106MB [2022-12-19 22:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][980/1519] eta 0:09:02 lr 0.000022 time 0.9756 (1.0057) model_time 0.9754 (1.0048) loss 0.6783 (0.8682) grad_norm 7.9288 (8.7461/1.9886) mem 68106MB [2022-12-19 22:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][990/1519] eta 0:08:52 lr 0.000022 time 0.9328 (1.0058) model_time 0.9325 (1.0050) loss 0.6824 (0.8673) grad_norm 8.4047 (8.7284/1.9918) mem 68106MB [2022-12-19 22:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1000/1519] eta 0:08:41 lr 0.000022 time 0.9250 (1.0057) model_time 0.9248 (1.0049) loss 0.7644 (0.8665) grad_norm 7.8947 (8.7211/2.0013) mem 68106MB [2022-12-19 22:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1010/1519] eta 0:08:31 lr 0.000022 time 0.9296 (1.0058) model_time 0.9295 (1.0050) loss 0.6854 (0.8666) grad_norm 10.0947 (8.7405/2.0043) mem 68106MB [2022-12-19 22:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1020/1519] eta 0:08:21 lr 0.000022 time 0.9260 (1.0057) model_time 0.9259 (1.0049) loss 1.0061 (0.8666) grad_norm 8.4214 (8.7491/2.0138) mem 68106MB [2022-12-19 22:08:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1030/1519] eta 0:08:11 lr 0.000022 time 0.9285 (1.0057) model_time 0.9284 (1.0049) loss 0.8121 (0.8665) grad_norm 7.3640 (8.7051/1.9269) mem 68106MB [2022-12-19 22:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1040/1519] eta 0:08:01 lr 0.000022 time 0.9487 (1.0058) model_time 0.9485 (1.0050) loss 0.8326 (0.8665) grad_norm 9.2545 (8.7227/1.9250) mem 68106MB [2022-12-19 22:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1050/1519] eta 0:07:51 lr 0.000022 time 0.9261 (1.0058) model_time 0.9259 (1.0050) loss 0.7683 (0.8661) grad_norm 7.8371 (8.7209/1.9107) mem 68106MB [2022-12-19 22:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1060/1519] eta 0:07:41 lr 0.000022 time 0.9200 (1.0058) model_time 0.9199 (1.0049) loss 0.7202 (0.8655) grad_norm 8.4363 (8.7025/1.9070) mem 68106MB [2022-12-19 22:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1070/1519] eta 0:07:31 lr 0.000022 time 0.9847 (1.0057) model_time 0.9846 (1.0049) loss 0.8488 (0.8652) grad_norm 10.5636 (8.7120/1.9127) mem 68106MB [2022-12-19 22:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1080/1519] eta 0:07:21 lr 0.000022 time 0.9334 (1.0058) model_time 0.9333 (1.0050) loss 0.8841 (0.8647) grad_norm 6.5977 (8.7023/1.9059) mem 68106MB [2022-12-19 22:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1090/1519] eta 0:07:11 lr 0.000022 time 0.9282 (1.0057) model_time 0.9280 (1.0049) loss 0.7400 (0.8639) grad_norm 8.5642 (8.6862/1.9119) mem 68106MB [2022-12-19 22:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1100/1519] eta 0:07:01 lr 0.000022 time 0.9223 (1.0057) model_time 0.9221 (1.0049) loss 0.8579 (0.8639) grad_norm 10.6218 (8.7036/1.9192) mem 68106MB [2022-12-19 22:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1110/1519] eta 0:06:51 lr 0.000022 time 0.8906 (1.0057) model_time 0.8905 (1.0049) loss 0.8907 (0.8639) grad_norm 11.4529 (8.7034/1.8487) mem 68106MB [2022-12-19 22:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1120/1519] eta 0:06:41 lr 0.000022 time 0.9248 (1.0056) model_time 0.9246 (1.0049) loss 0.9181 (0.8638) grad_norm 7.3016 (8.6975/1.8431) mem 68106MB [2022-12-19 22:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1130/1519] eta 0:06:31 lr 0.000022 time 0.9290 (1.0056) model_time 0.9289 (1.0048) loss 0.8654 (0.8636) grad_norm 7.8158 (8.6512/1.8240) mem 68106MB [2022-12-19 22:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1140/1519] eta 0:06:21 lr 0.000022 time 0.9317 (1.0055) model_time 0.9316 (1.0048) loss 1.3558 (0.8641) grad_norm 9.9192 (8.7085/1.8463) mem 68106MB [2022-12-19 22:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1150/1519] eta 0:06:11 lr 0.000022 time 0.9351 (1.0055) model_time 0.9350 (1.0047) loss 0.7120 (0.8644) grad_norm 10.8816 (8.7051/1.8378) mem 68106MB [2022-12-19 22:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1160/1519] eta 0:06:00 lr 0.000022 time 1.0022 (1.0055) model_time 1.0021 (1.0047) loss 0.7368 (0.8638) grad_norm 6.6134 (8.6751/1.8383) mem 68106MB [2022-12-19 22:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1170/1519] eta 0:05:50 lr 0.000022 time 0.9218 (1.0054) model_time 0.9217 (1.0047) loss 0.9869 (0.8644) grad_norm 10.9713 (8.6852/1.8553) mem 68106MB [2022-12-19 22:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1180/1519] eta 0:05:40 lr 0.000022 time 0.9211 (1.0054) model_time 0.9210 (1.0046) loss 0.7199 (0.8644) grad_norm 9.0596 (8.7006/1.8682) mem 68106MB [2022-12-19 22:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1190/1519] eta 0:05:30 lr 0.000022 time 0.9281 (1.0053) model_time 0.9280 (1.0046) loss 0.9412 (0.8640) grad_norm 8.3373 (8.7007/1.8636) mem 68106MB [2022-12-19 22:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1200/1519] eta 0:05:20 lr 0.000022 time 0.9266 (1.0053) model_time 0.9263 (1.0045) loss 1.0526 (0.8639) grad_norm 10.3540 (8.6996/1.8562) mem 68106MB [2022-12-19 22:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1210/1519] eta 0:05:10 lr 0.000022 time 0.9375 (1.0054) model_time 0.9374 (1.0046) loss 0.9147 (0.8642) grad_norm 10.0645 (8.6685/1.8088) mem 68106MB [2022-12-19 22:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1220/1519] eta 0:05:00 lr 0.000022 time 0.9193 (1.0054) model_time 0.9190 (1.0046) loss 0.7000 (0.8646) grad_norm 7.7665 (8.6625/1.8175) mem 68106MB [2022-12-19 22:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1230/1519] eta 0:04:50 lr 0.000022 time 0.9317 (1.0054) model_time 0.9316 (1.0046) loss 0.9267 (0.8642) grad_norm 14.8279 (8.7008/1.8542) mem 68106MB [2022-12-19 22:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1240/1519] eta 0:04:40 lr 0.000022 time 0.9224 (1.0054) model_time 0.9222 (1.0046) loss 0.7701 (0.8637) grad_norm 5.5776 (8.6786/1.8339) mem 68106MB [2022-12-19 22:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1250/1519] eta 0:04:30 lr 0.000022 time 0.9847 (1.0054) model_time 0.9845 (1.0046) loss 1.0410 (0.8637) grad_norm 7.6808 (8.6638/1.8327) mem 68106MB [2022-12-19 22:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1260/1519] eta 0:04:20 lr 0.000022 time 0.9293 (1.0054) model_time 0.9291 (1.0047) loss 0.8415 (0.8636) grad_norm 8.5391 (8.6469/1.8478) mem 68106MB [2022-12-19 22:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1270/1519] eta 0:04:10 lr 0.000022 time 0.9333 (1.0054) model_time 0.9331 (1.0046) loss 0.9246 (0.8643) grad_norm 10.2282 (8.6891/1.8968) mem 68106MB [2022-12-19 22:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1280/1519] eta 0:04:00 lr 0.000022 time 0.9320 (1.0054) model_time 0.9319 (1.0046) loss 0.8014 (0.8648) grad_norm 10.9135 (8.7055/1.8970) mem 68106MB [2022-12-19 22:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1290/1519] eta 0:03:50 lr 0.000022 time 0.9807 (1.0054) model_time 0.9802 (1.0047) loss 0.7230 (0.8650) grad_norm 6.8402 (8.7043/1.9115) mem 68106MB [2022-12-19 22:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1300/1519] eta 0:03:40 lr 0.000022 time 0.9104 (1.0055) model_time 0.9102 (1.0048) loss 1.1099 (0.8654) grad_norm 9.2589 (8.7076/1.9194) mem 68106MB [2022-12-19 22:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1310/1519] eta 0:03:30 lr 0.000022 time 0.9304 (1.0055) model_time 0.9302 (1.0047) loss 1.1805 (0.8660) grad_norm 10.9036 (8.7357/1.9559) mem 68106MB [2022-12-19 22:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1320/1519] eta 0:03:20 lr 0.000022 time 0.9235 (1.0054) model_time 0.9233 (1.0047) loss 0.8199 (0.8665) grad_norm 11.9356 (8.7965/1.9657) mem 68106MB [2022-12-19 22:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1330/1519] eta 0:03:10 lr 0.000022 time 0.9474 (1.0055) model_time 0.9472 (1.0047) loss 0.7978 (0.8671) grad_norm 10.8333 (8.7899/1.9568) mem 68106MB [2022-12-19 22:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1340/1519] eta 0:02:59 lr 0.000022 time 0.9243 (1.0054) model_time 0.9242 (1.0047) loss 0.8856 (0.8669) grad_norm 9.9274 (8.8029/1.9590) mem 68106MB [2022-12-19 22:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1350/1519] eta 0:02:49 lr 0.000022 time 0.9448 (1.0055) model_time 0.9447 (1.0048) loss 1.1429 (0.8668) grad_norm 16.8564 (8.8382/2.0234) mem 68106MB [2022-12-19 22:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1360/1519] eta 0:02:39 lr 0.000022 time 0.8981 (1.0056) model_time 0.8978 (1.0048) loss 0.7379 (0.8669) grad_norm 7.1236 (8.8343/2.0195) mem 68106MB [2022-12-19 22:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1370/1519] eta 0:02:29 lr 0.000022 time 0.9335 (1.0055) model_time 0.9333 (1.0048) loss 0.8432 (0.8675) grad_norm 8.0647 (8.8120/1.9813) mem 68106MB [2022-12-19 22:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1380/1519] eta 0:02:19 lr 0.000022 time 0.9387 (1.0055) model_time 0.9386 (1.0048) loss 0.8509 (0.8673) grad_norm 6.4300 (8.7882/1.9878) mem 68106MB [2022-12-19 22:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1390/1519] eta 0:02:09 lr 0.000022 time 0.9283 (1.0055) model_time 0.9281 (1.0047) loss 0.9130 (0.8671) grad_norm 9.7654 (8.7996/1.9831) mem 68106MB [2022-12-19 22:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1400/1519] eta 0:01:59 lr 0.000022 time 0.9684 (1.0055) model_time 0.9682 (1.0047) loss 0.9423 (0.8668) grad_norm 7.2970 (8.7968/1.9908) mem 68106MB [2022-12-19 22:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1410/1519] eta 0:01:49 lr 0.000022 time 0.9304 (1.0054) model_time 0.9302 (1.0047) loss 0.7243 (0.8667) grad_norm 8.0931 (8.7849/1.9879) mem 68106MB [2022-12-19 22:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1420/1519] eta 0:01:39 lr 0.000022 time 0.9234 (1.0054) model_time 0.9227 (1.0047) loss 0.6803 (0.8666) grad_norm 6.4568 (8.7788/1.9875) mem 68106MB [2022-12-19 22:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1430/1519] eta 0:01:29 lr 0.000022 time 0.9147 (1.0054) model_time 0.9145 (1.0047) loss 0.7321 (0.8662) grad_norm 9.4075 (8.7664/1.9507) mem 68106MB [2022-12-19 22:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1440/1519] eta 0:01:19 lr 0.000022 time 0.9333 (1.0054) model_time 0.9332 (1.0047) loss 0.9375 (0.8658) grad_norm 6.8297 (8.7293/1.9533) mem 68106MB [2022-12-19 22:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1450/1519] eta 0:01:09 lr 0.000022 time 0.9242 (1.0054) model_time 0.9241 (1.0046) loss 0.7913 (0.8653) grad_norm 7.0167 (8.7041/1.9571) mem 68106MB [2022-12-19 22:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1460/1519] eta 0:00:59 lr 0.000022 time 0.9312 (1.0053) model_time 0.9311 (1.0046) loss 1.1664 (0.8656) grad_norm 7.5254 (8.6817/1.9414) mem 68106MB [2022-12-19 22:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1470/1519] eta 0:00:49 lr 0.000022 time 0.9283 (1.0053) model_time 0.9280 (1.0046) loss 1.1033 (0.8658) grad_norm 9.0902 (8.6515/1.9291) mem 68106MB [2022-12-19 22:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1480/1519] eta 0:00:39 lr 0.000022 time 0.9305 (1.0053) model_time 0.9303 (1.0046) loss 0.7595 (0.8658) grad_norm 7.6060 (8.6556/1.9360) mem 68106MB [2022-12-19 22:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1490/1519] eta 0:00:29 lr 0.000022 time 0.9180 (1.0053) model_time 0.9175 (1.0046) loss 0.9209 (0.8657) grad_norm 10.8156 (8.6765/1.9416) mem 68106MB [2022-12-19 22:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1500/1519] eta 0:00:19 lr 0.000022 time 0.9186 (1.0052) model_time 0.9183 (1.0045) loss 0.7087 (0.8656) grad_norm 10.3431 (8.7080/1.9610) mem 68106MB [2022-12-19 22:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [40/100][1510/1519] eta 0:00:09 lr 0.000022 time 0.9190 (1.0052) model_time 0.9189 (1.0045) loss 1.0423 (0.8656) grad_norm 11.3069 (8.7381/1.9946) mem 68106MB [2022-12-19 22:16:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 40 training takes 0:25:26 [2022-12-19 22:16:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_40.pth saving...... [2022-12-19 22:16:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_40.pth saved !!! [2022-12-19 22:16:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.686 (0.686) Loss 0.5350 (0.5350) Acc@1 89.931 (89.931) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 22:16:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.333) Loss 0.4925 (0.4820) Acc@1 92.361 (92.077) Acc@5 98.264 (98.422) Mem 68106MB [2022-12-19 22:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.320) Loss 0.4623 (0.4810) Acc@1 91.667 (91.832) Acc@5 98.611 (98.495) Mem 68106MB [2022-12-19 22:16:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.313) Loss 0.5653 (0.4866) Acc@1 91.319 (91.723) Acc@5 98.264 (98.398) Mem 68106MB [2022-12-19 22:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.309) Loss 0.4494 (0.4792) Acc@1 92.708 (91.734) Acc@5 98.958 (98.450) Mem 68106MB [2022-12-19 22:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.307) Loss 0.4859 (0.4782) Acc@1 90.625 (91.789) Acc@5 99.306 (98.489) Mem 68106MB [2022-12-19 22:17:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.306) Loss 0.4963 (0.4780) Acc@1 90.625 (91.781) Acc@5 97.917 (98.475) Mem 68106MB [2022-12-19 22:17:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.302 (0.306) Loss 0.5123 (0.4790) Acc@1 91.319 (91.706) Acc@5 98.264 (98.455) Mem 68106MB [2022-12-19 22:17:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.305) Loss 0.4075 (0.4763) Acc@1 93.403 (91.761) Acc@5 98.611 (98.495) Mem 68106MB [2022-12-19 22:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:40] * Acc@1 91.724 Acc@5 98.490 [2022-12-19 22:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.7% [2022-12-19 22:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.80% [2022-12-19 22:17:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][0/1519] eta 0:54:35 lr 0.000022 time 2.1566 (2.1566) model_time 1.3172 (1.3172) loss 0.7176 (0.7176) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 22:17:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][10/1519] eta 0:27:58 lr 0.000022 time 0.9981 (1.1125) model_time 0.9980 (1.0359) loss 0.7611 (0.8663) grad_norm 8.6434 (8.1299/0.9386) mem 68106MB [2022-12-19 22:17:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][20/1519] eta 0:26:32 lr 0.000022 time 0.9278 (1.0621) model_time 0.9274 (1.0218) loss 0.8363 (0.9048) grad_norm 13.7330 (8.4805/2.1072) mem 68106MB [2022-12-19 22:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][30/1519] eta 0:25:54 lr 0.000022 time 0.9337 (1.0439) model_time 0.9336 (1.0165) loss 1.1769 (0.8808) grad_norm 7.3286 (8.4930/2.1240) mem 68106MB [2022-12-19 22:17:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][40/1519] eta 0:25:34 lr 0.000022 time 0.9760 (1.0378) model_time 0.9758 (1.0169) loss 0.7421 (0.8609) grad_norm 9.1143 (8.7173/2.0039) mem 68106MB [2022-12-19 22:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][50/1519] eta 0:25:15 lr 0.000022 time 0.9207 (1.0319) model_time 0.9205 (1.0151) loss 0.7398 (0.8673) grad_norm 7.9314 (9.1928/2.2144) mem 68106MB [2022-12-19 22:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][60/1519] eta 0:24:59 lr 0.000022 time 0.9312 (1.0275) model_time 0.9310 (1.0133) loss 0.6877 (0.8592) grad_norm 7.5054 (8.9427/2.1198) mem 68106MB [2022-12-19 22:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][70/1519] eta 0:24:44 lr 0.000022 time 0.9202 (1.0245) model_time 0.9201 (1.0123) loss 0.7727 (0.8637) grad_norm 6.5394 (8.8437/2.0304) mem 68106MB [2022-12-19 22:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][80/1519] eta 0:24:31 lr 0.000022 time 0.9304 (1.0223) model_time 0.9302 (1.0115) loss 0.7652 (0.8580) grad_norm 8.1014 (9.0075/2.1340) mem 68106MB [2022-12-19 22:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][90/1519] eta 0:24:18 lr 0.000022 time 0.9341 (1.0205) model_time 0.9339 (1.0109) loss 0.7388 (0.8490) grad_norm 10.1085 (9.0334/2.0399) mem 68106MB [2022-12-19 22:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][100/1519] eta 0:24:06 lr 0.000022 time 0.9298 (1.0196) model_time 0.9297 (1.0109) loss 0.9755 (0.8536) grad_norm 7.6993 (8.9819/2.0236) mem 68106MB [2022-12-19 22:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][110/1519] eta 0:23:54 lr 0.000022 time 0.9096 (1.0180) model_time 0.9094 (1.0101) loss 0.7306 (0.8560) grad_norm 8.4423 (8.9293/1.9541) mem 68106MB [2022-12-19 22:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][120/1519] eta 0:23:42 lr 0.000022 time 0.9312 (1.0165) model_time 0.9310 (1.0092) loss 0.8220 (0.8574) grad_norm 11.0476 (8.9078/1.9068) mem 68106MB [2022-12-19 22:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][130/1519] eta 0:23:30 lr 0.000022 time 0.9237 (1.0152) model_time 0.9235 (1.0084) loss 0.7522 (0.8587) grad_norm 8.3071 (8.9176/1.8701) mem 68106MB [2022-12-19 22:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][140/1519] eta 0:23:18 lr 0.000022 time 0.9312 (1.0141) model_time 0.9308 (1.0078) loss 0.6868 (0.8554) grad_norm 10.2527 (8.8493/1.8547) mem 68106MB [2022-12-19 22:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][150/1519] eta 0:23:06 lr 0.000022 time 0.9222 (1.0130) model_time 0.9220 (1.0070) loss 0.7896 (0.8591) grad_norm 6.4435 (8.7751/1.8592) mem 68106MB [2022-12-19 22:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][160/1519] eta 0:22:55 lr 0.000022 time 0.9341 (1.0124) model_time 0.9339 (1.0068) loss 0.8987 (0.8619) grad_norm 7.8051 (8.7354/1.8343) mem 68106MB [2022-12-19 22:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][170/1519] eta 0:22:44 lr 0.000022 time 0.9846 (1.0118) model_time 0.9845 (1.0065) loss 0.8581 (0.8588) grad_norm 7.3332 (8.7168/1.8168) mem 68106MB [2022-12-19 22:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][180/1519] eta 0:22:34 lr 0.000022 time 1.0330 (1.0116) model_time 1.0325 (1.0066) loss 0.8493 (0.8571) grad_norm 10.2551 (8.7257/1.8292) mem 68106MB [2022-12-19 22:20:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][190/1519] eta 0:22:23 lr 0.000022 time 0.9223 (1.0111) model_time 0.9221 (1.0063) loss 0.7986 (0.8581) grad_norm 7.1649 (8.6737/1.8037) mem 68106MB [2022-12-19 22:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][200/1519] eta 0:22:12 lr 0.000022 time 0.9326 (1.0105) model_time 0.9324 (1.0059) loss 0.6821 (0.8573) grad_norm 9.6049 (8.6582/1.8172) mem 68106MB [2022-12-19 22:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][210/1519] eta 0:22:01 lr 0.000022 time 0.9356 (1.0099) model_time 0.9355 (1.0055) loss 0.7475 (0.8555) grad_norm 8.2885 (8.6694/1.7943) mem 68106MB [2022-12-19 22:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][220/1519] eta 0:21:51 lr 0.000022 time 0.9820 (1.0096) model_time 0.9819 (1.0054) loss 0.8577 (0.8530) grad_norm 7.0937 (8.6482/1.7645) mem 68106MB [2022-12-19 22:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][230/1519] eta 0:21:40 lr 0.000022 time 0.9302 (1.0091) model_time 0.9301 (1.0050) loss 0.6862 (0.8477) grad_norm 6.1857 (8.5952/1.7557) mem 68106MB [2022-12-19 22:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][240/1519] eta 0:21:30 lr 0.000022 time 0.9435 (1.0087) model_time 0.9433 (1.0048) loss 0.7008 (0.8459) grad_norm 5.8896 (8.5439/1.7444) mem 68106MB [2022-12-19 22:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][250/1519] eta 0:21:19 lr 0.000022 time 0.9429 (1.0084) model_time 0.9428 (1.0046) loss 0.9154 (0.8448) grad_norm 8.9512 (8.5142/1.7194) mem 68106MB [2022-12-19 22:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][260/1519] eta 0:21:09 lr 0.000022 time 0.9310 (1.0080) model_time 0.9309 (1.0044) loss 0.6800 (0.8446) grad_norm 8.5926 (8.5135/1.6910) mem 68106MB [2022-12-19 22:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][270/1519] eta 0:20:58 lr 0.000022 time 0.9303 (1.0080) model_time 0.9299 (1.0045) loss 0.9593 (0.8445) grad_norm 7.3561 (8.5105/1.6761) mem 68106MB [2022-12-19 22:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][280/1519] eta 0:20:49 lr 0.000022 time 0.9295 (1.0081) model_time 0.9294 (1.0048) loss 0.7844 (0.8463) grad_norm 9.9287 (8.5099/1.6550) mem 68106MB [2022-12-19 22:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][290/1519] eta 0:20:38 lr 0.000022 time 0.9327 (1.0078) model_time 0.9325 (1.0046) loss 1.0165 (0.8463) grad_norm 7.3094 (8.5007/1.6401) mem 68106MB [2022-12-19 22:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][300/1519] eta 0:20:28 lr 0.000022 time 0.9312 (1.0076) model_time 0.9310 (1.0044) loss 0.7279 (0.8462) grad_norm 8.0045 (8.5256/1.6998) mem 68106MB [2022-12-19 22:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][310/1519] eta 0:20:17 lr 0.000022 time 0.9298 (1.0074) model_time 0.9297 (1.0043) loss 0.8255 (0.8493) grad_norm 5.8444 (8.5191/1.7046) mem 68106MB [2022-12-19 22:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][320/1519] eta 0:20:07 lr 0.000022 time 0.9375 (1.0074) model_time 0.9373 (1.0044) loss 0.9852 (0.8478) grad_norm 6.2619 (8.4841/1.7242) mem 68106MB [2022-12-19 22:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][330/1519] eta 0:19:58 lr 0.000022 time 0.9260 (1.0077) model_time 0.9258 (1.0047) loss 0.6857 (0.8494) grad_norm 9.7028 (8.5536/1.8039) mem 68106MB [2022-12-19 22:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][340/1519] eta 0:19:47 lr 0.000022 time 0.9362 (1.0075) model_time 0.9361 (1.0046) loss 0.8194 (0.8507) grad_norm 9.1348 (8.5584/1.8066) mem 68106MB [2022-12-19 22:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][350/1519] eta 0:19:38 lr 0.000022 time 0.9428 (1.0082) model_time 0.9426 (1.0055) loss 0.8030 (0.8505) grad_norm 8.3904 (8.5264/1.7976) mem 68106MB [2022-12-19 22:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][360/1519] eta 0:19:28 lr 0.000022 time 0.9121 (1.0085) model_time 0.9119 (1.0058) loss 0.8274 (0.8493) grad_norm 7.4905 (8.5320/1.7978) mem 68106MB [2022-12-19 22:23:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][370/1519] eta 0:19:18 lr 0.000022 time 0.9304 (1.0085) model_time 0.9302 (1.0059) loss 0.9706 (0.8508) grad_norm 7.5255 (8.5330/1.7839) mem 68106MB [2022-12-19 22:23:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][380/1519] eta 0:19:08 lr 0.000022 time 0.9229 (1.0085) model_time 0.9228 (1.0060) loss 0.8336 (0.8514) grad_norm 7.7421 (8.5583/1.8695) mem 68106MB [2022-12-19 22:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][390/1519] eta 0:18:58 lr 0.000022 time 0.9275 (1.0084) model_time 0.9274 (1.0059) loss 0.7550 (0.8514) grad_norm 10.1339 (8.5382/1.8708) mem 68106MB [2022-12-19 22:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][400/1519] eta 0:18:48 lr 0.000022 time 0.9123 (1.0085) model_time 0.9121 (1.0060) loss 0.6954 (0.8508) grad_norm 7.9573 (8.5134/1.8564) mem 68106MB [2022-12-19 22:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][410/1519] eta 0:18:38 lr 0.000022 time 0.9292 (1.0082) model_time 0.9290 (1.0058) loss 0.6891 (0.8514) grad_norm 8.0823 (8.5229/1.8583) mem 68106MB [2022-12-19 22:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][420/1519] eta 0:18:27 lr 0.000022 time 0.9287 (1.0080) model_time 0.9286 (1.0056) loss 0.7530 (0.8512) grad_norm 9.9031 (8.5546/1.8550) mem 68106MB [2022-12-19 22:24:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][430/1519] eta 0:18:17 lr 0.000022 time 0.9246 (1.0078) model_time 0.9242 (1.0055) loss 0.7266 (0.8491) grad_norm 9.8905 (8.5796/1.8867) mem 68106MB [2022-12-19 22:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][440/1519] eta 0:18:07 lr 0.000022 time 0.9382 (1.0076) model_time 0.9380 (1.0054) loss 0.8088 (0.8485) grad_norm 6.3704 (8.5624/1.8810) mem 68106MB [2022-12-19 22:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][450/1519] eta 0:17:57 lr 0.000022 time 0.9302 (1.0077) model_time 0.9301 (1.0055) loss 1.2445 (0.8501) grad_norm 6.9935 (8.5581/1.9289) mem 68106MB [2022-12-19 22:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][460/1519] eta 0:17:46 lr 0.000022 time 0.9264 (1.0075) model_time 0.9263 (1.0053) loss 0.7622 (0.8504) grad_norm 7.6760 (8.5321/1.9246) mem 68106MB [2022-12-19 22:25:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][470/1519] eta 0:17:36 lr 0.000022 time 0.9642 (1.0074) model_time 0.9640 (1.0053) loss 0.8707 (0.8501) grad_norm 6.5553 (8.5287/1.9140) mem 68106MB [2022-12-19 22:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][480/1519] eta 0:17:26 lr 0.000022 time 0.9308 (1.0072) model_time 0.9307 (1.0051) loss 0.7691 (0.8500) grad_norm 9.3002 (8.5207/1.9001) mem 68106MB [2022-12-19 22:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][490/1519] eta 0:17:16 lr 0.000022 time 0.9305 (1.0072) model_time 0.9304 (1.0051) loss 1.2408 (0.8504) grad_norm 7.9559 (8.5315/1.8924) mem 68106MB [2022-12-19 22:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][500/1519] eta 0:17:07 lr 0.000022 time 1.1892 (1.0082) model_time 1.1891 (1.0061) loss 0.6962 (0.8513) grad_norm 6.6743 (8.5121/1.8847) mem 68106MB [2022-12-19 22:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][510/1519] eta 0:16:57 lr 0.000022 time 0.9292 (1.0082) model_time 0.9291 (1.0061) loss 0.9000 (0.8525) grad_norm 9.1437 (8.5144/1.8851) mem 68106MB [2022-12-19 22:25:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][520/1519] eta 0:16:47 lr 0.000022 time 0.9308 (1.0080) model_time 0.9307 (1.0060) loss 0.7339 (0.8525) grad_norm 9.2055 (8.5125/1.8921) mem 68106MB [2022-12-19 22:26:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][530/1519] eta 0:16:36 lr 0.000022 time 0.9322 (1.0079) model_time 0.9320 (1.0059) loss 0.7429 (0.8527) grad_norm 6.7871 (8.4994/1.8855) mem 68106MB [2022-12-19 22:26:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][540/1519] eta 0:16:26 lr 0.000022 time 0.9352 (1.0079) model_time 0.9351 (1.0060) loss 1.1712 (0.8534) grad_norm 12.4279 (8.4971/1.8920) mem 68106MB [2022-12-19 22:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][550/1519] eta 0:16:16 lr 0.000022 time 0.9335 (1.0079) model_time 0.9334 (1.0060) loss 0.7740 (0.8536) grad_norm 7.8290 (8.4969/1.8881) mem 68106MB [2022-12-19 22:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][560/1519] eta 0:16:06 lr 0.000022 time 0.9287 (1.0077) model_time 0.9286 (1.0058) loss 0.7075 (0.8549) grad_norm 7.2141 (8.4940/1.8848) mem 68106MB [2022-12-19 22:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][570/1519] eta 0:15:56 lr 0.000022 time 0.9318 (1.0075) model_time 0.9316 (1.0057) loss 0.8272 (0.8543) grad_norm 9.0434 (8.4963/1.8862) mem 68106MB [2022-12-19 22:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][580/1519] eta 0:15:46 lr 0.000022 time 0.9306 (1.0075) model_time 0.9305 (1.0056) loss 0.6947 (0.8532) grad_norm 11.0983 (8.4953/1.8890) mem 68106MB [2022-12-19 22:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][590/1519] eta 0:15:35 lr 0.000022 time 0.9327 (1.0075) model_time 0.9325 (1.0057) loss 0.8205 (0.8539) grad_norm 11.1669 (8.5022/1.8961) mem 68106MB [2022-12-19 22:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][600/1519] eta 0:15:25 lr 0.000022 time 0.9337 (1.0073) model_time 0.9336 (1.0055) loss 0.9592 (0.8540) grad_norm 6.0675 (8.4879/1.8959) mem 68106MB [2022-12-19 22:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][610/1519] eta 0:15:15 lr 0.000022 time 0.9329 (1.0072) model_time 0.9328 (1.0054) loss 0.7781 (0.8533) grad_norm 9.5880 (8.4758/1.9038) mem 68106MB [2022-12-19 22:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][620/1519] eta 0:15:05 lr 0.000021 time 0.9256 (1.0070) model_time 0.9255 (1.0053) loss 1.0178 (0.8544) grad_norm 6.8171 (8.4501/1.8756) mem 68106MB [2022-12-19 22:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][630/1519] eta 0:14:55 lr 0.000021 time 0.9306 (1.0069) model_time 0.9305 (1.0052) loss 0.8685 (0.8544) grad_norm 9.8238 (8.4649/1.8604) mem 68106MB [2022-12-19 22:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][640/1519] eta 0:14:45 lr 0.000021 time 0.9299 (1.0069) model_time 0.9296 (1.0052) loss 0.8860 (0.8555) grad_norm 8.5691 (8.4421/1.8566) mem 68106MB [2022-12-19 22:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][650/1519] eta 0:14:34 lr 0.000021 time 1.0229 (1.0069) model_time 1.0227 (1.0052) loss 1.0402 (0.8567) grad_norm 7.2765 (8.3811/1.8136) mem 68106MB [2022-12-19 22:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][660/1519] eta 0:14:24 lr 0.000021 time 0.9341 (1.0069) model_time 0.9339 (1.0052) loss 0.8597 (0.8571) grad_norm 7.4512 (8.3915/1.8145) mem 68106MB [2022-12-19 22:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][670/1519] eta 0:14:14 lr 0.000021 time 0.9517 (1.0068) model_time 0.9516 (1.0051) loss 0.9146 (0.8575) grad_norm 8.7750 (8.3953/1.8227) mem 68106MB [2022-12-19 22:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][680/1519] eta 0:14:04 lr 0.000021 time 0.9835 (1.0068) model_time 0.9833 (1.0052) loss 0.8763 (0.8570) grad_norm 9.4491 (8.3687/1.7831) mem 68106MB [2022-12-19 22:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][690/1519] eta 0:13:54 lr 0.000021 time 0.9407 (1.0068) model_time 0.9402 (1.0052) loss 1.1750 (0.8578) grad_norm 9.6524 (8.3861/1.8071) mem 68106MB [2022-12-19 22:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][700/1519] eta 0:13:44 lr 0.000021 time 0.9314 (1.0068) model_time 0.9313 (1.0052) loss 1.0925 (0.8570) grad_norm 9.9847 (8.3844/1.8007) mem 68106MB [2022-12-19 22:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][710/1519] eta 0:13:34 lr 0.000021 time 0.9137 (1.0067) model_time 0.9136 (1.0052) loss 0.9760 (0.8573) grad_norm 8.4820 (8.3851/1.8123) mem 68106MB [2022-12-19 22:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][720/1519] eta 0:13:24 lr 0.000021 time 0.9326 (1.0067) model_time 0.9324 (1.0051) loss 1.0400 (0.8564) grad_norm 11.0056 (8.4167/1.8765) mem 68106MB [2022-12-19 22:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][730/1519] eta 0:13:14 lr 0.000021 time 0.9347 (1.0066) model_time 0.9346 (1.0051) loss 0.8909 (0.8569) grad_norm 11.0889 (8.4404/1.9126) mem 68106MB [2022-12-19 22:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][740/1519] eta 0:13:04 lr 0.000021 time 0.9328 (1.0065) model_time 0.9327 (1.0050) loss 0.6833 (0.8562) grad_norm 8.2907 (8.4548/1.9146) mem 68106MB [2022-12-19 22:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][750/1519] eta 0:12:53 lr 0.000021 time 0.9384 (1.0065) model_time 0.9383 (1.0050) loss 0.8166 (0.8555) grad_norm 6.6027 (8.4623/1.9152) mem 68106MB [2022-12-19 22:29:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][760/1519] eta 0:12:44 lr 0.000021 time 0.9153 (1.0067) model_time 0.9150 (1.0052) loss 1.3480 (0.8557) grad_norm 9.7444 (8.4627/1.9137) mem 68106MB [2022-12-19 22:30:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][770/1519] eta 0:12:34 lr 0.000021 time 0.9317 (1.0067) model_time 0.9315 (1.0052) loss 0.7511 (0.8563) grad_norm 7.2901 (8.4639/1.9070) mem 68106MB [2022-12-19 22:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][780/1519] eta 0:12:23 lr 0.000021 time 0.9309 (1.0066) model_time 0.9308 (1.0051) loss 0.9399 (0.8561) grad_norm 6.3853 (8.4476/1.9034) mem 68106MB [2022-12-19 22:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][790/1519] eta 0:12:13 lr 0.000021 time 0.9289 (1.0065) model_time 0.9287 (1.0051) loss 0.8456 (0.8559) grad_norm 9.0256 (8.4551/1.9058) mem 68106MB [2022-12-19 22:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][800/1519] eta 0:12:03 lr 0.000021 time 0.9318 (1.0065) model_time 0.9316 (1.0050) loss 0.8939 (0.8558) grad_norm 9.2784 (8.4562/1.8935) mem 68106MB [2022-12-19 22:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][810/1519] eta 0:11:53 lr 0.000021 time 0.9249 (1.0068) model_time 0.9247 (1.0054) loss 0.6799 (0.8552) grad_norm 6.8079 (8.4381/1.8907) mem 68106MB [2022-12-19 22:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][820/1519] eta 0:11:43 lr 0.000021 time 0.9284 (1.0069) model_time 0.9283 (1.0055) loss 0.6882 (0.8550) grad_norm 8.8034 (8.4501/1.9001) mem 68106MB [2022-12-19 22:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][830/1519] eta 0:11:33 lr 0.000021 time 0.9719 (1.0068) model_time 0.9718 (1.0054) loss 1.0663 (0.8558) grad_norm 8.8905 (8.4911/1.9301) mem 68106MB [2022-12-19 22:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][840/1519] eta 0:11:23 lr 0.000021 time 0.9279 (1.0068) model_time 0.9277 (1.0054) loss 0.9918 (0.8558) grad_norm 7.8300 (8.4921/1.9275) mem 68106MB [2022-12-19 22:31:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][850/1519] eta 0:11:13 lr 0.000021 time 0.9105 (1.0068) model_time 0.9103 (1.0054) loss 0.7138 (0.8572) grad_norm 5.9071 (8.4922/1.9307) mem 68106MB [2022-12-19 22:31:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][860/1519] eta 0:11:03 lr 0.000021 time 0.9328 (1.0068) model_time 0.9326 (1.0055) loss 0.7067 (0.8585) grad_norm 8.4125 (8.4917/1.9352) mem 68106MB [2022-12-19 22:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][870/1519] eta 0:10:53 lr 0.000021 time 0.9423 (1.0068) model_time 0.9421 (1.0054) loss 0.7885 (0.8581) grad_norm 7.0010 (8.4886/1.9510) mem 68106MB [2022-12-19 22:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][880/1519] eta 0:10:43 lr 0.000021 time 0.9340 (1.0067) model_time 0.9339 (1.0054) loss 0.9738 (0.8578) grad_norm 7.7877 (8.4896/1.9584) mem 68106MB [2022-12-19 22:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][890/1519] eta 0:10:33 lr 0.000021 time 0.9315 (1.0066) model_time 0.9314 (1.0053) loss 0.8274 (0.8572) grad_norm 7.3870 (8.5058/1.9704) mem 68106MB [2022-12-19 22:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][900/1519] eta 0:10:23 lr 0.000021 time 0.9299 (1.0067) model_time 0.9296 (1.0054) loss 0.7611 (0.8570) grad_norm 7.9143 (8.4932/1.9402) mem 68106MB [2022-12-19 22:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][910/1519] eta 0:10:13 lr 0.000021 time 0.9289 (1.0066) model_time 0.9288 (1.0053) loss 0.8135 (0.8575) grad_norm 7.6807 (8.4907/1.9293) mem 68106MB [2022-12-19 22:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][920/1519] eta 0:10:02 lr 0.000021 time 0.9314 (1.0066) model_time 0.9311 (1.0053) loss 1.1170 (0.8575) grad_norm 7.0831 (8.4952/1.9169) mem 68106MB [2022-12-19 22:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][930/1519] eta 0:09:52 lr 0.000021 time 0.9302 (1.0065) model_time 0.9300 (1.0052) loss 0.8727 (0.8572) grad_norm 9.7308 (8.4719/1.8742) mem 68106MB [2022-12-19 22:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][940/1519] eta 0:09:42 lr 0.000021 time 0.9287 (1.0064) model_time 0.9285 (1.0052) loss 0.7727 (0.8564) grad_norm 7.0718 (8.5065/1.9152) mem 68106MB [2022-12-19 22:33:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][950/1519] eta 0:09:32 lr 0.000021 time 0.9354 (1.0065) model_time 0.9353 (1.0052) loss 0.7156 (0.8557) grad_norm 7.8989 (8.5279/1.9088) mem 68106MB [2022-12-19 22:33:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][960/1519] eta 0:09:22 lr 0.000021 time 0.9307 (1.0064) model_time 0.9306 (1.0051) loss 0.8512 (0.8557) grad_norm 6.8129 (8.5204/1.8986) mem 68106MB [2022-12-19 22:33:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][970/1519] eta 0:09:12 lr 0.000021 time 0.9833 (1.0064) model_time 0.9831 (1.0051) loss 0.6781 (0.8561) grad_norm 8.5352 (8.5374/1.9177) mem 68106MB [2022-12-19 22:33:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][980/1519] eta 0:09:02 lr 0.000021 time 0.9299 (1.0063) model_time 0.9297 (1.0050) loss 1.1198 (0.8566) grad_norm 10.1480 (8.5101/1.8618) mem 68106MB [2022-12-19 22:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][990/1519] eta 0:08:52 lr 0.000021 time 0.9299 (1.0063) model_time 0.9298 (1.0051) loss 0.9162 (0.8563) grad_norm 11.5142 (8.5248/1.8560) mem 68106MB [2022-12-19 22:33:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1000/1519] eta 0:08:42 lr 0.000021 time 0.9395 (1.0063) model_time 0.9393 (1.0051) loss 0.7417 (0.8561) grad_norm 9.4457 (8.5470/1.8913) mem 68106MB [2022-12-19 22:34:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1010/1519] eta 0:08:32 lr 0.000021 time 0.9313 (1.0063) model_time 0.9312 (1.0051) loss 0.6960 (0.8564) grad_norm 9.6534 (8.5461/1.8846) mem 68106MB [2022-12-19 22:34:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1020/1519] eta 0:08:22 lr 0.000021 time 0.9358 (1.0063) model_time 0.9357 (1.0051) loss 0.7152 (0.8564) grad_norm 8.2207 (8.5299/1.8742) mem 68106MB [2022-12-19 22:34:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1030/1519] eta 0:08:12 lr 0.000021 time 0.9355 (1.0063) model_time 0.9353 (1.0051) loss 0.6994 (0.8567) grad_norm 6.5768 (8.5101/1.8472) mem 68106MB [2022-12-19 22:34:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1040/1519] eta 0:08:01 lr 0.000021 time 0.9301 (1.0062) model_time 0.9299 (1.0050) loss 0.8747 (0.8567) grad_norm 6.1345 (8.5073/1.8440) mem 68106MB [2022-12-19 22:34:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1050/1519] eta 0:07:51 lr 0.000021 time 0.9504 (1.0062) model_time 0.9503 (1.0050) loss 0.8924 (0.8569) grad_norm 7.8457 (8.5217/1.8023) mem 68106MB [2022-12-19 22:34:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1060/1519] eta 0:07:41 lr 0.000021 time 0.9326 (1.0061) model_time 0.9325 (1.0049) loss 0.9390 (0.8571) grad_norm 7.1300 (8.5300/1.8025) mem 68106MB [2022-12-19 22:35:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1070/1519] eta 0:07:31 lr 0.000021 time 0.9299 (1.0060) model_time 0.9298 (1.0048) loss 0.8035 (0.8567) grad_norm 10.8180 (8.5381/1.8001) mem 68106MB [2022-12-19 22:35:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1080/1519] eta 0:07:21 lr 0.000021 time 0.9263 (1.0060) model_time 0.9261 (1.0049) loss 0.7264 (0.8570) grad_norm 7.1598 (8.5443/1.8095) mem 68106MB [2022-12-19 22:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1090/1519] eta 0:07:11 lr 0.000021 time 0.9359 (1.0060) model_time 0.9357 (1.0048) loss 0.9591 (0.8569) grad_norm 9.4894 (8.5345/1.8014) mem 68106MB [2022-12-19 22:35:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1100/1519] eta 0:07:01 lr 0.000021 time 0.9293 (1.0059) model_time 0.9292 (1.0048) loss 0.7457 (0.8565) grad_norm 7.9380 (8.5481/1.8082) mem 68106MB [2022-12-19 22:35:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1110/1519] eta 0:06:51 lr 0.000021 time 0.9297 (1.0059) model_time 0.9296 (1.0047) loss 0.8731 (0.8560) grad_norm 7.3616 (8.5427/1.7982) mem 68106MB [2022-12-19 22:35:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1120/1519] eta 0:06:41 lr 0.000021 time 0.9283 (1.0059) model_time 0.9281 (1.0048) loss 0.7136 (0.8561) grad_norm 7.3953 (8.5408/1.7826) mem 68106MB [2022-12-19 22:36:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1130/1519] eta 0:06:31 lr 0.000021 time 0.9308 (1.0059) model_time 0.9307 (1.0048) loss 0.7179 (0.8561) grad_norm 9.3637 (8.5642/1.7784) mem 68106MB [2022-12-19 22:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1140/1519] eta 0:06:21 lr 0.000021 time 0.9341 (1.0059) model_time 0.9340 (1.0048) loss 1.1082 (0.8562) grad_norm 10.6498 (8.5704/1.7656) mem 68106MB [2022-12-19 22:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1150/1519] eta 0:06:11 lr 0.000021 time 1.0019 (1.0061) model_time 1.0018 (1.0050) loss 0.8779 (0.8557) grad_norm 7.3016 (8.5492/1.7633) mem 68106MB [2022-12-19 22:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1160/1519] eta 0:06:01 lr 0.000021 time 0.9432 (1.0061) model_time 0.9430 (1.0049) loss 0.6957 (0.8556) grad_norm 8.3039 (8.5588/1.7550) mem 68106MB [2022-12-19 22:36:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1170/1519] eta 0:05:51 lr 0.000021 time 1.1002 (1.0062) model_time 1.1001 (1.0051) loss 0.7073 (0.8553) grad_norm 9.1313 (8.5563/1.7381) mem 68106MB [2022-12-19 22:36:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1180/1519] eta 0:05:41 lr 0.000021 time 0.9377 (1.0062) model_time 0.9376 (1.0051) loss 0.7726 (0.8549) grad_norm 9.9700 (8.5549/1.7219) mem 68106MB [2022-12-19 22:37:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1190/1519] eta 0:05:31 lr 0.000021 time 0.9297 (1.0062) model_time 0.9296 (1.0051) loss 0.6681 (0.8547) grad_norm 7.7393 (8.5460/1.7507) mem 68106MB [2022-12-19 22:37:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1200/1519] eta 0:05:20 lr 0.000021 time 0.9273 (1.0062) model_time 0.9271 (1.0051) loss 1.1971 (0.8556) grad_norm 10.0714 (8.5719/1.7403) mem 68106MB [2022-12-19 22:37:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1210/1519] eta 0:05:10 lr 0.000021 time 1.0337 (1.0064) model_time 1.0335 (1.0053) loss 0.7984 (0.8555) grad_norm 9.4324 (8.5811/1.7307) mem 68106MB [2022-12-19 22:37:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1220/1519] eta 0:05:00 lr 0.000021 time 0.9298 (1.0064) model_time 0.9296 (1.0053) loss 0.7740 (0.8559) grad_norm 10.5558 (8.6011/1.7312) mem 68106MB [2022-12-19 22:37:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1230/1519] eta 0:04:50 lr 0.000021 time 0.9269 (1.0063) model_time 0.9261 (1.0053) loss 0.7191 (0.8553) grad_norm 6.6608 (8.5689/1.7478) mem 68106MB [2022-12-19 22:37:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1240/1519] eta 0:04:40 lr 0.000021 time 0.9362 (1.0063) model_time 0.9361 (1.0052) loss 0.7645 (0.8553) grad_norm 7.3847 (8.5660/1.7647) mem 68106MB [2022-12-19 22:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1250/1519] eta 0:04:30 lr 0.000021 time 0.9322 (1.0062) model_time 0.9321 (1.0052) loss 0.7231 (0.8553) grad_norm 8.1307 (8.5958/1.7595) mem 68106MB [2022-12-19 22:38:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1260/1519] eta 0:04:20 lr 0.000021 time 0.9306 (1.0062) model_time 0.9305 (1.0052) loss 0.7094 (0.8554) grad_norm 14.1388 (8.6071/1.7878) mem 68106MB [2022-12-19 22:38:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1270/1519] eta 0:04:10 lr 0.000021 time 0.9340 (1.0062) model_time 0.9338 (1.0051) loss 0.8083 (0.8561) grad_norm 7.2018 (8.6246/1.7941) mem 68106MB [2022-12-19 22:38:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1280/1519] eta 0:04:00 lr 0.000021 time 0.9337 (1.0061) model_time 0.9336 (1.0051) loss 0.6745 (0.8551) grad_norm 8.4871 (8.6393/1.8104) mem 68106MB [2022-12-19 22:38:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1290/1519] eta 0:03:50 lr 0.000021 time 0.9283 (1.0061) model_time 0.9281 (1.0051) loss 0.8064 (0.8553) grad_norm 8.1761 (8.5946/1.7883) mem 68106MB [2022-12-19 22:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1300/1519] eta 0:03:40 lr 0.000021 time 0.9435 (1.0063) model_time 0.9434 (1.0053) loss 0.7726 (0.8551) grad_norm 9.0090 (8.6056/1.7884) mem 68106MB [2022-12-19 22:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1310/1519] eta 0:03:30 lr 0.000021 time 0.9310 (1.0065) model_time 0.9309 (1.0054) loss 0.8603 (0.8550) grad_norm 7.2744 (8.6119/1.7792) mem 68106MB [2022-12-19 22:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1320/1519] eta 0:03:20 lr 0.000021 time 0.9318 (1.0065) model_time 0.9316 (1.0054) loss 0.9497 (0.8551) grad_norm 8.9022 (8.5701/1.7128) mem 68106MB [2022-12-19 22:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1330/1519] eta 0:03:10 lr 0.000021 time 0.9317 (1.0064) model_time 0.9315 (1.0054) loss 0.7270 (0.8552) grad_norm 7.7194 (8.5456/1.6762) mem 68106MB [2022-12-19 22:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1340/1519] eta 0:03:00 lr 0.000021 time 0.9261 (1.0064) model_time 0.9260 (1.0054) loss 0.7233 (0.8550) grad_norm 9.3595 (8.5302/1.6944) mem 68106MB [2022-12-19 22:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1350/1519] eta 0:02:50 lr 0.000021 time 0.9317 (1.0063) model_time 0.9314 (1.0053) loss 0.7334 (0.8554) grad_norm 11.7728 (8.5362/1.6914) mem 68106MB [2022-12-19 22:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1360/1519] eta 0:02:39 lr 0.000021 time 0.9312 (1.0062) model_time 0.9311 (1.0052) loss 0.8623 (0.8559) grad_norm 10.9888 (8.5506/1.6908) mem 68106MB [2022-12-19 22:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1370/1519] eta 0:02:29 lr 0.000021 time 0.9323 (1.0062) model_time 0.9322 (1.0052) loss 0.9094 (0.8559) grad_norm 6.7078 (8.5449/1.6945) mem 68106MB [2022-12-19 22:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1380/1519] eta 0:02:19 lr 0.000021 time 0.9255 (1.0062) model_time 0.9254 (1.0052) loss 0.7215 (0.8559) grad_norm 9.0730 (8.5480/1.6898) mem 68106MB [2022-12-19 22:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1390/1519] eta 0:02:09 lr 0.000021 time 1.0258 (1.0064) model_time 1.0256 (1.0054) loss 0.7352 (0.8556) grad_norm 8.2532 (8.5363/1.6877) mem 68106MB [2022-12-19 22:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1400/1519] eta 0:01:59 lr 0.000021 time 0.9306 (1.0063) model_time 0.9305 (1.0053) loss 0.9487 (0.8558) grad_norm 7.0977 (8.5315/1.6881) mem 68106MB [2022-12-19 22:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1410/1519] eta 0:01:49 lr 0.000021 time 0.9354 (1.0063) model_time 0.9353 (1.0053) loss 0.7976 (0.8557) grad_norm 9.0916 (8.5721/1.7627) mem 68106MB [2022-12-19 22:40:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1420/1519] eta 0:01:39 lr 0.000021 time 0.9440 (1.0062) model_time 0.9438 (1.0053) loss 0.9063 (0.8556) grad_norm 7.5780 (8.5557/1.7527) mem 68106MB [2022-12-19 22:41:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1430/1519] eta 0:01:29 lr 0.000021 time 0.9753 (1.0062) model_time 0.9751 (1.0052) loss 0.9060 (0.8554) grad_norm 9.8316 (8.5257/1.7155) mem 68106MB [2022-12-19 22:41:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1440/1519] eta 0:01:19 lr 0.000021 time 0.9319 (1.0062) model_time 0.9318 (1.0052) loss 1.0542 (0.8558) grad_norm 6.4011 (8.5302/1.7217) mem 68106MB [2022-12-19 22:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1450/1519] eta 0:01:09 lr 0.000021 time 0.9307 (1.0062) model_time 0.9305 (1.0053) loss 0.8570 (0.8561) grad_norm 8.9989 (8.5684/1.7681) mem 68106MB [2022-12-19 22:41:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1460/1519] eta 0:00:59 lr 0.000021 time 0.9270 (1.0062) model_time 0.9269 (1.0052) loss 0.8477 (0.8567) grad_norm 7.6049 (8.5537/1.7669) mem 68106MB [2022-12-19 22:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1470/1519] eta 0:00:49 lr 0.000021 time 0.9385 (1.0062) model_time 0.9384 (1.0052) loss 0.8427 (0.8562) grad_norm 10.1535 (8.5661/1.7471) mem 68106MB [2022-12-19 22:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1480/1519] eta 0:00:39 lr 0.000021 time 0.9293 (1.0063) model_time 0.9292 (1.0053) loss 1.1513 (0.8561) grad_norm 10.0040 (8.5650/1.7425) mem 68106MB [2022-12-19 22:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1490/1519] eta 0:00:29 lr 0.000021 time 0.9338 (1.0063) model_time 0.9337 (1.0053) loss 0.9396 (0.8563) grad_norm 6.5962 (8.5484/1.7308) mem 68106MB [2022-12-19 22:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1500/1519] eta 0:00:19 lr 0.000021 time 0.9476 (1.0063) model_time 0.9474 (1.0054) loss 0.6985 (0.8560) grad_norm 9.3739 (8.5692/1.7358) mem 68106MB [2022-12-19 22:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [41/100][1510/1519] eta 0:00:09 lr 0.000021 time 0.9375 (1.0063) model_time 0.9374 (1.0054) loss 0.8083 (0.8563) grad_norm 12.9106 (8.5904/1.7545) mem 68106MB [2022-12-19 22:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 41 training takes 0:25:28 [2022-12-19 22:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_41.pth saving...... [2022-12-19 22:43:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_41.pth saved !!! [2022-12-19 22:43:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.658 (0.658) Loss 0.5018 (0.5018) Acc@1 90.625 (90.625) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 22:43:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.332) Loss 0.5046 (0.4832) Acc@1 92.014 (91.919) Acc@5 98.264 (98.580) Mem 68106MB [2022-12-19 22:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4470 (0.4807) Acc@1 92.361 (92.130) Acc@5 98.611 (98.479) Mem 68106MB [2022-12-19 22:43:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.300 (0.310) Loss 0.6026 (0.4875) Acc@1 88.889 (91.891) Acc@5 97.917 (98.398) Mem 68106MB [2022-12-19 22:43:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4842 (0.4819) Acc@1 91.667 (91.836) Acc@5 98.264 (98.450) Mem 68106MB [2022-12-19 22:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.305) Loss 0.4600 (0.4789) Acc@1 91.319 (91.905) Acc@5 99.653 (98.482) Mem 68106MB [2022-12-19 22:43:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.4887 (0.4786) Acc@1 90.625 (91.929) Acc@5 97.917 (98.475) Mem 68106MB [2022-12-19 22:43:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.296 (0.303) Loss 0.5044 (0.4796) Acc@1 92.708 (91.926) Acc@5 98.611 (98.460) Mem 68106MB [2022-12-19 22:43:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.4036 (0.4774) Acc@1 93.056 (91.928) Acc@5 98.611 (98.487) Mem 68106MB [2022-12-19 22:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:41] * Acc@1 91.904 Acc@5 98.486 [2022-12-19 22:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.9% [2022-12-19 22:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-19 22:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-19 22:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.90% [2022-12-19 22:43:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][0/1519] eta 0:37:41 lr 0.000021 time 1.4886 (1.4886) model_time 0.9941 (0.9941) loss 0.9069 (0.9069) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 22:44:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][10/1519] eta 0:26:27 lr 0.000021 time 0.9263 (1.0519) model_time 0.9262 (1.0066) loss 0.6696 (0.7863) grad_norm 6.9153 (9.2152/2.9423) mem 68106MB [2022-12-19 22:44:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][20/1519] eta 0:25:37 lr 0.000021 time 0.9239 (1.0258) model_time 0.9237 (1.0019) loss 1.0117 (0.8255) grad_norm 7.0780 (8.6656/2.2336) mem 68106MB [2022-12-19 22:44:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][30/1519] eta 0:25:14 lr 0.000021 time 0.9378 (1.0172) model_time 0.9376 (1.0009) loss 0.8265 (0.8619) grad_norm 10.8793 (8.7892/1.9433) mem 68106MB [2022-12-19 22:44:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][40/1519] eta 0:24:58 lr 0.000021 time 0.9231 (1.0130) model_time 0.9229 (1.0005) loss 0.7564 (0.8816) grad_norm 10.3733 (9.0353/2.4482) mem 68106MB [2022-12-19 22:44:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][50/1519] eta 0:24:43 lr 0.000021 time 0.9279 (1.0101) model_time 0.9277 (1.0000) loss 0.8165 (0.8833) grad_norm 8.1931 (8.9521/2.2715) mem 68106MB [2022-12-19 22:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][60/1519] eta 0:24:31 lr 0.000021 time 0.9349 (1.0084) model_time 0.9347 (1.0000) loss 0.8450 (0.8674) grad_norm 7.6384 (8.5707/2.2618) mem 68106MB [2022-12-19 22:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][70/1519] eta 0:24:19 lr 0.000021 time 0.9573 (1.0073) model_time 0.9571 (0.9999) loss 0.6785 (0.8578) grad_norm 6.1618 (8.4836/2.1399) mem 68106MB [2022-12-19 22:45:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][80/1519] eta 0:24:09 lr 0.000021 time 0.9233 (1.0076) model_time 0.9231 (1.0011) loss 0.8430 (0.8673) grad_norm 9.3884 (8.5178/2.0115) mem 68106MB [2022-12-19 22:45:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][90/1519] eta 0:23:59 lr 0.000021 time 0.9303 (1.0072) model_time 0.9301 (1.0014) loss 0.8872 (0.8674) grad_norm 7.9430 (8.4926/1.9312) mem 68106MB [2022-12-19 22:45:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][100/1519] eta 0:23:50 lr 0.000021 time 0.9041 (1.0084) model_time 0.9039 (1.0031) loss 0.6972 (0.8703) grad_norm 7.7057 (8.3785/1.8667) mem 68106MB [2022-12-19 22:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][110/1519] eta 0:23:39 lr 0.000021 time 0.9322 (1.0077) model_time 0.9319 (1.0028) loss 0.7850 (0.8666) grad_norm 8.0630 (8.4021/1.7994) mem 68106MB [2022-12-19 22:45:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][120/1519] eta 0:23:28 lr 0.000021 time 0.9352 (1.0071) model_time 0.9350 (1.0026) loss 0.9058 (0.8664) grad_norm 10.1935 (8.3724/1.7697) mem 68106MB [2022-12-19 22:46:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][130/1519] eta 0:23:18 lr 0.000021 time 0.9330 (1.0067) model_time 0.9328 (1.0025) loss 0.7930 (0.8688) grad_norm 13.5004 (8.5715/1.8910) mem 68106MB [2022-12-19 22:46:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][140/1519] eta 0:23:08 lr 0.000021 time 0.9332 (1.0070) model_time 0.9330 (1.0031) loss 0.6808 (0.8617) grad_norm 10.4113 (8.5617/1.8440) mem 68106MB [2022-12-19 22:46:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][150/1519] eta 0:23:00 lr 0.000021 time 0.9256 (1.0084) model_time 0.9254 (1.0047) loss 0.8196 (0.8595) grad_norm 11.4551 (8.6321/1.8396) mem 68106MB [2022-12-19 22:46:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][160/1519] eta 0:22:51 lr 0.000021 time 0.9314 (1.0089) model_time 0.9313 (1.0054) loss 1.0395 (0.8564) grad_norm 7.8052 (8.6209/1.8443) mem 68106MB [2022-12-19 22:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][170/1519] eta 0:22:40 lr 0.000021 time 0.9296 (1.0085) model_time 0.9294 (1.0052) loss 0.7605 (0.8554) grad_norm 5.9852 (8.5571/1.8270) mem 68106MB [2022-12-19 22:46:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][180/1519] eta 0:22:29 lr 0.000021 time 0.9331 (1.0079) model_time 0.9329 (1.0047) loss 0.8277 (0.8565) grad_norm 11.0176 (8.5390/1.8165) mem 68106MB [2022-12-19 22:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][190/1519] eta 0:22:19 lr 0.000021 time 0.9318 (1.0080) model_time 0.9314 (1.0050) loss 0.6945 (0.8549) grad_norm 11.9204 (8.6429/1.9155) mem 68106MB [2022-12-19 22:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][200/1519] eta 0:22:08 lr 0.000021 time 0.9314 (1.0075) model_time 0.9312 (1.0046) loss 0.6936 (0.8541) grad_norm 8.0265 (8.7093/1.9135) mem 68106MB [2022-12-19 22:47:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][210/1519] eta 0:21:58 lr 0.000021 time 0.9268 (1.0072) model_time 0.9267 (1.0044) loss 0.8552 (0.8547) grad_norm 8.5751 (8.6753/1.8985) mem 68106MB [2022-12-19 22:47:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][220/1519] eta 0:21:47 lr 0.000021 time 0.9391 (1.0069) model_time 0.9390 (1.0042) loss 1.0045 (0.8551) grad_norm 6.5925 (8.6075/1.8833) mem 68106MB [2022-12-19 22:47:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][230/1519] eta 0:21:37 lr 0.000021 time 0.9437 (1.0066) model_time 0.9434 (1.0040) loss 0.6730 (0.8542) grad_norm 8.4505 (8.5832/1.8666) mem 68106MB [2022-12-19 22:47:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][240/1519] eta 0:21:27 lr 0.000021 time 0.9308 (1.0065) model_time 0.9307 (1.0040) loss 1.0509 (0.8565) grad_norm 8.3738 (8.5556/1.8421) mem 68106MB [2022-12-19 22:48:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][250/1519] eta 0:21:17 lr 0.000021 time 0.9152 (1.0067) model_time 0.9149 (1.0043) loss 0.6833 (0.8537) grad_norm 7.5533 (8.5202/1.8203) mem 68106MB [2022-12-19 22:48:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][260/1519] eta 0:21:07 lr 0.000021 time 0.9336 (1.0066) model_time 0.9335 (1.0043) loss 0.8220 (0.8528) grad_norm 8.6678 (8.5309/1.8113) mem 68106MB [2022-12-19 22:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][270/1519] eta 0:20:57 lr 0.000021 time 0.9296 (1.0065) model_time 0.9294 (1.0042) loss 1.0313 (0.8561) grad_norm 7.7397 (8.5531/1.7957) mem 68106MB [2022-12-19 22:48:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][280/1519] eta 0:20:47 lr 0.000021 time 0.9284 (1.0071) model_time 0.9283 (1.0050) loss 0.7291 (0.8556) grad_norm 6.7771 (8.4911/1.7954) mem 68106MB [2022-12-19 22:48:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][290/1519] eta 0:20:37 lr 0.000021 time 1.0287 (1.0071) model_time 1.0285 (1.0050) loss 1.0499 (0.8538) grad_norm 7.2034 (8.4971/1.8177) mem 68106MB [2022-12-19 22:48:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][300/1519] eta 0:20:27 lr 0.000021 time 0.9327 (1.0069) model_time 0.9325 (1.0048) loss 0.7970 (0.8574) grad_norm 6.6049 (8.4991/1.7973) mem 68106MB [2022-12-19 22:49:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][310/1519] eta 0:20:17 lr 0.000021 time 0.9736 (1.0070) model_time 0.9735 (1.0050) loss 0.7156 (0.8551) grad_norm 9.3790 (8.4871/1.7917) mem 68106MB [2022-12-19 22:49:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][320/1519] eta 0:20:07 lr 0.000021 time 0.9325 (1.0069) model_time 0.9323 (1.0049) loss 0.6838 (0.8583) grad_norm 11.1584 (8.5320/1.8106) mem 68106MB [2022-12-19 22:49:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][330/1519] eta 0:19:57 lr 0.000021 time 0.9339 (1.0068) model_time 0.9338 (1.0049) loss 1.2558 (0.8603) grad_norm 7.8760 (8.5539/1.7953) mem 68106MB [2022-12-19 22:49:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][340/1519] eta 0:19:46 lr 0.000021 time 0.9362 (1.0066) model_time 0.9361 (1.0048) loss 1.1387 (0.8621) grad_norm 7.0579 (8.5893/1.8166) mem 68106MB [2022-12-19 22:49:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][350/1519] eta 0:19:36 lr 0.000021 time 0.9306 (1.0064) model_time 0.9304 (1.0046) loss 1.0513 (0.8627) grad_norm 6.7183 (8.5478/1.8123) mem 68106MB [2022-12-19 22:49:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][360/1519] eta 0:19:26 lr 0.000021 time 0.9328 (1.0062) model_time 0.9327 (1.0045) loss 0.7892 (0.8629) grad_norm 8.6397 (8.5387/1.7916) mem 68106MB [2022-12-19 22:50:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][370/1519] eta 0:19:15 lr 0.000021 time 0.9308 (1.0060) model_time 0.9306 (1.0043) loss 0.7611 (0.8610) grad_norm 7.9563 (8.5220/1.7748) mem 68106MB [2022-12-19 22:50:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][380/1519] eta 0:19:05 lr 0.000021 time 0.9278 (1.0059) model_time 0.9277 (1.0042) loss 0.6867 (0.8591) grad_norm 10.6029 (8.5376/1.7804) mem 68106MB [2022-12-19 22:50:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][390/1519] eta 0:18:55 lr 0.000021 time 0.9906 (1.0061) model_time 0.9904 (1.0044) loss 0.9287 (0.8585) grad_norm 11.1021 (8.5494/1.7742) mem 68106MB [2022-12-19 22:50:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][400/1519] eta 0:18:45 lr 0.000021 time 0.9292 (1.0059) model_time 0.9291 (1.0043) loss 0.8986 (0.8581) grad_norm 10.3178 (8.5508/1.7573) mem 68106MB [2022-12-19 22:50:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][410/1519] eta 0:18:35 lr 0.000021 time 0.9367 (1.0058) model_time 0.9366 (1.0042) loss 0.7364 (0.8572) grad_norm 7.4929 (8.5195/1.7514) mem 68106MB [2022-12-19 22:50:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][420/1519] eta 0:18:25 lr 0.000021 time 0.9334 (1.0059) model_time 0.9333 (1.0043) loss 0.9672 (0.8565) grad_norm 7.2629 (8.5069/1.7508) mem 68106MB [2022-12-19 22:51:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][430/1519] eta 0:18:15 lr 0.000021 time 0.9312 (1.0058) model_time 0.9310 (1.0043) loss 0.9732 (0.8611) grad_norm 8.3566 (8.5061/1.7435) mem 68106MB [2022-12-19 22:51:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][440/1519] eta 0:18:05 lr 0.000021 time 0.9293 (1.0057) model_time 0.9291 (1.0042) loss 1.1494 (0.8606) grad_norm 9.4090 (8.4916/1.7445) mem 68106MB [2022-12-19 22:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][450/1519] eta 0:17:55 lr 0.000021 time 0.9277 (1.0057) model_time 0.9275 (1.0042) loss 1.1781 (0.8634) grad_norm 7.3908 (8.4830/1.7274) mem 68106MB [2022-12-19 22:51:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][460/1519] eta 0:17:45 lr 0.000021 time 0.9458 (1.0059) model_time 0.9457 (1.0045) loss 1.0235 (0.8636) grad_norm 7.8789 (8.4608/1.7174) mem 68106MB [2022-12-19 22:51:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][470/1519] eta 0:17:35 lr 0.000021 time 1.0481 (1.0064) model_time 1.0480 (1.0050) loss 0.8343 (0.8627) grad_norm 11.5008 (8.4828/1.7264) mem 68106MB [2022-12-19 22:51:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][480/1519] eta 0:17:25 lr 0.000021 time 0.9395 (1.0064) model_time 0.9394 (1.0050) loss 1.1822 (0.8638) grad_norm 9.6193 (8.4902/1.7243) mem 68106MB [2022-12-19 22:52:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][490/1519] eta 0:17:15 lr 0.000021 time 0.9807 (1.0064) model_time 0.9806 (1.0050) loss 1.0627 (0.8650) grad_norm 8.1777 (8.5245/1.7811) mem 68106MB [2022-12-19 22:52:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][500/1519] eta 0:17:05 lr 0.000021 time 0.9387 (1.0065) model_time 0.9386 (1.0051) loss 1.2077 (0.8637) grad_norm 9.3922 (8.5194/1.7735) mem 68106MB [2022-12-19 22:52:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][510/1519] eta 0:16:55 lr 0.000021 time 0.9290 (1.0064) model_time 0.9289 (1.0050) loss 1.0381 (0.8643) grad_norm 7.6659 (8.5111/1.7646) mem 68106MB [2022-12-19 22:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][520/1519] eta 0:16:45 lr 0.000021 time 0.9161 (1.0063) model_time 0.9160 (1.0049) loss 0.9334 (0.8632) grad_norm 11.0495 (8.5322/1.7838) mem 68106MB [2022-12-19 22:52:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][530/1519] eta 0:16:35 lr 0.000021 time 0.9325 (1.0062) model_time 0.9324 (1.0048) loss 0.7568 (0.8628) grad_norm 5.8877 (8.5154/1.7776) mem 68106MB [2022-12-19 22:52:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][540/1519] eta 0:16:24 lr 0.000021 time 0.9278 (1.0060) model_time 0.9276 (1.0047) loss 0.8831 (0.8624) grad_norm 9.2263 (8.5192/1.7705) mem 68106MB [2022-12-19 22:53:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][550/1519] eta 0:16:14 lr 0.000021 time 0.9766 (1.0060) model_time 0.9764 (1.0047) loss 0.7506 (0.8621) grad_norm 6.6005 (8.5014/1.7654) mem 68106MB [2022-12-19 22:53:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][560/1519] eta 0:16:04 lr 0.000021 time 0.9259 (1.0060) model_time 0.9257 (1.0047) loss 0.8766 (0.8612) grad_norm 7.4093 (8.4942/1.7585) mem 68106MB [2022-12-19 22:53:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][570/1519] eta 0:15:55 lr 0.000021 time 0.9924 (1.0064) model_time 0.9922 (1.0051) loss 0.8285 (0.8614) grad_norm 7.6343 (8.5065/1.7572) mem 68106MB [2022-12-19 22:53:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][580/1519] eta 0:15:44 lr 0.000021 time 0.9265 (1.0063) model_time 0.9263 (1.0051) loss 1.0112 (0.8612) grad_norm 7.2441 (8.5044/1.7493) mem 68106MB [2022-12-19 22:53:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][590/1519] eta 0:15:34 lr 0.000021 time 0.9284 (1.0063) model_time 0.9283 (1.0051) loss 0.7606 (0.8626) grad_norm 11.2824 (8.5119/1.7449) mem 68106MB [2022-12-19 22:53:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][600/1519] eta 0:15:24 lr 0.000021 time 0.9252 (1.0062) model_time 0.9250 (1.0049) loss 0.9211 (0.8634) grad_norm 11.9098 (8.5423/1.7975) mem 68106MB [2022-12-19 22:54:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][610/1519] eta 0:15:14 lr 0.000021 time 0.9277 (1.0061) model_time 0.9275 (1.0049) loss 1.0020 (0.8631) grad_norm 8.6174 (8.5269/1.7603) mem 68106MB [2022-12-19 22:54:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][620/1519] eta 0:15:04 lr 0.000021 time 0.9249 (1.0060) model_time 0.9247 (1.0048) loss 0.7956 (0.8639) grad_norm 6.7773 (8.5248/1.7642) mem 68106MB [2022-12-19 22:54:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][630/1519] eta 0:14:54 lr 0.000021 time 0.9233 (1.0063) model_time 0.9232 (1.0051) loss 0.8044 (0.8636) grad_norm 10.8865 (8.5350/1.7851) mem 68106MB [2022-12-19 22:54:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][640/1519] eta 0:14:44 lr 0.000021 time 0.9279 (1.0064) model_time 0.9278 (1.0052) loss 1.0020 (0.8625) grad_norm 8.6928 (8.4967/1.7306) mem 68106MB [2022-12-19 22:54:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][650/1519] eta 0:14:34 lr 0.000021 time 1.0158 (1.0064) model_time 1.0157 (1.0052) loss 0.7077 (0.8619) grad_norm 6.6536 (8.4705/1.7369) mem 68106MB [2022-12-19 22:54:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][660/1519] eta 0:14:24 lr 0.000021 time 0.9247 (1.0063) model_time 0.9246 (1.0051) loss 0.9001 (0.8612) grad_norm 9.7041 (8.5030/1.7349) mem 68106MB [2022-12-19 22:55:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][670/1519] eta 0:14:14 lr 0.000021 time 0.9283 (1.0062) model_time 0.9282 (1.0050) loss 1.0467 (0.8612) grad_norm 7.1321 (8.4909/1.7369) mem 68106MB [2022-12-19 22:55:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][680/1519] eta 0:14:04 lr 0.000021 time 0.9233 (1.0062) model_time 0.9231 (1.0051) loss 0.9556 (0.8613) grad_norm 5.5044 (8.4906/1.7684) mem 68106MB [2022-12-19 22:55:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][690/1519] eta 0:13:54 lr 0.000021 time 0.9275 (1.0061) model_time 0.9274 (1.0050) loss 0.7802 (0.8601) grad_norm 11.0390 (8.5015/1.7867) mem 68106MB [2022-12-19 22:55:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][700/1519] eta 0:13:43 lr 0.000021 time 0.9261 (1.0061) model_time 0.9259 (1.0050) loss 0.8837 (0.8604) grad_norm 7.2063 (8.5213/1.8059) mem 68106MB [2022-12-19 22:55:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][710/1519] eta 0:13:33 lr 0.000021 time 0.9262 (1.0060) model_time 0.9261 (1.0049) loss 0.8367 (0.8598) grad_norm 7.2026 (8.4989/1.8176) mem 68106MB [2022-12-19 22:55:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][720/1519] eta 0:13:23 lr 0.000021 time 0.9575 (1.0060) model_time 0.9573 (1.0049) loss 0.9258 (0.8601) grad_norm 6.5310 (8.4750/1.8297) mem 68106MB [2022-12-19 22:56:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][730/1519] eta 0:13:13 lr 0.000021 time 0.9285 (1.0060) model_time 0.9284 (1.0049) loss 1.0070 (0.8613) grad_norm 10.5311 (8.4369/1.7969) mem 68106MB [2022-12-19 22:56:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][740/1519] eta 0:13:03 lr 0.000021 time 0.9266 (1.0060) model_time 0.9264 (1.0050) loss 0.7405 (0.8623) grad_norm 6.5937 (8.4395/1.7991) mem 68106MB [2022-12-19 22:56:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][750/1519] eta 0:12:53 lr 0.000021 time 0.9380 (1.0060) model_time 0.9378 (1.0049) loss 1.1102 (0.8623) grad_norm 11.9824 (8.5135/1.9851) mem 68106MB [2022-12-19 22:56:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][760/1519] eta 0:12:43 lr 0.000021 time 0.9305 (1.0059) model_time 0.9304 (1.0049) loss 0.9222 (0.8622) grad_norm 8.0946 (8.5210/1.9998) mem 68106MB [2022-12-19 22:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][770/1519] eta 0:12:33 lr 0.000021 time 0.9262 (1.0059) model_time 0.9261 (1.0049) loss 0.7288 (0.8620) grad_norm 9.0260 (8.5685/2.0604) mem 68106MB [2022-12-19 22:57:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][780/1519] eta 0:12:23 lr 0.000021 time 0.9481 (1.0059) model_time 0.9479 (1.0049) loss 0.8300 (0.8617) grad_norm 7.0548 (8.5653/2.0570) mem 68106MB [2022-12-19 22:57:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][790/1519] eta 0:12:13 lr 0.000021 time 0.9204 (1.0060) model_time 0.9203 (1.0049) loss 0.7886 (0.8612) grad_norm 8.3022 (8.5277/2.0197) mem 68106MB [2022-12-19 22:57:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][800/1519] eta 0:12:03 lr 0.000021 time 0.9169 (1.0060) model_time 0.9167 (1.0050) loss 0.7441 (0.8604) grad_norm 9.0936 (8.5368/2.0337) mem 68106MB [2022-12-19 22:57:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][810/1519] eta 0:11:53 lr 0.000021 time 0.9176 (1.0060) model_time 0.9175 (1.0050) loss 0.8491 (0.8599) grad_norm 8.0899 (8.5352/2.0317) mem 68106MB [2022-12-19 22:57:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][820/1519] eta 0:11:43 lr 0.000021 time 0.9373 (1.0059) model_time 0.9371 (1.0049) loss 0.7533 (0.8593) grad_norm 11.0911 (8.5564/2.0371) mem 68106MB [2022-12-19 22:57:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][830/1519] eta 0:11:33 lr 0.000021 time 0.9407 (1.0058) model_time 0.9406 (1.0048) loss 0.8226 (0.8592) grad_norm 7.8942 (8.5559/2.0357) mem 68106MB [2022-12-19 22:58:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][840/1519] eta 0:11:22 lr 0.000021 time 0.9273 (1.0057) model_time 0.9272 (1.0047) loss 0.7119 (0.8591) grad_norm 9.5472 (8.5679/2.0366) mem 68106MB [2022-12-19 22:58:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][850/1519] eta 0:11:12 lr 0.000021 time 0.9282 (1.0057) model_time 0.9279 (1.0047) loss 0.8607 (0.8584) grad_norm 9.3864 (8.5780/2.0379) mem 68106MB [2022-12-19 22:58:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][860/1519] eta 0:11:02 lr 0.000021 time 0.9301 (1.0056) model_time 0.9300 (1.0046) loss 0.9252 (0.8589) grad_norm 7.7848 (8.5611/2.0324) mem 68106MB [2022-12-19 22:58:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][870/1519] eta 0:10:52 lr 0.000021 time 0.9312 (1.0056) model_time 0.9309 (1.0046) loss 0.6771 (0.8580) grad_norm 11.0097 (8.5540/2.0392) mem 68106MB [2022-12-19 22:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][880/1519] eta 0:10:42 lr 0.000021 time 0.9689 (1.0056) model_time 0.9687 (1.0046) loss 1.0342 (0.8577) grad_norm 11.0380 (8.5931/2.0353) mem 68106MB [2022-12-19 22:58:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][890/1519] eta 0:10:32 lr 0.000021 time 0.9318 (1.0058) model_time 0.9317 (1.0048) loss 0.8649 (0.8568) grad_norm 6.2709 (8.5753/2.0221) mem 68106MB [2022-12-19 22:59:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][900/1519] eta 0:10:22 lr 0.000021 time 0.9217 (1.0059) model_time 0.9215 (1.0050) loss 0.6839 (0.8560) grad_norm 7.0763 (8.5679/2.0250) mem 68106MB [2022-12-19 22:59:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][910/1519] eta 0:10:12 lr 0.000021 time 0.9205 (1.0060) model_time 0.9204 (1.0050) loss 1.2365 (0.8556) grad_norm 13.3175 (8.5843/2.0338) mem 68106MB [2022-12-19 22:59:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][920/1519] eta 0:10:02 lr 0.000021 time 0.9293 (1.0059) model_time 0.9292 (1.0050) loss 1.2143 (0.8559) grad_norm 6.1385 (8.5375/2.0325) mem 68106MB [2022-12-19 22:59:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][930/1519] eta 0:09:52 lr 0.000021 time 0.9309 (1.0058) model_time 0.9308 (1.0049) loss 1.1130 (0.8568) grad_norm 8.8943 (8.5421/2.0409) mem 68106MB [2022-12-19 22:59:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][940/1519] eta 0:09:42 lr 0.000021 time 0.9354 (1.0058) model_time 0.9351 (1.0049) loss 0.7259 (0.8564) grad_norm 8.3473 (8.5002/2.0250) mem 68106MB [2022-12-19 22:59:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][950/1519] eta 0:09:32 lr 0.000021 time 0.9376 (1.0059) model_time 0.9375 (1.0050) loss 1.3724 (0.8570) grad_norm 8.0511 (8.4944/2.0287) mem 68106MB [2022-12-19 23:00:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][960/1519] eta 0:09:22 lr 0.000021 time 0.9303 (1.0058) model_time 0.9301 (1.0049) loss 0.7488 (0.8573) grad_norm 8.6610 (8.5151/2.0461) mem 68106MB [2022-12-19 23:00:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][970/1519] eta 0:09:12 lr 0.000021 time 0.9197 (1.0058) model_time 0.9196 (1.0049) loss 0.6909 (0.8576) grad_norm 9.6673 (8.5272/2.0502) mem 68106MB [2022-12-19 23:00:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][980/1519] eta 0:09:02 lr 0.000021 time 0.9238 (1.0057) model_time 0.9237 (1.0048) loss 0.9073 (0.8571) grad_norm 8.0068 (8.5142/2.0406) mem 68106MB [2022-12-19 23:00:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][990/1519] eta 0:08:52 lr 0.000021 time 0.9278 (1.0057) model_time 0.9276 (1.0048) loss 0.8299 (0.8584) grad_norm 8.0464 (8.4959/2.0398) mem 68106MB [2022-12-19 23:00:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1000/1519] eta 0:08:41 lr 0.000021 time 0.9194 (1.0057) model_time 0.9193 (1.0048) loss 0.8497 (0.8583) grad_norm 10.0712 (8.5159/2.0502) mem 68106MB [2022-12-19 23:00:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1010/1519] eta 0:08:31 lr 0.000021 time 0.9334 (1.0056) model_time 0.9332 (1.0047) loss 0.8906 (0.8581) grad_norm 7.7423 (8.5216/2.0488) mem 68106MB [2022-12-19 23:01:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1020/1519] eta 0:08:21 lr 0.000021 time 0.9250 (1.0055) model_time 0.9248 (1.0046) loss 0.7863 (0.8578) grad_norm 7.3421 (8.5294/2.0393) mem 68106MB [2022-12-19 23:01:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1030/1519] eta 0:08:11 lr 0.000021 time 0.9221 (1.0055) model_time 0.9218 (1.0046) loss 0.9054 (0.8581) grad_norm 7.9064 (8.5256/2.0341) mem 68106MB [2022-12-19 23:01:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1040/1519] eta 0:08:01 lr 0.000021 time 0.9265 (1.0054) model_time 0.9262 (1.0045) loss 0.9062 (0.8581) grad_norm 7.6776 (8.5151/2.0303) mem 68106MB [2022-12-19 23:01:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1050/1519] eta 0:07:51 lr 0.000021 time 0.9201 (1.0054) model_time 0.9200 (1.0046) loss 0.9743 (0.8583) grad_norm 10.8526 (8.5132/2.0412) mem 68106MB [2022-12-19 23:01:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1060/1519] eta 0:07:41 lr 0.000021 time 1.0154 (1.0055) model_time 1.0152 (1.0047) loss 0.8467 (0.8586) grad_norm 8.5621 (8.5344/2.0493) mem 68106MB [2022-12-19 23:01:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1070/1519] eta 0:07:31 lr 0.000021 time 0.9584 (1.0055) model_time 0.9582 (1.0046) loss 0.7163 (0.8585) grad_norm 8.7182 (8.5218/2.0458) mem 68106MB [2022-12-19 23:02:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1080/1519] eta 0:07:21 lr 0.000021 time 0.9336 (1.0055) model_time 0.9332 (1.0047) loss 0.7556 (0.8590) grad_norm 9.4477 (8.5185/2.0429) mem 68106MB [2022-12-19 23:02:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1090/1519] eta 0:07:11 lr 0.000021 time 0.9264 (1.0055) model_time 0.9262 (1.0046) loss 0.8460 (0.8591) grad_norm 7.6389 (8.5007/2.0194) mem 68106MB [2022-12-19 23:02:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1100/1519] eta 0:07:01 lr 0.000021 time 0.9316 (1.0057) model_time 0.9313 (1.0049) loss 0.8871 (0.8592) grad_norm 9.9813 (8.5081/2.0155) mem 68106MB [2022-12-19 23:02:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1110/1519] eta 0:06:51 lr 0.000021 time 0.9321 (1.0057) model_time 0.9319 (1.0048) loss 0.9181 (0.8589) grad_norm 9.3780 (8.5420/2.0961) mem 68106MB [2022-12-19 23:02:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1120/1519] eta 0:06:41 lr 0.000021 time 0.9310 (1.0057) model_time 0.9308 (1.0048) loss 1.1848 (0.8593) grad_norm 7.0542 (8.4923/2.0862) mem 68106MB [2022-12-19 23:02:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1130/1519] eta 0:06:31 lr 0.000021 time 0.9208 (1.0057) model_time 0.9202 (1.0049) loss 0.6897 (0.8597) grad_norm 5.4347 (8.4917/2.0891) mem 68106MB [2022-12-19 23:03:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1140/1519] eta 0:06:21 lr 0.000021 time 0.9232 (1.0056) model_time 0.9230 (1.0048) loss 1.1267 (0.8600) grad_norm 9.4564 (8.4779/2.0866) mem 68106MB [2022-12-19 23:03:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1150/1519] eta 0:06:11 lr 0.000021 time 0.9216 (1.0056) model_time 0.9215 (1.0047) loss 0.7274 (0.8600) grad_norm 9.3619 (8.5076/2.0846) mem 68106MB [2022-12-19 23:03:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1160/1519] eta 0:06:00 lr 0.000021 time 0.9271 (1.0055) model_time 0.9270 (1.0047) loss 0.8479 (0.8605) grad_norm 8.2595 (8.5214/2.0874) mem 68106MB [2022-12-19 23:03:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1170/1519] eta 0:05:50 lr 0.000021 time 0.9263 (1.0054) model_time 0.9262 (1.0046) loss 0.9331 (0.8610) grad_norm 9.3941 (8.5007/2.0831) mem 68106MB [2022-12-19 23:03:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1180/1519] eta 0:05:40 lr 0.000021 time 0.9176 (1.0054) model_time 0.9174 (1.0046) loss 0.7773 (0.8609) grad_norm 18.8523 (8.5333/2.1632) mem 68106MB [2022-12-19 23:03:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1190/1519] eta 0:05:30 lr 0.000021 time 0.9371 (1.0053) model_time 0.9369 (1.0045) loss 1.0858 (0.8611) grad_norm 8.4487 (8.5306/2.1610) mem 68106MB [2022-12-19 23:04:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1200/1519] eta 0:05:20 lr 0.000021 time 0.9751 (1.0055) model_time 0.9750 (1.0047) loss 1.1494 (0.8619) grad_norm 10.5474 (8.5045/2.1127) mem 68106MB [2022-12-19 23:04:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1210/1519] eta 0:05:10 lr 0.000021 time 0.9752 (1.0055) model_time 0.9749 (1.0047) loss 0.8640 (0.8627) grad_norm 8.3637 (8.5146/2.1264) mem 68106MB [2022-12-19 23:04:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1220/1519] eta 0:05:00 lr 0.000021 time 0.9194 (1.0055) model_time 0.9193 (1.0047) loss 0.7747 (0.8631) grad_norm 8.0452 (8.5203/2.1290) mem 68106MB [2022-12-19 23:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1230/1519] eta 0:04:50 lr 0.000021 time 0.9339 (1.0055) model_time 0.9337 (1.0047) loss 0.7355 (0.8624) grad_norm 9.3131 (8.5111/2.1129) mem 68106MB [2022-12-19 23:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1240/1519] eta 0:04:40 lr 0.000021 time 0.9421 (1.0055) model_time 0.9419 (1.0047) loss 1.1619 (0.8626) grad_norm 8.8486 (8.5255/2.1087) mem 68106MB [2022-12-19 23:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1250/1519] eta 0:04:30 lr 0.000021 time 0.9551 (1.0054) model_time 0.9549 (1.0046) loss 0.8440 (0.8630) grad_norm 6.6959 (8.5348/2.1036) mem 68106MB [2022-12-19 23:05:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1260/1519] eta 0:04:20 lr 0.000021 time 0.8872 (1.0054) model_time 0.8870 (1.0047) loss 0.8446 (0.8626) grad_norm 6.1013 (8.5220/2.1154) mem 68106MB [2022-12-19 23:05:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1270/1519] eta 0:04:10 lr 0.000021 time 0.9246 (1.0054) model_time 0.9245 (1.0046) loss 0.9168 (0.8629) grad_norm 8.7046 (8.5361/2.1104) mem 68106MB [2022-12-19 23:05:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1280/1519] eta 0:04:00 lr 0.000021 time 0.9230 (1.0057) model_time 0.9229 (1.0049) loss 1.1904 (0.8628) grad_norm 8.2634 (8.5276/2.0918) mem 68106MB [2022-12-19 23:05:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1290/1519] eta 0:03:50 lr 0.000021 time 0.9227 (1.0056) model_time 0.9225 (1.0049) loss 0.9287 (0.8630) grad_norm 8.7353 (8.5384/2.0899) mem 68106MB [2022-12-19 23:05:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1300/1519] eta 0:03:40 lr 0.000021 time 0.9277 (1.0056) model_time 0.9276 (1.0048) loss 0.7661 (0.8626) grad_norm 18.5651 (8.5699/2.1488) mem 68106MB [2022-12-19 23:05:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1310/1519] eta 0:03:30 lr 0.000021 time 0.9238 (1.0056) model_time 0.9236 (1.0049) loss 0.8001 (0.8621) grad_norm 6.8083 (8.5776/2.1489) mem 68106MB [2022-12-19 23:06:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1320/1519] eta 0:03:20 lr 0.000021 time 0.9289 (1.0056) model_time 0.9286 (1.0048) loss 0.9512 (0.8623) grad_norm 11.4461 (8.6294/2.1647) mem 68106MB [2022-12-19 23:06:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1330/1519] eta 0:03:10 lr 0.000021 time 0.9384 (1.0055) model_time 0.9382 (1.0048) loss 0.7299 (0.8618) grad_norm 9.0201 (8.6308/2.1597) mem 68106MB [2022-12-19 23:06:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1340/1519] eta 0:02:59 lr 0.000021 time 0.9210 (1.0055) model_time 0.9208 (1.0047) loss 0.7698 (0.8618) grad_norm 6.0963 (8.6480/2.1866) mem 68106MB [2022-12-19 23:06:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1350/1519] eta 0:02:49 lr 0.000021 time 0.9247 (1.0054) model_time 0.9246 (1.0047) loss 1.1310 (0.8617) grad_norm 7.9233 (8.5501/2.0138) mem 68106MB [2022-12-19 23:06:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1360/1519] eta 0:02:39 lr 0.000021 time 0.9266 (1.0054) model_time 0.9264 (1.0047) loss 0.8878 (0.8621) grad_norm 9.0322 (8.5390/1.9864) mem 68106MB [2022-12-19 23:06:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1370/1519] eta 0:02:29 lr 0.000021 time 0.9149 (1.0055) model_time 0.9148 (1.0048) loss 0.8440 (0.8623) grad_norm 9.6255 (8.5108/1.9261) mem 68106MB [2022-12-19 23:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1380/1519] eta 0:02:19 lr 0.000021 time 0.9849 (1.0056) model_time 0.9848 (1.0048) loss 0.6811 (0.8619) grad_norm 10.0144 (8.5184/1.9254) mem 68106MB [2022-12-19 23:07:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1390/1519] eta 0:02:09 lr 0.000021 time 0.9706 (1.0056) model_time 0.9704 (1.0048) loss 0.7285 (0.8621) grad_norm 7.0805 (8.5014/1.9268) mem 68106MB [2022-12-19 23:07:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1400/1519] eta 0:01:59 lr 0.000021 time 0.9283 (1.0055) model_time 0.9281 (1.0048) loss 0.7210 (0.8622) grad_norm 8.2059 (8.4600/1.8963) mem 68106MB [2022-12-19 23:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1410/1519] eta 0:01:49 lr 0.000021 time 0.9187 (1.0056) model_time 0.9186 (1.0048) loss 0.6839 (0.8627) grad_norm 8.6476 (8.4646/1.8918) mem 68106MB [2022-12-19 23:07:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1420/1519] eta 0:01:39 lr 0.000021 time 0.9237 (1.0056) model_time 0.9236 (1.0048) loss 0.9539 (0.8626) grad_norm 10.8233 (8.4573/1.8965) mem 68106MB [2022-12-19 23:07:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1430/1519] eta 0:01:29 lr 0.000021 time 0.9245 (1.0056) model_time 0.9244 (1.0049) loss 0.8448 (0.8622) grad_norm 9.1212 (8.4679/1.8973) mem 68106MB [2022-12-19 23:08:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1440/1519] eta 0:01:19 lr 0.000021 time 0.9096 (1.0056) model_time 0.9094 (1.0049) loss 0.9065 (0.8621) grad_norm 8.1449 (8.4517/1.8946) mem 68106MB [2022-12-19 23:08:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1450/1519] eta 0:01:09 lr 0.000021 time 0.9231 (1.0056) model_time 0.9230 (1.0048) loss 0.7175 (0.8619) grad_norm 7.6541 (8.4651/1.8930) mem 68106MB [2022-12-19 23:08:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1460/1519] eta 0:00:59 lr 0.000021 time 0.9221 (1.0055) model_time 0.9220 (1.0048) loss 1.2068 (0.8625) grad_norm 6.5888 (8.4631/1.8975) mem 68106MB [2022-12-19 23:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1470/1519] eta 0:00:49 lr 0.000021 time 0.9376 (1.0055) model_time 0.9373 (1.0048) loss 1.0401 (0.8629) grad_norm 8.1230 (8.4772/1.8994) mem 68106MB [2022-12-19 23:08:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1480/1519] eta 0:00:39 lr 0.000021 time 0.9253 (1.0055) model_time 0.9252 (1.0047) loss 1.3417 (0.8633) grad_norm 8.0397 (8.4440/1.8983) mem 68106MB [2022-12-19 23:08:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1490/1519] eta 0:00:29 lr 0.000021 time 0.9243 (1.0054) model_time 0.9240 (1.0047) loss 1.0553 (0.8630) grad_norm 11.2193 (8.4534/1.9012) mem 68106MB [2022-12-19 23:09:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1500/1519] eta 0:00:19 lr 0.000021 time 0.9160 (1.0053) model_time 0.9158 (1.0046) loss 1.1268 (0.8636) grad_norm 7.5237 (8.4599/1.8972) mem 68106MB [2022-12-19 23:09:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [42/100][1510/1519] eta 0:00:09 lr 0.000021 time 0.9193 (1.0054) model_time 0.9192 (1.0047) loss 0.8545 (0.8635) grad_norm 8.0869 (8.4324/1.8801) mem 68106MB [2022-12-19 23:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 42 training takes 0:25:27 [2022-12-19 23:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_42.pth saving...... [2022-12-19 23:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_42.pth saved !!! [2022-12-19 23:09:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.670 (0.670) Loss 0.5245 (0.5245) Acc@1 91.667 (91.667) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-19 23:09:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.333) Loss 0.5100 (0.4893) Acc@1 91.667 (92.172) Acc@5 97.569 (98.390) Mem 68106MB [2022-12-19 23:09:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.315) Loss 0.4638 (0.4872) Acc@1 92.014 (92.146) Acc@5 98.264 (98.380) Mem 68106MB [2022-12-19 23:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.310) Loss 0.6154 (0.4939) Acc@1 89.931 (91.924) Acc@5 97.569 (98.342) Mem 68106MB [2022-12-19 23:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.307) Loss 0.4800 (0.4859) Acc@1 90.972 (91.929) Acc@5 98.611 (98.425) Mem 68106MB [2022-12-19 23:10:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.305 (0.305) Loss 0.4866 (0.4840) Acc@1 90.972 (91.973) Acc@5 99.653 (98.475) Mem 68106MB [2022-12-19 23:10:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.4812 (0.4836) Acc@1 90.972 (91.911) Acc@5 98.264 (98.457) Mem 68106MB [2022-12-19 23:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5115 (0.4834) Acc@1 91.667 (91.882) Acc@5 98.611 (98.450) Mem 68106MB [2022-12-19 23:10:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.302) Loss 0.3994 (0.4810) Acc@1 93.056 (91.877) Acc@5 98.958 (98.491) Mem 68106MB [2022-12-19 23:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:42] * Acc@1 91.847 Acc@5 98.494 [2022-12-19 23:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.8% [2022-12-19 23:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.90% [2022-12-19 23:10:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][0/1519] eta 0:45:20 lr 0.000021 time 1.7912 (1.7912) model_time 1.0584 (1.0584) loss 0.7019 (0.7019) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 23:10:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][10/1519] eta 0:27:04 lr 0.000021 time 0.9323 (1.0765) model_time 0.9321 (1.0092) loss 0.7144 (0.9141) grad_norm 9.3328 (9.3259/1.0580) mem 68106MB [2022-12-19 23:10:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][20/1519] eta 0:26:01 lr 0.000021 time 0.9252 (1.0420) model_time 0.9250 (1.0065) loss 0.9602 (0.9156) grad_norm 9.0964 (9.1657/1.8744) mem 68106MB [2022-12-19 23:10:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][30/1519] eta 0:25:31 lr 0.000021 time 0.9367 (1.0284) model_time 0.9366 (1.0042) loss 0.7067 (0.9024) grad_norm 9.2477 (9.4028/1.8490) mem 68106MB [2022-12-19 23:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][40/1519] eta 0:25:11 lr 0.000021 time 0.9337 (1.0219) model_time 0.9335 (1.0034) loss 0.8056 (0.8972) grad_norm 9.2077 (9.4469/1.7199) mem 68106MB [2022-12-19 23:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][50/1519] eta 0:24:58 lr 0.000021 time 0.9925 (1.0200) model_time 0.9924 (1.0051) loss 0.6922 (0.8860) grad_norm 9.1767 (9.3326/1.6267) mem 68106MB [2022-12-19 23:11:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][60/1519] eta 0:24:41 lr 0.000021 time 0.9713 (1.0155) model_time 0.9711 (1.0030) loss 0.8777 (0.8848) grad_norm inf (9.0339/1.7286) mem 68106MB [2022-12-19 23:11:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][70/1519] eta 0:24:32 lr 0.000021 time 0.9055 (1.0159) model_time 0.9054 (1.0051) loss 0.6982 (0.8826) grad_norm 7.7716 (9.2060/1.8355) mem 68106MB [2022-12-19 23:11:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][80/1519] eta 0:24:19 lr 0.000021 time 0.9340 (1.0139) model_time 0.9339 (1.0044) loss 0.9125 (0.8739) grad_norm 8.5901 (9.0870/1.7515) mem 68106MB [2022-12-19 23:11:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][90/1519] eta 0:24:07 lr 0.000021 time 0.9239 (1.0129) model_time 0.9238 (1.0044) loss 1.2060 (0.8725) grad_norm 7.9133 (8.9065/1.7770) mem 68106MB [2022-12-19 23:11:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][100/1519] eta 0:23:58 lr 0.000021 time 0.9254 (1.0135) model_time 0.9253 (1.0058) loss 0.7485 (0.8721) grad_norm 6.1536 (8.7195/1.7833) mem 68106MB [2022-12-19 23:12:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][110/1519] eta 0:23:45 lr 0.000021 time 0.9221 (1.0119) model_time 0.9220 (1.0049) loss 0.9062 (0.8750) grad_norm 9.7554 (8.6507/1.7363) mem 68106MB [2022-12-19 23:12:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][120/1519] eta 0:23:36 lr 0.000021 time 0.9184 (1.0128) model_time 0.9182 (1.0063) loss 0.7667 (0.8708) grad_norm 7.1328 (8.6101/1.6879) mem 68106MB [2022-12-19 23:12:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][130/1519] eta 0:23:26 lr 0.000021 time 0.9704 (1.0130) model_time 0.9703 (1.0069) loss 0.8803 (0.8726) grad_norm 12.0680 (8.8538/2.2252) mem 68106MB [2022-12-19 23:12:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][140/1519] eta 0:23:15 lr 0.000021 time 0.9229 (1.0121) model_time 0.9227 (1.0064) loss 0.9238 (0.8798) grad_norm 7.8900 (8.7221/2.2158) mem 68106MB [2022-12-19 23:12:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][150/1519] eta 0:23:04 lr 0.000021 time 0.9446 (1.0113) model_time 0.9445 (1.0060) loss 1.2327 (0.8815) grad_norm 7.0466 (8.6639/2.1709) mem 68106MB [2022-12-19 23:12:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][160/1519] eta 0:22:54 lr 0.000021 time 0.9225 (1.0113) model_time 0.9223 (1.0064) loss 0.8441 (0.8782) grad_norm 7.8194 (8.5704/2.1366) mem 68106MB [2022-12-19 23:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][170/1519] eta 0:22:43 lr 0.000021 time 0.9257 (1.0107) model_time 0.9256 (1.0060) loss 1.0854 (0.8770) grad_norm 6.0325 (8.5103/2.1053) mem 68106MB [2022-12-19 23:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][180/1519] eta 0:22:32 lr 0.000021 time 0.9213 (1.0102) model_time 0.9212 (1.0058) loss 0.7503 (0.8741) grad_norm 6.9299 (8.4693/2.0646) mem 68106MB [2022-12-19 23:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][190/1519] eta 0:22:22 lr 0.000021 time 0.9208 (1.0099) model_time 0.9207 (1.0057) loss 1.0995 (0.8753) grad_norm 8.5298 (8.4620/2.0335) mem 68106MB [2022-12-19 23:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][200/1519] eta 0:22:11 lr 0.000021 time 0.9205 (1.0093) model_time 0.9204 (1.0053) loss 0.7404 (0.8788) grad_norm 8.5348 (8.4531/2.0012) mem 68106MB [2022-12-19 23:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][210/1519] eta 0:22:00 lr 0.000021 time 0.9301 (1.0090) model_time 0.9299 (1.0051) loss 0.8811 (0.8789) grad_norm 7.7079 (8.4655/1.9668) mem 68106MB [2022-12-19 23:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][220/1519] eta 0:21:50 lr 0.000021 time 0.9185 (1.0086) model_time 0.9184 (1.0049) loss 0.7694 (0.8774) grad_norm 9.3787 (8.5128/1.9730) mem 68106MB [2022-12-19 23:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][230/1519] eta 0:21:40 lr 0.000021 time 0.9968 (1.0088) model_time 0.9966 (1.0053) loss 0.9550 (0.8772) grad_norm 8.6424 (8.5104/1.9597) mem 68106MB [2022-12-19 23:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][240/1519] eta 0:21:30 lr 0.000021 time 1.0033 (1.0089) model_time 1.0032 (1.0054) loss 0.8385 (0.8776) grad_norm 10.0684 (8.5048/1.9329) mem 68106MB [2022-12-19 23:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][250/1519] eta 0:21:20 lr 0.000021 time 0.9424 (1.0090) model_time 0.9423 (1.0057) loss 0.7626 (0.8753) grad_norm 5.4248 (8.4637/1.9275) mem 68106MB [2022-12-19 23:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][260/1519] eta 0:21:09 lr 0.000021 time 0.9353 (1.0086) model_time 0.9352 (1.0054) loss 1.0094 (0.8778) grad_norm 8.5528 (8.4732/1.9214) mem 68106MB [2022-12-19 23:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][270/1519] eta 0:20:59 lr 0.000021 time 0.9206 (1.0081) model_time 0.9205 (1.0051) loss 0.8047 (0.8768) grad_norm 7.6880 (8.5187/1.9123) mem 68106MB [2022-12-19 23:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][280/1519] eta 0:20:48 lr 0.000021 time 0.9208 (1.0078) model_time 0.9206 (1.0048) loss 0.6968 (0.8766) grad_norm 9.1218 (8.5555/1.8998) mem 68106MB [2022-12-19 23:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][290/1519] eta 0:20:38 lr 0.000021 time 0.9217 (1.0078) model_time 0.9215 (1.0049) loss 0.6961 (0.8737) grad_norm 8.7166 (8.5472/1.9117) mem 68106MB [2022-12-19 23:15:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][300/1519] eta 0:20:28 lr 0.000021 time 0.9232 (1.0077) model_time 0.9231 (1.0049) loss 1.1070 (0.8756) grad_norm 9.4440 (8.5579/1.8832) mem 68106MB [2022-12-19 23:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][310/1519] eta 0:20:18 lr 0.000021 time 0.9313 (1.0075) model_time 0.9312 (1.0048) loss 0.8839 (0.8744) grad_norm 9.2615 (8.5542/1.8605) mem 68106MB [2022-12-19 23:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][320/1519] eta 0:20:07 lr 0.000021 time 0.9305 (1.0075) model_time 0.9304 (1.0048) loss 1.0654 (0.8746) grad_norm 10.1476 (8.5414/1.8627) mem 68106MB [2022-12-19 23:15:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][330/1519] eta 0:19:57 lr 0.000021 time 0.9271 (1.0073) model_time 0.9270 (1.0047) loss 0.7036 (0.8740) grad_norm 9.0973 (8.5258/1.8535) mem 68106MB [2022-12-19 23:15:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][340/1519] eta 0:19:47 lr 0.000021 time 0.9221 (1.0070) model_time 0.9220 (1.0045) loss 0.8589 (0.8729) grad_norm 8.1539 (8.5228/1.8312) mem 68106MB [2022-12-19 23:16:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][350/1519] eta 0:19:36 lr 0.000021 time 0.9295 (1.0067) model_time 0.9294 (1.0043) loss 1.0001 (0.8748) grad_norm 8.7447 (8.5158/1.8104) mem 68106MB [2022-12-19 23:16:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][360/1519] eta 0:19:26 lr 0.000021 time 0.9302 (1.0067) model_time 0.9301 (1.0043) loss 1.0766 (0.8746) grad_norm 8.0574 (8.4892/1.8031) mem 68106MB [2022-12-19 23:16:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][370/1519] eta 0:19:16 lr 0.000021 time 0.9301 (1.0067) model_time 0.9300 (1.0043) loss 0.8053 (0.8754) grad_norm 9.7288 (8.5188/1.8033) mem 68106MB [2022-12-19 23:16:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][380/1519] eta 0:19:06 lr 0.000021 time 0.9291 (1.0064) model_time 0.9290 (1.0042) loss 1.0539 (0.8742) grad_norm 12.9859 (8.5923/1.9267) mem 68106MB [2022-12-19 23:16:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][390/1519] eta 0:18:56 lr 0.000021 time 0.9290 (1.0064) model_time 0.9289 (1.0042) loss 0.8115 (0.8762) grad_norm 9.9918 (8.5523/1.9329) mem 68106MB [2022-12-19 23:16:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][400/1519] eta 0:18:46 lr 0.000021 time 0.9264 (1.0063) model_time 0.9262 (1.0041) loss 0.7528 (0.8762) grad_norm 7.6684 (8.5394/1.9111) mem 68106MB [2022-12-19 23:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][410/1519] eta 0:18:36 lr 0.000021 time 0.9256 (1.0065) model_time 0.9255 (1.0043) loss 0.9002 (0.8777) grad_norm 11.3618 (8.5416/1.9268) mem 68106MB [2022-12-19 23:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][420/1519] eta 0:18:25 lr 0.000021 time 0.9252 (1.0063) model_time 0.9251 (1.0042) loss 0.7397 (0.8779) grad_norm 9.2248 (8.5327/1.9127) mem 68106MB [2022-12-19 23:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][430/1519] eta 0:18:15 lr 0.000021 time 1.0016 (1.0064) model_time 1.0014 (1.0043) loss 0.8210 (0.8759) grad_norm 6.9559 (8.5466/1.9017) mem 68106MB [2022-12-19 23:17:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][440/1519] eta 0:18:05 lr 0.000021 time 0.9178 (1.0064) model_time 0.9176 (1.0044) loss 1.2694 (0.8773) grad_norm 16.2069 (8.5881/1.9527) mem 68106MB [2022-12-19 23:17:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][450/1519] eta 0:17:55 lr 0.000021 time 0.9255 (1.0063) model_time 0.9253 (1.0043) loss 0.9099 (0.8759) grad_norm 14.4931 (8.6073/1.9791) mem 68106MB [2022-12-19 23:17:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][460/1519] eta 0:17:45 lr 0.000021 time 0.9208 (1.0061) model_time 0.9206 (1.0041) loss 0.8217 (0.8741) grad_norm 9.0233 (8.6186/1.9801) mem 68106MB [2022-12-19 23:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][470/1519] eta 0:17:35 lr 0.000021 time 1.1674 (1.0066) model_time 1.1673 (1.0047) loss 0.7228 (0.8726) grad_norm 6.1562 (8.6191/1.9765) mem 68106MB [2022-12-19 23:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][480/1519] eta 0:17:25 lr 0.000021 time 0.9267 (1.0065) model_time 0.9266 (1.0046) loss 0.7279 (0.8722) grad_norm 8.5468 (8.6075/1.9589) mem 68106MB [2022-12-19 23:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][490/1519] eta 0:17:15 lr 0.000021 time 0.9255 (1.0063) model_time 0.9254 (1.0045) loss 0.8858 (0.8723) grad_norm 7.0193 (8.6189/1.9477) mem 68106MB [2022-12-19 23:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][500/1519] eta 0:17:05 lr 0.000021 time 0.9271 (1.0062) model_time 0.9270 (1.0044) loss 0.9676 (0.8712) grad_norm 10.0458 (8.6427/1.9459) mem 68106MB [2022-12-19 23:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][510/1519] eta 0:16:55 lr 0.000021 time 0.9295 (1.0060) model_time 0.9293 (1.0043) loss 0.9819 (0.8711) grad_norm 8.1895 (8.6218/1.9367) mem 68106MB [2022-12-19 23:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][520/1519] eta 0:16:44 lr 0.000021 time 0.9249 (1.0060) model_time 0.9248 (1.0042) loss 1.0227 (0.8715) grad_norm 14.4042 (8.6491/1.9562) mem 68106MB [2022-12-19 23:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][530/1519] eta 0:16:34 lr 0.000021 time 0.9330 (1.0059) model_time 0.9328 (1.0042) loss 0.8557 (0.8725) grad_norm 9.3888 (8.6691/1.9558) mem 68106MB [2022-12-19 23:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][540/1519] eta 0:16:24 lr 0.000021 time 0.9746 (1.0059) model_time 0.9745 (1.0042) loss 0.9116 (0.8716) grad_norm 8.2684 (8.6673/1.9433) mem 68106MB [2022-12-19 23:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][550/1519] eta 0:16:14 lr 0.000021 time 0.9268 (1.0061) model_time 0.9267 (1.0044) loss 0.9355 (0.8717) grad_norm 8.1255 (8.6702/1.9308) mem 68106MB [2022-12-19 23:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][560/1519] eta 0:16:04 lr 0.000020 time 0.9091 (1.0061) model_time 0.9090 (1.0045) loss 0.6893 (0.8703) grad_norm 7.6458 (8.6516/1.9224) mem 68106MB [2022-12-19 23:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][570/1519] eta 0:15:54 lr 0.000020 time 0.9279 (1.0061) model_time 0.9277 (1.0045) loss 0.6790 (0.8692) grad_norm 8.8633 (8.6489/1.9086) mem 68106MB [2022-12-19 23:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][580/1519] eta 0:15:44 lr 0.000020 time 0.9286 (1.0061) model_time 0.9284 (1.0045) loss 0.9077 (0.8681) grad_norm 7.6485 (8.6620/1.8986) mem 68106MB [2022-12-19 23:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][590/1519] eta 0:15:34 lr 0.000020 time 0.9253 (1.0060) model_time 0.9251 (1.0044) loss 0.8793 (0.8682) grad_norm 6.1770 (8.6669/1.9023) mem 68106MB [2022-12-19 23:20:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][600/1519] eta 0:15:24 lr 0.000020 time 1.1854 (1.0063) model_time 1.1853 (1.0048) loss 0.6826 (0.8669) grad_norm 10.5322 (8.6920/1.9031) mem 68106MB [2022-12-19 23:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][610/1519] eta 0:15:14 lr 0.000020 time 1.0860 (1.0066) model_time 1.0856 (1.0050) loss 0.9057 (0.8668) grad_norm 7.0199 (8.6787/1.9042) mem 68106MB [2022-12-19 23:20:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][620/1519] eta 0:15:04 lr 0.000020 time 0.9317 (1.0065) model_time 0.9315 (1.0049) loss 0.7472 (0.8675) grad_norm 11.2565 (8.6711/1.8859) mem 68106MB [2022-12-19 23:20:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][630/1519] eta 0:14:54 lr 0.000020 time 0.9699 (1.0065) model_time 0.9698 (1.0050) loss 0.7830 (0.8676) grad_norm 6.6605 (8.6309/1.8785) mem 68106MB [2022-12-19 23:20:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][640/1519] eta 0:14:44 lr 0.000020 time 0.9233 (1.0064) model_time 0.9232 (1.0049) loss 0.7920 (0.8669) grad_norm 7.2317 (8.6072/1.8884) mem 68106MB [2022-12-19 23:21:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][650/1519] eta 0:14:34 lr 0.000020 time 0.9247 (1.0065) model_time 0.9246 (1.0051) loss 0.6936 (0.8654) grad_norm 11.5659 (8.6218/1.8941) mem 68106MB [2022-12-19 23:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][660/1519] eta 0:14:24 lr 0.000020 time 0.9247 (1.0064) model_time 0.9245 (1.0049) loss 1.1440 (0.8666) grad_norm 12.3533 (8.6535/1.9129) mem 68106MB [2022-12-19 23:21:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][670/1519] eta 0:14:14 lr 0.000020 time 0.9403 (1.0064) model_time 0.9401 (1.0050) loss 0.7257 (0.8660) grad_norm 15.5925 (8.6854/1.9746) mem 68106MB [2022-12-19 23:21:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][680/1519] eta 0:14:04 lr 0.000020 time 0.9207 (1.0065) model_time 0.9206 (1.0051) loss 1.0574 (0.8666) grad_norm 9.2467 (8.7191/2.0057) mem 68106MB [2022-12-19 23:21:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][690/1519] eta 0:13:54 lr 0.000020 time 0.9271 (1.0064) model_time 0.9269 (1.0050) loss 0.9095 (0.8664) grad_norm 11.2214 (8.7446/2.0026) mem 68106MB [2022-12-19 23:21:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][700/1519] eta 0:13:44 lr 0.000020 time 0.9327 (1.0068) model_time 0.9326 (1.0054) loss 0.7010 (0.8664) grad_norm 7.7752 (8.7668/1.9960) mem 68106MB [2022-12-19 23:22:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][710/1519] eta 0:13:34 lr 0.000020 time 0.9212 (1.0067) model_time 0.9210 (1.0054) loss 0.6994 (0.8671) grad_norm 15.0433 (8.7930/2.0272) mem 68106MB [2022-12-19 23:22:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][720/1519] eta 0:13:24 lr 0.000020 time 1.1826 (1.0070) model_time 1.1824 (1.0056) loss 0.6949 (0.8670) grad_norm 7.8393 (8.7916/2.0274) mem 68106MB [2022-12-19 23:22:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][730/1519] eta 0:13:14 lr 0.000020 time 0.9183 (1.0069) model_time 0.9182 (1.0055) loss 0.8674 (0.8670) grad_norm 8.1342 (8.7355/1.9072) mem 68106MB [2022-12-19 23:22:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][740/1519] eta 0:13:04 lr 0.000020 time 0.9242 (1.0068) model_time 0.9240 (1.0055) loss 0.9985 (0.8673) grad_norm 7.6185 (8.7583/1.8983) mem 68106MB [2022-12-19 23:22:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][750/1519] eta 0:12:54 lr 0.000020 time 0.9244 (1.0067) model_time 0.9243 (1.0054) loss 0.9010 (0.8668) grad_norm 10.3892 (8.7774/1.8990) mem 68106MB [2022-12-19 23:22:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][760/1519] eta 0:12:44 lr 0.000020 time 0.9226 (1.0067) model_time 0.9225 (1.0054) loss 0.7066 (0.8667) grad_norm 7.6449 (8.8010/1.8883) mem 68106MB [2022-12-19 23:23:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][770/1519] eta 0:12:33 lr 0.000020 time 0.9209 (1.0066) model_time 0.9207 (1.0053) loss 0.7073 (0.8667) grad_norm 11.1883 (8.8210/1.8836) mem 68106MB [2022-12-19 23:23:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][780/1519] eta 0:12:23 lr 0.000020 time 0.9845 (1.0065) model_time 0.9843 (1.0052) loss 0.9481 (0.8665) grad_norm 7.5231 (8.8316/1.8820) mem 68106MB [2022-12-19 23:23:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][790/1519] eta 0:12:13 lr 0.000020 time 0.9236 (1.0066) model_time 0.9235 (1.0053) loss 0.7692 (0.8651) grad_norm 14.7231 (8.8586/1.9081) mem 68106MB [2022-12-19 23:23:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][800/1519] eta 0:12:03 lr 0.000020 time 0.9209 (1.0065) model_time 0.9207 (1.0052) loss 0.7850 (0.8649) grad_norm 9.2898 (8.8610/1.9021) mem 68106MB [2022-12-19 23:23:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][810/1519] eta 0:11:53 lr 0.000020 time 0.9224 (1.0064) model_time 0.9222 (1.0051) loss 0.8261 (0.8650) grad_norm 8.9008 (8.8479/1.9047) mem 68106MB [2022-12-19 23:23:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][820/1519] eta 0:11:43 lr 0.000020 time 0.9261 (1.0063) model_time 0.9260 (1.0051) loss 0.7086 (0.8646) grad_norm 9.2349 (8.8237/1.8985) mem 68106MB [2022-12-19 23:24:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][830/1519] eta 0:11:33 lr 0.000020 time 0.9242 (1.0062) model_time 0.9241 (1.0049) loss 0.8286 (0.8643) grad_norm 8.8804 (8.8364/1.8926) mem 68106MB [2022-12-19 23:24:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][840/1519] eta 0:11:23 lr 0.000020 time 0.9252 (1.0061) model_time 0.9250 (1.0049) loss 0.7656 (0.8643) grad_norm 7.5341 (8.8338/1.8904) mem 68106MB [2022-12-19 23:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][850/1519] eta 0:11:13 lr 0.000020 time 0.9241 (1.0060) model_time 0.9240 (1.0048) loss 0.6676 (0.8636) grad_norm 8.8288 (8.8805/1.9202) mem 68106MB [2022-12-19 23:24:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][860/1519] eta 0:11:02 lr 0.000020 time 0.9240 (1.0060) model_time 0.9238 (1.0048) loss 0.8972 (0.8641) grad_norm 7.3006 (8.8629/1.9158) mem 68106MB [2022-12-19 23:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][870/1519] eta 0:10:52 lr 0.000020 time 0.9214 (1.0060) model_time 0.9212 (1.0048) loss 0.7095 (0.8640) grad_norm 6.9034 (8.8088/1.9345) mem 68106MB [2022-12-19 23:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][880/1519] eta 0:10:42 lr 0.000020 time 0.9248 (1.0060) model_time 0.9246 (1.0048) loss 0.6906 (0.8636) grad_norm 7.3581 (8.7940/1.9411) mem 68106MB [2022-12-19 23:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][890/1519] eta 0:10:32 lr 0.000020 time 0.9243 (1.0059) model_time 0.9241 (1.0047) loss 0.9089 (0.8638) grad_norm 6.6289 (8.7790/1.9364) mem 68106MB [2022-12-19 23:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][900/1519] eta 0:10:22 lr 0.000020 time 1.0635 (1.0060) model_time 1.0634 (1.0048) loss 0.8359 (0.8639) grad_norm 7.3079 (8.7865/1.9579) mem 68106MB [2022-12-19 23:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][910/1519] eta 0:10:12 lr 0.000020 time 0.9267 (1.0060) model_time 0.9266 (1.0048) loss 0.7465 (0.8645) grad_norm 7.6370 (8.7833/1.9581) mem 68106MB [2022-12-19 23:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][920/1519] eta 0:10:02 lr 0.000020 time 0.9261 (1.0059) model_time 0.9260 (1.0047) loss 0.7546 (0.8634) grad_norm 10.6113 (8.7920/1.9488) mem 68106MB [2022-12-19 23:25:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][930/1519] eta 0:09:52 lr 0.000020 time 0.9224 (1.0060) model_time 0.9223 (1.0049) loss 0.9520 (0.8634) grad_norm 8.4573 (8.8220/1.9932) mem 68106MB [2022-12-19 23:25:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][940/1519] eta 0:09:42 lr 0.000020 time 0.9239 (1.0059) model_time 0.9237 (1.0048) loss 0.7126 (0.8627) grad_norm 6.0159 (8.8226/1.9998) mem 68106MB [2022-12-19 23:26:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][950/1519] eta 0:09:32 lr 0.000020 time 0.9268 (1.0060) model_time 0.9267 (1.0049) loss 1.1717 (0.8632) grad_norm 9.4433 (8.8275/2.0003) mem 68106MB [2022-12-19 23:26:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][960/1519] eta 0:09:22 lr 0.000020 time 0.9340 (1.0063) model_time 0.9339 (1.0052) loss 0.7718 (0.8634) grad_norm 7.3097 (8.8350/1.9946) mem 68106MB [2022-12-19 23:26:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][970/1519] eta 0:09:12 lr 0.000020 time 0.9255 (1.0065) model_time 0.9253 (1.0054) loss 1.2323 (0.8642) grad_norm 10.6970 (8.8259/1.9947) mem 68106MB [2022-12-19 23:26:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][980/1519] eta 0:09:02 lr 0.000020 time 0.9232 (1.0064) model_time 0.9231 (1.0053) loss 1.1288 (0.8645) grad_norm 9.5782 (8.7822/1.9134) mem 68106MB [2022-12-19 23:26:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][990/1519] eta 0:08:52 lr 0.000020 time 1.0959 (1.0065) model_time 1.0956 (1.0054) loss 0.7690 (0.8653) grad_norm 8.7795 (8.8157/1.8912) mem 68106MB [2022-12-19 23:26:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1000/1519] eta 0:08:42 lr 0.000020 time 0.9200 (1.0064) model_time 0.9198 (1.0053) loss 0.8462 (0.8650) grad_norm 8.7397 (8.8193/1.8994) mem 68106MB [2022-12-19 23:27:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1010/1519] eta 0:08:32 lr 0.000020 time 0.9227 (1.0064) model_time 0.9225 (1.0053) loss 0.6974 (0.8645) grad_norm 32.9625 (8.9070/2.3491) mem 68106MB [2022-12-19 23:27:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1020/1519] eta 0:08:22 lr 0.000020 time 0.9969 (1.0064) model_time 0.9967 (1.0053) loss 0.7064 (0.8648) grad_norm 7.1919 (8.9092/2.3531) mem 68106MB [2022-12-19 23:27:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1030/1519] eta 0:08:12 lr 0.000020 time 0.9284 (1.0063) model_time 0.9283 (1.0052) loss 1.0735 (0.8644) grad_norm 7.1666 (8.9070/2.3543) mem 68106MB [2022-12-19 23:27:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1040/1519] eta 0:08:02 lr 0.000020 time 0.9243 (1.0063) model_time 0.9242 (1.0052) loss 1.0195 (0.8639) grad_norm 11.8458 (8.8793/2.3279) mem 68106MB [2022-12-19 23:27:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1050/1519] eta 0:07:51 lr 0.000020 time 0.9245 (1.0063) model_time 0.9244 (1.0052) loss 0.8796 (0.8645) grad_norm 8.9298 (8.8572/2.3046) mem 68106MB [2022-12-19 23:27:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1060/1519] eta 0:07:41 lr 0.000020 time 0.9268 (1.0062) model_time 0.9267 (1.0052) loss 0.8311 (0.8640) grad_norm 8.7838 (8.8342/2.3004) mem 68106MB [2022-12-19 23:28:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1070/1519] eta 0:07:31 lr 0.000020 time 0.9281 (1.0063) model_time 0.9280 (1.0052) loss 0.9260 (0.8637) grad_norm 17.7784 (8.8548/2.3508) mem 68106MB [2022-12-19 23:28:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1080/1519] eta 0:07:21 lr 0.000020 time 0.9262 (1.0062) model_time 0.9260 (1.0052) loss 0.7174 (0.8633) grad_norm 7.2115 (8.8499/2.3542) mem 68106MB [2022-12-19 23:28:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1090/1519] eta 0:07:11 lr 0.000020 time 0.9255 (1.0062) model_time 0.9254 (1.0051) loss 1.1310 (0.8636) grad_norm 7.0562 (8.8427/2.3741) mem 68106MB [2022-12-19 23:28:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1100/1519] eta 0:07:01 lr 0.000020 time 0.9243 (1.0062) model_time 0.9242 (1.0052) loss 0.7150 (0.8633) grad_norm 6.7624 (8.8684/2.4667) mem 68106MB [2022-12-19 23:28:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1110/1519] eta 0:06:51 lr 0.000020 time 0.9251 (1.0062) model_time 0.9250 (1.0052) loss 0.8646 (0.8635) grad_norm 15.3538 (8.9103/2.4902) mem 68106MB [2022-12-19 23:29:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1120/1519] eta 0:06:41 lr 0.000020 time 0.9280 (1.0061) model_time 0.9279 (1.0051) loss 0.8134 (0.8632) grad_norm 7.4392 (8.8706/2.4736) mem 68106MB [2022-12-19 23:29:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1130/1519] eta 0:06:31 lr 0.000020 time 0.9240 (1.0062) model_time 0.9239 (1.0051) loss 0.7937 (0.8634) grad_norm 8.1286 (8.8493/2.4735) mem 68106MB [2022-12-19 23:29:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1140/1519] eta 0:06:21 lr 0.000020 time 0.9247 (1.0061) model_time 0.9246 (1.0051) loss 0.6991 (0.8632) grad_norm 11.6289 (8.8494/2.4782) mem 68106MB [2022-12-19 23:29:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1150/1519] eta 0:06:11 lr 0.000020 time 0.9327 (1.0060) model_time 0.9325 (1.0050) loss 0.7939 (0.8632) grad_norm 7.0079 (8.8227/2.4837) mem 68106MB [2022-12-19 23:29:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1160/1519] eta 0:06:01 lr 0.000020 time 0.9236 (1.0060) model_time 0.9235 (1.0050) loss 0.7017 (0.8635) grad_norm 9.0445 (8.8409/2.4812) mem 68106MB [2022-12-19 23:29:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1170/1519] eta 0:05:51 lr 0.000020 time 0.9295 (1.0060) model_time 0.9294 (1.0050) loss 0.7773 (0.8636) grad_norm 6.6397 (8.8272/2.4859) mem 68106MB [2022-12-19 23:30:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1180/1519] eta 0:05:41 lr 0.000020 time 0.9286 (1.0060) model_time 0.9285 (1.0050) loss 0.9733 (0.8638) grad_norm 8.5369 (8.8062/2.4844) mem 68106MB [2022-12-19 23:30:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1190/1519] eta 0:05:31 lr 0.000020 time 0.9424 (1.0062) model_time 0.9423 (1.0052) loss 0.8719 (0.8642) grad_norm 8.8445 (8.8143/2.5099) mem 68106MB [2022-12-19 23:30:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1200/1519] eta 0:05:20 lr 0.000020 time 0.9897 (1.0062) model_time 0.9896 (1.0052) loss 0.6914 (0.8638) grad_norm 8.4576 (8.7839/2.5017) mem 68106MB [2022-12-19 23:30:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1210/1519] eta 0:05:10 lr 0.000020 time 0.9204 (1.0061) model_time 0.9203 (1.0052) loss 0.8523 (0.8638) grad_norm 10.5504 (8.7849/2.5029) mem 68106MB [2022-12-19 23:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1220/1519] eta 0:05:00 lr 0.000020 time 0.9455 (1.0064) model_time 0.9454 (1.0054) loss 0.6759 (0.8635) grad_norm 7.6463 (8.8058/2.5156) mem 68106MB [2022-12-19 23:30:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1230/1519] eta 0:04:50 lr 0.000020 time 0.9326 (1.0063) model_time 0.9325 (1.0054) loss 0.7556 (0.8632) grad_norm 6.4812 (8.8150/2.5138) mem 68106MB [2022-12-19 23:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1240/1519] eta 0:04:40 lr 0.000020 time 0.9331 (1.0064) model_time 0.9330 (1.0054) loss 0.8290 (0.8625) grad_norm 7.9606 (8.8186/2.5035) mem 68106MB [2022-12-19 23:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1250/1519] eta 0:04:30 lr 0.000020 time 0.9306 (1.0064) model_time 0.9304 (1.0054) loss 0.7603 (0.8624) grad_norm 8.6620 (8.7901/2.5078) mem 68106MB [2022-12-19 23:31:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1260/1519] eta 0:04:20 lr 0.000020 time 0.9535 (1.0064) model_time 0.9534 (1.0054) loss 1.1513 (0.8622) grad_norm 5.9362 (8.7583/2.4959) mem 68106MB [2022-12-19 23:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1270/1519] eta 0:04:10 lr 0.000020 time 0.9879 (1.0064) model_time 0.9878 (1.0054) loss 0.8785 (0.8622) grad_norm 7.8552 (8.6942/2.4299) mem 68106MB [2022-12-19 23:31:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1280/1519] eta 0:04:00 lr 0.000020 time 0.9199 (1.0063) model_time 0.9197 (1.0054) loss 0.6700 (0.8622) grad_norm 6.6220 (8.6533/2.4146) mem 68106MB [2022-12-19 23:31:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1290/1519] eta 0:03:50 lr 0.000020 time 0.9196 (1.0063) model_time 0.9194 (1.0054) loss 1.0146 (0.8618) grad_norm 6.2289 (8.6335/2.4202) mem 68106MB [2022-12-19 23:32:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1300/1519] eta 0:03:40 lr 0.000020 time 0.9282 (1.0063) model_time 0.9281 (1.0054) loss 0.8500 (0.8620) grad_norm 7.7456 (8.6301/2.4197) mem 68106MB [2022-12-19 23:32:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1310/1519] eta 0:03:30 lr 0.000020 time 0.9443 (1.0063) model_time 0.9442 (1.0054) loss 0.6903 (0.8617) grad_norm 7.4275 (8.6020/2.3932) mem 68106MB [2022-12-19 23:32:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1320/1519] eta 0:03:20 lr 0.000020 time 0.9390 (1.0063) model_time 0.9389 (1.0054) loss 0.7051 (0.8613) grad_norm 8.3559 (8.5994/2.3974) mem 68106MB [2022-12-19 23:32:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1330/1519] eta 0:03:10 lr 0.000020 time 0.9294 (1.0062) model_time 0.9293 (1.0053) loss 1.2375 (0.8617) grad_norm 7.3359 (8.6194/2.4034) mem 68106MB [2022-12-19 23:32:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1340/1519] eta 0:03:00 lr 0.000020 time 0.9268 (1.0062) model_time 0.9266 (1.0053) loss 0.8103 (0.8617) grad_norm 8.2954 (8.6343/2.4001) mem 68106MB [2022-12-19 23:32:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1350/1519] eta 0:02:50 lr 0.000020 time 0.9329 (1.0062) model_time 0.9327 (1.0053) loss 0.7065 (0.8617) grad_norm 7.3539 (8.6216/2.3994) mem 68106MB [2022-12-19 23:33:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1360/1519] eta 0:02:39 lr 0.000020 time 0.9313 (1.0062) model_time 0.9311 (1.0053) loss 0.9808 (0.8619) grad_norm 8.4670 (8.6176/2.4040) mem 68106MB [2022-12-19 23:33:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1370/1519] eta 0:02:29 lr 0.000020 time 0.9273 (1.0063) model_time 0.9272 (1.0054) loss 0.9844 (0.8621) grad_norm 7.7155 (8.6054/2.3999) mem 68106MB [2022-12-19 23:33:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1380/1519] eta 0:02:19 lr 0.000020 time 0.9177 (1.0062) model_time 0.9176 (1.0053) loss 0.6843 (0.8616) grad_norm 11.3751 (8.6276/2.4033) mem 68106MB [2022-12-19 23:33:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1390/1519] eta 0:02:09 lr 0.000020 time 0.9249 (1.0061) model_time 0.9248 (1.0053) loss 0.7472 (0.8611) grad_norm 9.8571 (8.6311/2.3972) mem 68106MB [2022-12-19 23:33:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1400/1519] eta 0:01:59 lr 0.000020 time 0.9271 (1.0061) model_time 0.9269 (1.0052) loss 0.9096 (0.8608) grad_norm 6.3951 (8.6324/2.4018) mem 68106MB [2022-12-19 23:33:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1410/1519] eta 0:01:49 lr 0.000020 time 0.9230 (1.0062) model_time 0.9228 (1.0053) loss 0.7375 (0.8605) grad_norm 11.5276 (8.6518/2.4068) mem 68106MB [2022-12-19 23:34:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1420/1519] eta 0:01:39 lr 0.000020 time 0.9217 (1.0062) model_time 0.9216 (1.0053) loss 0.8302 (0.8607) grad_norm 8.5259 (8.6601/2.4068) mem 68106MB [2022-12-19 23:34:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1430/1519] eta 0:01:29 lr 0.000020 time 0.9334 (1.0062) model_time 0.9332 (1.0053) loss 0.7431 (0.8612) grad_norm 8.5459 (8.6469/2.4044) mem 68106MB [2022-12-19 23:34:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1440/1519] eta 0:01:19 lr 0.000020 time 0.9208 (1.0062) model_time 0.9207 (1.0053) loss 0.9759 (0.8610) grad_norm 6.3513 (8.6438/2.4085) mem 68106MB [2022-12-19 23:34:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1450/1519] eta 0:01:09 lr 0.000020 time 0.9223 (1.0061) model_time 0.9222 (1.0053) loss 0.8524 (0.8613) grad_norm 6.7857 (8.6024/2.3785) mem 68106MB [2022-12-19 23:34:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1460/1519] eta 0:00:59 lr 0.000020 time 0.9211 (1.0061) model_time 0.9210 (1.0052) loss 0.8507 (0.8609) grad_norm 8.3719 (8.6141/2.3826) mem 68106MB [2022-12-19 23:34:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1470/1519] eta 0:00:49 lr 0.000020 time 0.9232 (1.0061) model_time 0.9231 (1.0052) loss 1.0671 (0.8612) grad_norm 12.4508 (8.6520/2.3774) mem 68106MB [2022-12-19 23:35:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1480/1519] eta 0:00:39 lr 0.000020 time 0.9232 (1.0061) model_time 0.9230 (1.0052) loss 0.8620 (0.8610) grad_norm 9.4212 (8.6448/2.3693) mem 68106MB [2022-12-19 23:35:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1490/1519] eta 0:00:29 lr 0.000020 time 0.9279 (1.0061) model_time 0.9278 (1.0053) loss 1.1383 (0.8609) grad_norm 7.3907 (8.6435/2.3633) mem 68106MB [2022-12-19 23:35:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1500/1519] eta 0:00:19 lr 0.000020 time 0.9554 (1.0061) model_time 0.9553 (1.0052) loss 1.3027 (0.8613) grad_norm 7.0372 (8.6384/2.3499) mem 68106MB [2022-12-19 23:35:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [43/100][1510/1519] eta 0:00:09 lr 0.000020 time 0.9227 (1.0060) model_time 0.9226 (1.0052) loss 1.0921 (0.8612) grad_norm 10.2184 (8.6501/2.3591) mem 68106MB [2022-12-19 23:35:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 43 training takes 0:25:28 [2022-12-19 23:35:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_43.pth saving...... [2022-12-19 23:36:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_43.pth saved !!! [2022-12-19 23:36:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.672 (0.672) Loss 0.5113 (0.5113) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-19 23:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.299 (0.331) Loss 0.5155 (0.4914) Acc@1 92.014 (92.140) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-19 23:36:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4604 (0.4906) Acc@1 92.014 (91.964) Acc@5 98.958 (98.347) Mem 68106MB [2022-12-19 23:36:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.310) Loss 0.6007 (0.4936) Acc@1 89.236 (91.868) Acc@5 98.264 (98.320) Mem 68106MB [2022-12-19 23:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4774 (0.4842) Acc@1 90.972 (92.005) Acc@5 98.264 (98.399) Mem 68106MB [2022-12-19 23:36:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.306) Loss 0.4991 (0.4820) Acc@1 90.278 (91.966) Acc@5 98.958 (98.434) Mem 68106MB [2022-12-19 23:36:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.4892 (0.4815) Acc@1 91.319 (91.968) Acc@5 97.917 (98.412) Mem 68106MB [2022-12-19 23:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.303 (0.305) Loss 0.5243 (0.4820) Acc@1 92.014 (91.921) Acc@5 98.264 (98.420) Mem 68106MB [2022-12-19 23:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.304) Loss 0.4153 (0.4809) Acc@1 93.403 (91.924) Acc@5 97.917 (98.431) Mem 68106MB [2022-12-19 23:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:43] * Acc@1 91.892 Acc@5 98.441 [2022-12-19 23:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.9% [2022-12-19 23:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.90% [2022-12-19 23:36:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][0/1519] eta 0:48:01 lr 0.000020 time 1.8968 (1.8968) model_time 1.3678 (1.3678) loss 0.7656 (0.7656) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-19 23:36:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][10/1519] eta 0:27:25 lr 0.000020 time 0.9264 (1.0905) model_time 0.9263 (1.0421) loss 1.0681 (0.8241) grad_norm 7.1048 (7.7649/1.1637) mem 68106MB [2022-12-19 23:36:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][20/1519] eta 0:26:09 lr 0.000020 time 0.9276 (1.0472) model_time 0.9275 (1.0217) loss 0.6966 (0.8247) grad_norm 9.1213 (8.2094/1.1222) mem 68106MB [2022-12-19 23:37:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][30/1519] eta 0:25:40 lr 0.000020 time 0.9248 (1.0345) model_time 0.9246 (1.0171) loss 0.7419 (0.8442) grad_norm 9.1475 (8.2139/1.1732) mem 68106MB [2022-12-19 23:37:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][40/1519] eta 0:25:25 lr 0.000020 time 0.9221 (1.0312) model_time 0.9220 (1.0180) loss 1.0128 (0.8426) grad_norm 9.8890 (8.2409/1.1346) mem 68106MB [2022-12-19 23:37:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][50/1519] eta 0:25:06 lr 0.000020 time 0.9187 (1.0257) model_time 0.9186 (1.0151) loss 0.6829 (0.8405) grad_norm 6.8023 (8.0560/1.2118) mem 68106MB [2022-12-19 23:37:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][60/1519] eta 0:24:49 lr 0.000020 time 0.8887 (1.0212) model_time 0.8886 (1.0122) loss 0.7632 (0.8471) grad_norm 7.4817 (7.8423/1.2881) mem 68106MB [2022-12-19 23:37:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][70/1519] eta 0:24:36 lr 0.000020 time 0.9228 (1.0191) model_time 0.9227 (1.0113) loss 1.0508 (0.8577) grad_norm 9.5254 (7.8792/1.3156) mem 68106MB [2022-12-19 23:37:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][80/1519] eta 0:24:23 lr 0.000020 time 0.9282 (1.0170) model_time 0.9280 (1.0102) loss 0.9934 (0.8607) grad_norm 6.6958 (7.8927/1.3758) mem 68106MB [2022-12-19 23:38:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][90/1519] eta 0:24:10 lr 0.000020 time 0.9398 (1.0150) model_time 0.9397 (1.0089) loss 0.7292 (0.8532) grad_norm 5.9593 (7.9277/1.4280) mem 68106MB [2022-12-19 23:38:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][100/1519] eta 0:23:58 lr 0.000020 time 0.9225 (1.0137) model_time 0.9224 (1.0081) loss 0.6911 (0.8491) grad_norm 10.7643 (7.9946/1.4243) mem 68106MB [2022-12-19 23:38:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][110/1519] eta 0:23:46 lr 0.000020 time 0.9233 (1.0121) model_time 0.9232 (1.0070) loss 1.0856 (0.8523) grad_norm 7.2782 (7.9993/1.4011) mem 68106MB [2022-12-19 23:38:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][120/1519] eta 0:23:34 lr 0.000020 time 0.9238 (1.0109) model_time 0.9236 (1.0063) loss 0.7064 (0.8518) grad_norm 9.1231 (8.0455/1.3586) mem 68106MB [2022-12-19 23:38:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][130/1519] eta 0:23:22 lr 0.000020 time 0.9362 (1.0101) model_time 0.9361 (1.0057) loss 0.7968 (0.8507) grad_norm 9.2752 (8.0110/1.3856) mem 68106MB [2022-12-19 23:38:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][140/1519] eta 0:23:12 lr 0.000020 time 0.9240 (1.0097) model_time 0.9234 (1.0056) loss 0.7753 (0.8465) grad_norm 7.6252 (8.0564/1.3996) mem 68106MB [2022-12-19 23:39:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][150/1519] eta 0:23:02 lr 0.000020 time 0.9431 (1.0098) model_time 0.9430 (1.0060) loss 0.7573 (0.8428) grad_norm 6.4017 (8.1298/1.5134) mem 68106MB [2022-12-19 23:39:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][160/1519] eta 0:22:52 lr 0.000020 time 1.0081 (1.0103) model_time 1.0080 (1.0067) loss 0.8182 (0.8418) grad_norm 7.0266 (8.0517/1.5035) mem 68106MB [2022-12-19 23:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][170/1519] eta 0:22:41 lr 0.000020 time 0.9255 (1.0095) model_time 0.9254 (1.0061) loss 0.9421 (0.8414) grad_norm 6.3787 (8.0230/1.4867) mem 68106MB [2022-12-19 23:39:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][180/1519] eta 0:22:31 lr 0.000020 time 0.9285 (1.0090) model_time 0.9284 (1.0057) loss 0.9397 (0.8424) grad_norm 7.4695 (8.0733/1.4713) mem 68106MB [2022-12-19 23:39:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][190/1519] eta 0:22:22 lr 0.000020 time 0.9249 (1.0098) model_time 0.9248 (1.0067) loss 0.6865 (0.8420) grad_norm 7.0451 (8.0819/1.4458) mem 68106MB [2022-12-19 23:39:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][200/1519] eta 0:22:11 lr 0.000020 time 0.9133 (1.0098) model_time 0.9132 (1.0068) loss 0.8609 (0.8399) grad_norm 9.9537 (8.1246/1.4821) mem 68106MB [2022-12-19 23:40:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][210/1519] eta 0:22:02 lr 0.000020 time 0.9327 (1.0099) model_time 0.9326 (1.0071) loss 0.8805 (0.8424) grad_norm 8.8603 (8.1803/1.5242) mem 68106MB [2022-12-19 23:40:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][220/1519] eta 0:21:51 lr 0.000020 time 0.9267 (1.0095) model_time 0.9266 (1.0067) loss 0.8113 (0.8424) grad_norm 8.8314 (8.1528/1.5018) mem 68106MB [2022-12-19 23:40:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][230/1519] eta 0:21:40 lr 0.000020 time 0.9208 (1.0092) model_time 0.9207 (1.0065) loss 0.7565 (0.8441) grad_norm 7.8079 (8.1677/1.4996) mem 68106MB [2022-12-19 23:40:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][240/1519] eta 0:21:30 lr 0.000020 time 0.9212 (1.0087) model_time 0.9211 (1.0062) loss 0.8401 (0.8479) grad_norm 7.3808 (8.1416/1.4783) mem 68106MB [2022-12-19 23:40:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][250/1519] eta 0:21:19 lr 0.000020 time 0.9251 (1.0083) model_time 0.9250 (1.0059) loss 0.7145 (0.8487) grad_norm 7.5251 (8.1212/1.4583) mem 68106MB [2022-12-19 23:40:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][260/1519] eta 0:21:09 lr 0.000020 time 0.9826 (1.0083) model_time 0.9825 (1.0059) loss 0.7025 (0.8495) grad_norm 12.1413 (8.1773/1.4872) mem 68106MB [2022-12-19 23:41:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][270/1519] eta 0:20:59 lr 0.000020 time 0.9277 (1.0081) model_time 0.9276 (1.0058) loss 0.9521 (0.8480) grad_norm 7.8542 (8.1464/1.4759) mem 68106MB [2022-12-19 23:41:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][280/1519] eta 0:20:48 lr 0.000020 time 0.9226 (1.0079) model_time 0.9225 (1.0057) loss 0.9713 (0.8500) grad_norm 6.0176 (8.1320/1.4700) mem 68106MB [2022-12-19 23:41:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][290/1519] eta 0:20:38 lr 0.000020 time 0.9248 (1.0076) model_time 0.9246 (1.0055) loss 0.6726 (0.8505) grad_norm 6.7895 (8.1467/1.4912) mem 68106MB [2022-12-19 23:41:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][300/1519] eta 0:20:28 lr 0.000020 time 0.9225 (1.0074) model_time 0.9223 (1.0054) loss 0.8662 (0.8480) grad_norm 8.0956 (8.2234/1.5704) mem 68106MB [2022-12-19 23:41:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][310/1519] eta 0:20:17 lr 0.000020 time 0.9335 (1.0073) model_time 0.9333 (1.0053) loss 1.2018 (0.8481) grad_norm 6.0979 (8.2367/1.5970) mem 68106MB [2022-12-19 23:41:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][320/1519] eta 0:20:07 lr 0.000020 time 0.9219 (1.0074) model_time 0.9217 (1.0054) loss 0.7394 (0.8485) grad_norm 10.4923 (8.2755/1.6105) mem 68106MB [2022-12-19 23:42:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][330/1519] eta 0:19:57 lr 0.000020 time 0.9287 (1.0073) model_time 0.9285 (1.0054) loss 0.9294 (0.8476) grad_norm 7.0639 (8.3720/1.9790) mem 68106MB [2022-12-19 23:42:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][340/1519] eta 0:19:48 lr 0.000020 time 1.2167 (1.0083) model_time 1.2165 (1.0064) loss 0.9638 (0.8481) grad_norm 6.3871 (8.3640/1.9678) mem 68106MB [2022-12-19 23:42:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][350/1519] eta 0:19:38 lr 0.000020 time 0.9233 (1.0082) model_time 0.9232 (1.0064) loss 0.8582 (0.8479) grad_norm 7.7671 (8.3315/1.9546) mem 68106MB [2022-12-19 23:42:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][360/1519] eta 0:19:28 lr 0.000020 time 0.9233 (1.0082) model_time 0.9231 (1.0064) loss 0.7772 (0.8480) grad_norm 8.4433 (8.3674/1.9544) mem 68106MB [2022-12-19 23:42:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][370/1519] eta 0:19:18 lr 0.000020 time 0.9296 (1.0080) model_time 0.9294 (1.0062) loss 1.2401 (0.8516) grad_norm 10.5295 (8.3691/1.9370) mem 68106MB [2022-12-19 23:42:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][380/1519] eta 0:19:08 lr 0.000020 time 0.9271 (1.0080) model_time 0.9270 (1.0062) loss 0.8735 (0.8503) grad_norm 9.6425 (8.3849/1.9169) mem 68106MB [2022-12-19 23:43:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][390/1519] eta 0:18:57 lr 0.000020 time 0.9310 (1.0077) model_time 0.9308 (1.0060) loss 0.8459 (0.8500) grad_norm 6.9230 (8.3832/1.9019) mem 68106MB [2022-12-19 23:43:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][400/1519] eta 0:18:47 lr 0.000020 time 0.9348 (1.0076) model_time 0.9346 (1.0060) loss 0.8494 (0.8495) grad_norm 11.9832 (8.4113/1.9063) mem 68106MB [2022-12-19 23:43:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][410/1519] eta 0:18:37 lr 0.000020 time 0.9341 (1.0074) model_time 0.9339 (1.0058) loss 0.7182 (0.8488) grad_norm 9.0339 (8.4077/1.8946) mem 68106MB [2022-12-19 23:43:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][420/1519] eta 0:18:26 lr 0.000020 time 0.9228 (1.0071) model_time 0.9227 (1.0056) loss 0.7132 (0.8476) grad_norm 7.5801 (8.3933/1.8783) mem 68106MB [2022-12-19 23:43:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][430/1519] eta 0:18:16 lr 0.000020 time 0.9264 (1.0069) model_time 0.9263 (1.0053) loss 0.7464 (0.8462) grad_norm 6.6520 (8.4112/1.8793) mem 68106MB [2022-12-19 23:43:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][440/1519] eta 0:18:06 lr 0.000020 time 0.9867 (1.0069) model_time 0.9866 (1.0054) loss 0.9471 (0.8468) grad_norm 13.5240 (8.4219/1.8983) mem 68106MB [2022-12-19 23:44:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][450/1519] eta 0:17:56 lr 0.000020 time 0.9263 (1.0070) model_time 0.9262 (1.0055) loss 0.7013 (0.8476) grad_norm 9.9798 (8.4192/1.8833) mem 68106MB [2022-12-19 23:44:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][460/1519] eta 0:17:46 lr 0.000020 time 0.9252 (1.0069) model_time 0.9251 (1.0055) loss 0.6822 (0.8464) grad_norm 8.1688 (8.4146/1.8715) mem 68106MB [2022-12-19 23:44:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][470/1519] eta 0:17:36 lr 0.000020 time 0.9347 (1.0068) model_time 0.9346 (1.0053) loss 0.7978 (0.8484) grad_norm 9.5375 (8.4335/1.8719) mem 68106MB [2022-12-19 23:44:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][480/1519] eta 0:17:25 lr 0.000020 time 0.9457 (1.0067) model_time 0.9456 (1.0052) loss 0.7456 (0.8484) grad_norm 10.0975 (8.4152/1.8720) mem 68106MB [2022-12-19 23:44:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][490/1519] eta 0:17:15 lr 0.000020 time 0.9328 (1.0067) model_time 0.9327 (1.0053) loss 0.8296 (0.8467) grad_norm 6.5743 (8.4195/1.8742) mem 68106MB [2022-12-19 23:44:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][500/1519] eta 0:17:05 lr 0.000020 time 0.9107 (1.0066) model_time 0.9105 (1.0052) loss 0.9988 (0.8474) grad_norm 7.9381 (8.3987/1.8658) mem 68106MB [2022-12-19 23:45:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][510/1519] eta 0:16:55 lr 0.000020 time 0.9276 (1.0065) model_time 0.9275 (1.0051) loss 0.9899 (0.8481) grad_norm 13.7175 (8.4091/1.8885) mem 68106MB [2022-12-19 23:45:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][520/1519] eta 0:16:46 lr 0.000020 time 0.8795 (1.0070) model_time 0.8793 (1.0057) loss 0.6809 (0.8473) grad_norm 7.8771 (8.3983/1.8752) mem 68106MB [2022-12-19 23:45:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][530/1519] eta 0:16:35 lr 0.000020 time 0.9375 (1.0069) model_time 0.9374 (1.0056) loss 0.7353 (0.8479) grad_norm 5.6949 (8.3626/1.8805) mem 68106MB [2022-12-19 23:45:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][540/1519] eta 0:16:25 lr 0.000020 time 0.9382 (1.0069) model_time 0.9381 (1.0056) loss 0.7899 (0.8471) grad_norm 9.2355 (8.3596/1.8667) mem 68106MB [2022-12-19 23:45:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][550/1519] eta 0:16:15 lr 0.000020 time 0.9418 (1.0069) model_time 0.9417 (1.0056) loss 1.0040 (0.8474) grad_norm 9.4180 (8.3684/1.8584) mem 68106MB [2022-12-19 23:45:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][560/1519] eta 0:16:05 lr 0.000020 time 0.9245 (1.0068) model_time 0.9243 (1.0055) loss 0.6793 (0.8473) grad_norm 7.6644 (8.3687/1.8612) mem 68106MB [2022-12-19 23:46:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][570/1519] eta 0:15:55 lr 0.000020 time 0.9355 (1.0066) model_time 0.9354 (1.0053) loss 0.9112 (0.8472) grad_norm 10.8309 (8.3978/1.8641) mem 68106MB [2022-12-19 23:46:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][580/1519] eta 0:15:45 lr 0.000020 time 0.9687 (1.0065) model_time 0.9686 (1.0053) loss 0.7634 (0.8487) grad_norm 7.6489 (8.3992/1.8600) mem 68106MB [2022-12-19 23:46:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][590/1519] eta 0:15:34 lr 0.000020 time 0.9336 (1.0064) model_time 0.9335 (1.0051) loss 0.8813 (0.8486) grad_norm 13.0022 (8.4067/1.8689) mem 68106MB [2022-12-19 23:46:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][600/1519] eta 0:15:24 lr 0.000020 time 0.9362 (1.0063) model_time 0.9361 (1.0051) loss 0.8694 (0.8497) grad_norm 9.3133 (8.4073/1.8547) mem 68106MB [2022-12-19 23:46:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][610/1519] eta 0:15:14 lr 0.000020 time 0.9307 (1.0062) model_time 0.9306 (1.0050) loss 0.9689 (0.8503) grad_norm 9.0595 (8.4310/1.8544) mem 68106MB [2022-12-19 23:46:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][620/1519] eta 0:15:04 lr 0.000020 time 0.9266 (1.0060) model_time 0.9262 (1.0048) loss 0.8809 (0.8495) grad_norm 10.1285 (8.4192/1.8575) mem 68106MB [2022-12-19 23:47:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][630/1519] eta 0:14:54 lr 0.000020 time 0.9395 (1.0059) model_time 0.9393 (1.0047) loss 0.8329 (0.8483) grad_norm 7.9344 (8.4024/1.8597) mem 68106MB [2022-12-19 23:47:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][640/1519] eta 0:14:44 lr 0.000020 time 0.9346 (1.0061) model_time 0.9344 (1.0049) loss 0.8391 (0.8477) grad_norm 10.4111 (8.4107/1.8636) mem 68106MB [2022-12-19 23:47:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][650/1519] eta 0:14:34 lr 0.000020 time 0.9252 (1.0060) model_time 0.9250 (1.0049) loss 0.8085 (0.8467) grad_norm 7.3814 (8.4324/1.8699) mem 68106MB [2022-12-19 23:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][660/1519] eta 0:14:24 lr 0.000020 time 0.9281 (1.0062) model_time 0.9279 (1.0051) loss 1.0812 (0.8473) grad_norm 11.6038 (8.4970/1.8915) mem 68106MB [2022-12-19 23:47:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][670/1519] eta 0:14:14 lr 0.000020 time 0.9335 (1.0065) model_time 0.9334 (1.0053) loss 0.9326 (0.8469) grad_norm 8.1918 (8.4975/1.8967) mem 68106MB [2022-12-19 23:47:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][680/1519] eta 0:14:04 lr 0.000020 time 0.9324 (1.0063) model_time 0.9323 (1.0052) loss 0.9711 (0.8471) grad_norm 8.8873 (8.5131/1.9012) mem 68106MB [2022-12-19 23:48:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][690/1519] eta 0:13:54 lr 0.000020 time 0.9325 (1.0064) model_time 0.9323 (1.0052) loss 0.6869 (0.8469) grad_norm 6.5803 (8.5045/1.8939) mem 68106MB [2022-12-19 23:48:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][700/1519] eta 0:13:44 lr 0.000020 time 0.9935 (1.0063) model_time 0.9934 (1.0052) loss 1.2582 (0.8471) grad_norm 7.9315 (8.4847/1.8941) mem 68106MB [2022-12-19 23:48:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][710/1519] eta 0:13:34 lr 0.000020 time 0.9327 (1.0062) model_time 0.9326 (1.0051) loss 0.7085 (0.8456) grad_norm 6.8940 (8.4930/1.8935) mem 68106MB [2022-12-19 23:48:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][720/1519] eta 0:13:23 lr 0.000020 time 0.9369 (1.0061) model_time 0.9368 (1.0051) loss 0.7853 (0.8463) grad_norm 9.3701 (8.5038/1.9002) mem 68106MB [2022-12-19 23:48:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][730/1519] eta 0:13:13 lr 0.000020 time 0.9350 (1.0061) model_time 0.9348 (1.0050) loss 0.8205 (0.8470) grad_norm 8.0757 (8.5958/2.2511) mem 68106MB [2022-12-19 23:48:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][740/1519] eta 0:13:03 lr 0.000020 time 0.9285 (1.0060) model_time 0.9284 (1.0049) loss 1.3244 (0.8475) grad_norm 8.8111 (8.5837/2.2568) mem 68106MB [2022-12-19 23:49:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][750/1519] eta 0:12:53 lr 0.000020 time 0.9633 (1.0059) model_time 0.9632 (1.0049) loss 0.7112 (0.8468) grad_norm 9.7362 (8.6005/2.2638) mem 68106MB [2022-12-19 23:49:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][760/1519] eta 0:12:43 lr 0.000020 time 0.9403 (1.0059) model_time 0.9401 (1.0049) loss 0.6933 (0.8464) grad_norm 11.0512 (8.6497/2.2624) mem 68106MB [2022-12-19 23:49:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][770/1519] eta 0:12:33 lr 0.000020 time 0.9100 (1.0060) model_time 0.9099 (1.0049) loss 0.8291 (0.8473) grad_norm 10.2975 (8.6670/2.2738) mem 68106MB [2022-12-19 23:49:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][780/1519] eta 0:12:23 lr 0.000020 time 0.9319 (1.0058) model_time 0.9317 (1.0048) loss 0.8483 (0.8475) grad_norm 9.2213 (8.6780/2.2937) mem 68106MB [2022-12-19 23:49:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][790/1519] eta 0:12:13 lr 0.000020 time 0.9331 (1.0057) model_time 0.9329 (1.0047) loss 0.7143 (0.8471) grad_norm 6.7867 (8.6657/2.3004) mem 68106MB [2022-12-19 23:49:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][800/1519] eta 0:12:03 lr 0.000020 time 0.9109 (1.0057) model_time 0.9108 (1.0047) loss 0.7473 (0.8467) grad_norm 9.1905 (8.6542/2.2916) mem 68106MB [2022-12-19 23:50:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][810/1519] eta 0:11:52 lr 0.000020 time 0.9273 (1.0056) model_time 0.9272 (1.0046) loss 1.0713 (0.8465) grad_norm 12.1617 (8.6549/2.3118) mem 68106MB [2022-12-19 23:50:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][820/1519] eta 0:11:42 lr 0.000020 time 0.9354 (1.0056) model_time 0.9353 (1.0046) loss 1.0253 (0.8466) grad_norm 6.6282 (8.6673/2.3122) mem 68106MB [2022-12-19 23:50:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][830/1519] eta 0:11:32 lr 0.000020 time 0.9330 (1.0057) model_time 0.9328 (1.0047) loss 0.7131 (0.8458) grad_norm 8.5371 (8.6665/2.3102) mem 68106MB [2022-12-19 23:50:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][840/1519] eta 0:11:22 lr 0.000020 time 0.9471 (1.0058) model_time 0.9470 (1.0048) loss 0.7982 (0.8460) grad_norm 9.9722 (8.6821/2.3086) mem 68106MB [2022-12-19 23:50:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][850/1519] eta 0:11:12 lr 0.000020 time 0.9277 (1.0058) model_time 0.9275 (1.0048) loss 1.0374 (0.8468) grad_norm 10.1777 (8.7125/2.3270) mem 68106MB [2022-12-19 23:50:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][860/1519] eta 0:11:02 lr 0.000020 time 0.9211 (1.0057) model_time 0.9209 (1.0047) loss 0.7622 (0.8465) grad_norm 7.1916 (8.6996/2.3210) mem 68106MB [2022-12-19 23:51:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][870/1519] eta 0:10:52 lr 0.000020 time 0.9291 (1.0058) model_time 0.9290 (1.0049) loss 0.9145 (0.8462) grad_norm 6.7899 (8.7141/2.3180) mem 68106MB [2022-12-19 23:51:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][880/1519] eta 0:10:42 lr 0.000020 time 0.9172 (1.0057) model_time 0.9171 (1.0048) loss 0.7923 (0.8464) grad_norm 8.4492 (8.7150/2.3177) mem 68106MB [2022-12-19 23:51:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][890/1519] eta 0:10:32 lr 0.000020 time 0.9294 (1.0056) model_time 0.9293 (1.0047) loss 1.3534 (0.8462) grad_norm 9.9660 (8.7205/2.3083) mem 68106MB [2022-12-19 23:51:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][900/1519] eta 0:10:22 lr 0.000020 time 0.9277 (1.0056) model_time 0.9275 (1.0047) loss 0.6738 (0.8471) grad_norm 10.3076 (8.7095/2.2867) mem 68106MB [2022-12-19 23:51:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][910/1519] eta 0:10:12 lr 0.000020 time 0.9305 (1.0055) model_time 0.9303 (1.0046) loss 0.7309 (0.8465) grad_norm 11.6817 (8.7172/2.2865) mem 68106MB [2022-12-19 23:51:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][920/1519] eta 0:10:02 lr 0.000020 time 0.9317 (1.0055) model_time 0.9315 (1.0046) loss 0.7187 (0.8461) grad_norm 6.3185 (8.6905/2.2973) mem 68106MB [2022-12-19 23:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][930/1519] eta 0:09:52 lr 0.000020 time 0.9381 (1.0055) model_time 0.9379 (1.0045) loss 0.8162 (0.8460) grad_norm 8.1090 (8.6604/2.1388) mem 68106MB [2022-12-19 23:52:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][940/1519] eta 0:09:42 lr 0.000020 time 0.9318 (1.0054) model_time 0.9316 (1.0045) loss 0.8468 (0.8463) grad_norm 8.7373 (8.6832/2.1348) mem 68106MB [2022-12-19 23:52:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][950/1519] eta 0:09:32 lr 0.000020 time 0.9211 (1.0055) model_time 0.9210 (1.0046) loss 0.9849 (0.8469) grad_norm 6.6810 (8.7085/2.1501) mem 68106MB [2022-12-19 23:52:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][960/1519] eta 0:09:22 lr 0.000020 time 0.9201 (1.0055) model_time 0.9200 (1.0046) loss 0.9090 (0.8474) grad_norm 14.6560 (8.7104/2.1823) mem 68106MB [2022-12-19 23:52:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][970/1519] eta 0:09:12 lr 0.000020 time 0.9239 (1.0055) model_time 0.9238 (1.0046) loss 0.6825 (0.8486) grad_norm 9.7331 (8.7096/2.1847) mem 68106MB [2022-12-19 23:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][980/1519] eta 0:09:01 lr 0.000020 time 0.9246 (1.0055) model_time 0.9245 (1.0046) loss 0.7684 (0.8484) grad_norm 7.6429 (8.7234/2.2058) mem 68106MB [2022-12-19 23:53:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][990/1519] eta 0:08:51 lr 0.000020 time 0.9272 (1.0054) model_time 0.9270 (1.0046) loss 0.7883 (0.8480) grad_norm 12.0405 (8.7793/2.2331) mem 68106MB [2022-12-19 23:53:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1000/1519] eta 0:08:41 lr 0.000020 time 0.9195 (1.0054) model_time 0.9194 (1.0045) loss 0.9363 (0.8484) grad_norm 6.7578 (8.7992/2.2528) mem 68106MB [2022-12-19 23:53:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1010/1519] eta 0:08:31 lr 0.000020 time 0.9242 (1.0054) model_time 0.9240 (1.0045) loss 0.7151 (0.8485) grad_norm 7.4064 (8.7823/2.2571) mem 68106MB [2022-12-19 23:53:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1020/1519] eta 0:08:21 lr 0.000020 time 0.9240 (1.0055) model_time 0.9239 (1.0046) loss 0.9131 (0.8484) grad_norm 9.3196 (8.8176/2.2584) mem 68106MB [2022-12-19 23:53:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1030/1519] eta 0:08:11 lr 0.000020 time 0.9265 (1.0054) model_time 0.9264 (1.0046) loss 0.7923 (0.8493) grad_norm 10.2582 (8.8042/2.2524) mem 68106MB [2022-12-19 23:53:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1040/1519] eta 0:08:01 lr 0.000020 time 0.9308 (1.0054) model_time 0.9305 (1.0045) loss 0.7785 (0.8493) grad_norm 8.7220 (8.7872/2.2371) mem 68106MB [2022-12-19 23:54:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1050/1519] eta 0:07:51 lr 0.000020 time 0.9286 (1.0053) model_time 0.9285 (1.0045) loss 0.7503 (0.8500) grad_norm 8.3407 (8.7881/2.2377) mem 68106MB [2022-12-19 23:54:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1060/1519] eta 0:07:41 lr 0.000020 time 0.9296 (1.0052) model_time 0.9294 (1.0044) loss 0.8706 (0.8503) grad_norm 6.5605 (8.7681/2.2430) mem 68106MB [2022-12-19 23:54:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1070/1519] eta 0:07:31 lr 0.000020 time 0.9912 (1.0054) model_time 0.9910 (1.0045) loss 0.7969 (0.8496) grad_norm 5.0976 (8.7277/2.2486) mem 68106MB [2022-12-19 23:54:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1080/1519] eta 0:07:21 lr 0.000020 time 0.9200 (1.0053) model_time 0.9198 (1.0045) loss 0.8673 (0.8499) grad_norm 7.0146 (8.7493/2.2783) mem 68106MB [2022-12-19 23:54:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1090/1519] eta 0:07:11 lr 0.000020 time 0.9351 (1.0053) model_time 0.9343 (1.0045) loss 0.7420 (0.8502) grad_norm 8.5919 (8.7468/2.2647) mem 68106MB [2022-12-19 23:54:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1100/1519] eta 0:07:01 lr 0.000020 time 0.9471 (1.0053) model_time 0.9470 (1.0045) loss 0.8214 (0.8498) grad_norm 7.2216 (8.7451/2.2639) mem 68106MB [2022-12-19 23:55:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1110/1519] eta 0:06:51 lr 0.000020 time 0.9307 (1.0053) model_time 0.9305 (1.0045) loss 0.9250 (0.8502) grad_norm 9.0323 (8.7558/2.2421) mem 68106MB [2022-12-19 23:55:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1120/1519] eta 0:06:41 lr 0.000020 time 0.9362 (1.0053) model_time 0.9360 (1.0044) loss 0.7624 (0.8505) grad_norm 15.2091 (8.7758/2.2721) mem 68106MB [2022-12-19 23:55:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1130/1519] eta 0:06:31 lr 0.000020 time 0.9306 (1.0054) model_time 0.9304 (1.0046) loss 0.7488 (0.8510) grad_norm 12.3681 (8.8171/2.2641) mem 68106MB [2022-12-19 23:55:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1140/1519] eta 0:06:21 lr 0.000020 time 0.9264 (1.0054) model_time 0.9263 (1.0046) loss 0.7462 (0.8513) grad_norm 6.7717 (8.7963/2.2786) mem 68106MB [2022-12-19 23:55:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1150/1519] eta 0:06:11 lr 0.000020 time 0.9344 (1.0055) model_time 0.9343 (1.0046) loss 0.8329 (0.8514) grad_norm 8.6715 (8.7874/2.2784) mem 68106MB [2022-12-19 23:55:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1160/1519] eta 0:06:01 lr 0.000020 time 0.9810 (1.0057) model_time 0.9809 (1.0049) loss 1.2151 (0.8519) grad_norm 8.2023 (8.7917/2.2699) mem 68106MB [2022-12-19 23:56:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1170/1519] eta 0:05:50 lr 0.000020 time 0.9329 (1.0056) model_time 0.9328 (1.0048) loss 0.7352 (0.8510) grad_norm 8.3030 (8.7473/2.2708) mem 68106MB [2022-12-19 23:56:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1180/1519] eta 0:05:40 lr 0.000020 time 0.9139 (1.0058) model_time 0.9137 (1.0050) loss 1.0234 (0.8508) grad_norm 10.9065 (8.7434/2.2705) mem 68106MB [2022-12-19 23:56:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1190/1519] eta 0:05:30 lr 0.000020 time 0.9309 (1.0058) model_time 0.9308 (1.0050) loss 0.8250 (0.8514) grad_norm 8.7631 (8.7280/2.2617) mem 68106MB [2022-12-19 23:56:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1200/1519] eta 0:05:20 lr 0.000020 time 0.9326 (1.0058) model_time 0.9325 (1.0050) loss 0.7805 (0.8508) grad_norm 7.8523 (8.7295/2.2686) mem 68106MB [2022-12-19 23:56:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1210/1519] eta 0:05:10 lr 0.000020 time 0.9315 (1.0058) model_time 0.9312 (1.0050) loss 1.0129 (0.8512) grad_norm 7.1003 (8.7140/2.2663) mem 68106MB [2022-12-19 23:56:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1220/1519] eta 0:05:00 lr 0.000020 time 0.9306 (1.0057) model_time 0.9305 (1.0049) loss 0.9176 (0.8514) grad_norm 9.2905 (8.7182/2.2670) mem 68106MB [2022-12-19 23:57:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1230/1519] eta 0:04:50 lr 0.000020 time 0.9222 (1.0056) model_time 0.9220 (1.0049) loss 0.7069 (0.8511) grad_norm 7.4153 (8.7505/2.2657) mem 68106MB [2022-12-19 23:57:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1240/1519] eta 0:04:40 lr 0.000020 time 0.9184 (1.0056) model_time 0.9180 (1.0048) loss 1.0903 (0.8511) grad_norm 8.5913 (8.7458/2.2651) mem 68106MB [2022-12-19 23:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1250/1519] eta 0:04:30 lr 0.000020 time 1.0112 (1.0056) model_time 1.0111 (1.0048) loss 0.8649 (0.8517) grad_norm 9.2464 (8.7335/2.2564) mem 68106MB [2022-12-19 23:57:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1260/1519] eta 0:04:20 lr 0.000020 time 0.9887 (1.0056) model_time 0.9881 (1.0048) loss 0.6831 (0.8514) grad_norm 11.7320 (8.6993/2.2382) mem 68106MB [2022-12-19 23:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1270/1519] eta 0:04:10 lr 0.000020 time 1.0246 (1.0056) model_time 1.0244 (1.0048) loss 0.9300 (0.8511) grad_norm 7.3280 (8.6915/2.2298) mem 68106MB [2022-12-19 23:57:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1280/1519] eta 0:04:00 lr 0.000020 time 0.9449 (1.0056) model_time 0.9448 (1.0048) loss 0.7344 (0.8511) grad_norm 5.9509 (8.7187/2.2754) mem 68106MB [2022-12-19 23:58:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1290/1519] eta 0:03:50 lr 0.000020 time 0.9276 (1.0056) model_time 0.9273 (1.0048) loss 0.8361 (0.8517) grad_norm 8.6661 (8.7311/2.2755) mem 68106MB [2022-12-19 23:58:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1300/1519] eta 0:03:40 lr 0.000020 time 0.9328 (1.0055) model_time 0.9326 (1.0047) loss 0.6849 (0.8519) grad_norm 6.5337 (8.7544/2.2958) mem 68106MB [2022-12-19 23:58:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1310/1519] eta 0:03:30 lr 0.000020 time 0.9327 (1.0056) model_time 0.9325 (1.0049) loss 0.8646 (0.8519) grad_norm 7.7885 (8.7560/2.2959) mem 68106MB [2022-12-19 23:58:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1320/1519] eta 0:03:20 lr 0.000020 time 0.9311 (1.0056) model_time 0.9310 (1.0048) loss 0.7829 (0.8520) grad_norm 8.9937 (8.7474/2.2926) mem 68106MB [2022-12-19 23:58:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1330/1519] eta 0:03:10 lr 0.000020 time 0.9289 (1.0057) model_time 0.9288 (1.0049) loss 1.1570 (0.8524) grad_norm 8.6981 (8.6594/1.9451) mem 68106MB [2022-12-19 23:58:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1340/1519] eta 0:03:00 lr 0.000020 time 0.9728 (1.0056) model_time 0.9726 (1.0049) loss 0.6950 (0.8528) grad_norm 10.0683 (8.6694/1.9351) mem 68106MB [2022-12-19 23:59:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1350/1519] eta 0:02:49 lr 0.000020 time 0.9415 (1.0056) model_time 0.9414 (1.0049) loss 0.8560 (0.8528) grad_norm 8.0396 (8.6277/1.9087) mem 68106MB [2022-12-19 23:59:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1360/1519] eta 0:02:39 lr 0.000020 time 0.9308 (1.0056) model_time 0.9307 (1.0048) loss 1.1004 (0.8529) grad_norm 7.8677 (8.5928/1.9015) mem 68106MB [2022-12-19 23:59:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1370/1519] eta 0:02:29 lr 0.000020 time 0.9370 (1.0055) model_time 0.9366 (1.0048) loss 0.6848 (0.8526) grad_norm 8.0855 (8.5726/1.8914) mem 68106MB [2022-12-19 23:59:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1380/1519] eta 0:02:19 lr 0.000020 time 0.9349 (1.0055) model_time 0.9348 (1.0048) loss 0.9566 (0.8523) grad_norm 8.4183 (8.5426/1.8669) mem 68106MB [2022-12-19 23:59:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1390/1519] eta 0:02:09 lr 0.000020 time 0.9286 (1.0055) model_time 0.9285 (1.0048) loss 0.6931 (0.8519) grad_norm 8.8670 (8.5500/1.8606) mem 68106MB [2022-12-19 23:59:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1400/1519] eta 0:01:59 lr 0.000020 time 0.9375 (1.0056) model_time 0.9373 (1.0048) loss 0.7458 (0.8516) grad_norm 9.4010 (8.5517/1.8613) mem 68106MB [2022-12-20 00:00:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1410/1519] eta 0:01:49 lr 0.000020 time 0.9277 (1.0055) model_time 0.9275 (1.0048) loss 0.7551 (0.8516) grad_norm 7.1246 (8.5369/1.8263) mem 68106MB [2022-12-20 00:00:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1420/1519] eta 0:01:39 lr 0.000020 time 0.9268 (1.0056) model_time 0.9266 (1.0048) loss 0.9194 (0.8517) grad_norm 8.2559 (8.5544/1.8398) mem 68106MB [2022-12-20 00:00:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1430/1519] eta 0:01:29 lr 0.000020 time 0.9272 (1.0056) model_time 0.9270 (1.0048) loss 0.7980 (0.8513) grad_norm 9.5176 (8.5579/1.8393) mem 68106MB [2022-12-20 00:00:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1440/1519] eta 0:01:19 lr 0.000020 time 1.0023 (1.0056) model_time 1.0022 (1.0049) loss 1.0800 (0.8515) grad_norm 8.3383 (8.5423/1.8410) mem 68106MB [2022-12-20 00:00:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1450/1519] eta 0:01:09 lr 0.000020 time 1.0024 (1.0056) model_time 1.0021 (1.0049) loss 0.9820 (0.8514) grad_norm 9.0034 (8.5434/1.8182) mem 68106MB [2022-12-20 00:01:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1460/1519] eta 0:00:59 lr 0.000020 time 1.0532 (1.0057) model_time 1.0531 (1.0050) loss 0.7508 (0.8508) grad_norm 8.9457 (8.5433/1.8159) mem 68106MB [2022-12-20 00:01:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1470/1519] eta 0:00:49 lr 0.000020 time 0.9182 (1.0057) model_time 0.9180 (1.0050) loss 0.6915 (0.8511) grad_norm 7.2263 (8.5534/1.8300) mem 68106MB [2022-12-20 00:01:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1480/1519] eta 0:00:39 lr 0.000020 time 0.9334 (1.0057) model_time 0.9332 (1.0049) loss 0.6768 (0.8509) grad_norm 7.4631 (8.5577/1.8336) mem 68106MB [2022-12-20 00:01:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1490/1519] eta 0:00:29 lr 0.000020 time 0.9356 (1.0056) model_time 0.9354 (1.0049) loss 0.6901 (0.8506) grad_norm 7.1206 (8.5429/1.8392) mem 68106MB [2022-12-20 00:01:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1500/1519] eta 0:00:19 lr 0.000020 time 0.9327 (1.0056) model_time 0.9325 (1.0049) loss 0.7904 (0.8508) grad_norm 9.3473 (8.5165/1.8349) mem 68106MB [2022-12-20 00:01:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [44/100][1510/1519] eta 0:00:09 lr 0.000020 time 0.9142 (1.0056) model_time 0.9141 (1.0049) loss 0.7734 (0.8508) grad_norm 9.2300 (8.5189/1.8153) mem 68106MB [2022-12-20 00:01:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 44 training takes 0:25:27 [2022-12-20 00:01:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_44.pth saving...... [2022-12-20 00:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_44.pth saved !!! [2022-12-20 00:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.649 (0.649) Loss 0.5426 (0.5426) Acc@1 90.278 (90.278) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 00:02:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.329) Loss 0.5182 (0.4980) Acc@1 91.667 (92.045) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 00:02:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.313) Loss 0.4393 (0.4931) Acc@1 93.056 (92.163) Acc@5 98.958 (98.462) Mem 68106MB [2022-12-20 00:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.308) Loss 0.6145 (0.4999) Acc@1 90.278 (91.891) Acc@5 97.569 (98.331) Mem 68106MB [2022-12-20 00:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.305) Loss 0.4858 (0.4918) Acc@1 90.972 (91.980) Acc@5 98.264 (98.408) Mem 68106MB [2022-12-20 00:02:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.304 (0.305) Loss 0.4975 (0.4901) Acc@1 89.583 (91.898) Acc@5 99.306 (98.489) Mem 68106MB [2022-12-20 00:02:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.5143 (0.4900) Acc@1 90.625 (91.889) Acc@5 98.264 (98.463) Mem 68106MB [2022-12-20 00:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5174 (0.4907) Acc@1 93.056 (91.828) Acc@5 98.264 (98.455) Mem 68106MB [2022-12-20 00:02:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.300 (0.302) Loss 0.4120 (0.4887) Acc@1 93.403 (91.868) Acc@5 98.958 (98.470) Mem 68106MB [2022-12-20 00:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:44] * Acc@1 91.830 Acc@5 98.477 [2022-12-20 00:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.8% [2022-12-20 00:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 91.90% [2022-12-20 00:02:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][0/1519] eta 0:47:23 lr 0.000020 time 1.8717 (1.8717) model_time 1.1316 (1.1316) loss 0.7533 (0.7533) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 00:03:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][10/1519] eta 0:27:06 lr 0.000020 time 0.9267 (1.0779) model_time 0.9263 (1.0102) loss 0.7617 (0.8195) grad_norm 8.7291 (8.8076/0.9169) mem 68106MB [2022-12-20 00:03:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][20/1519] eta 0:26:01 lr 0.000020 time 0.9415 (1.0418) model_time 0.9414 (1.0062) loss 0.8848 (0.8202) grad_norm 9.3740 (8.7025/0.9297) mem 68106MB [2022-12-20 00:03:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][30/1519] eta 0:25:32 lr 0.000020 time 0.9292 (1.0292) model_time 0.9290 (1.0049) loss 1.1616 (0.8216) grad_norm 8.7478 (8.6380/0.8895) mem 68106MB [2022-12-20 00:03:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][40/1519] eta 0:25:11 lr 0.000020 time 0.9352 (1.0219) model_time 0.9351 (1.0034) loss 0.7923 (0.8191) grad_norm 9.0940 (8.4919/1.0223) mem 68106MB [2022-12-20 00:03:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][50/1519] eta 0:25:01 lr 0.000020 time 0.9260 (1.0222) model_time 0.9255 (1.0072) loss 1.0743 (0.8209) grad_norm 11.5106 (8.5658/1.1581) mem 68106MB [2022-12-20 00:03:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][60/1519] eta 0:24:46 lr 0.000020 time 0.9262 (1.0185) model_time 0.9261 (1.0059) loss 0.7870 (0.8238) grad_norm 7.5651 (8.4368/1.1710) mem 68106MB [2022-12-20 00:04:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][70/1519] eta 0:24:32 lr 0.000020 time 0.9220 (1.0160) model_time 0.9218 (1.0051) loss 1.0761 (0.8220) grad_norm 8.4250 (8.4345/1.3544) mem 68106MB [2022-12-20 00:04:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][80/1519] eta 0:24:20 lr 0.000020 time 0.9367 (1.0147) model_time 0.9365 (1.0051) loss 0.7416 (0.8136) grad_norm 7.3920 (8.3966/1.3392) mem 68106MB [2022-12-20 00:04:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][90/1519] eta 0:24:07 lr 0.000020 time 0.9314 (1.0132) model_time 0.9311 (1.0047) loss 1.1980 (0.8244) grad_norm 7.9480 (8.3958/1.3169) mem 68106MB [2022-12-20 00:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][100/1519] eta 0:23:59 lr 0.000020 time 0.9325 (1.0144) model_time 0.9323 (1.0066) loss 0.8616 (0.8235) grad_norm 8.1217 (8.4277/1.2694) mem 68106MB [2022-12-20 00:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][110/1519] eta 0:23:48 lr 0.000020 time 1.0142 (1.0142) model_time 1.0141 (1.0071) loss 0.9360 (0.8304) grad_norm 8.2387 (8.5231/1.3432) mem 68106MB [2022-12-20 00:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][120/1519] eta 0:23:37 lr 0.000020 time 0.9257 (1.0131) model_time 0.9253 (1.0066) loss 0.9953 (0.8315) grad_norm 8.5724 (8.4985/1.3604) mem 68106MB [2022-12-20 00:05:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][130/1519] eta 0:23:25 lr 0.000020 time 0.9300 (1.0122) model_time 0.9299 (1.0061) loss 0.6951 (0.8379) grad_norm 7.5947 (8.4919/1.4743) mem 68106MB [2022-12-20 00:05:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][140/1519] eta 0:23:14 lr 0.000020 time 0.9397 (1.0115) model_time 0.9395 (1.0058) loss 0.9279 (0.8411) grad_norm 6.9862 (8.4003/1.5074) mem 68106MB [2022-12-20 00:05:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][150/1519] eta 0:23:07 lr 0.000020 time 0.9305 (1.0132) model_time 0.9303 (1.0079) loss 0.9039 (0.8419) grad_norm 7.9182 (8.3922/1.4705) mem 68106MB [2022-12-20 00:05:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][160/1519] eta 0:22:56 lr 0.000020 time 1.0056 (1.0131) model_time 1.0054 (1.0080) loss 1.0533 (0.8407) grad_norm 7.6596 (8.3833/1.4360) mem 68106MB [2022-12-20 00:05:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][170/1519] eta 0:22:45 lr 0.000020 time 0.9341 (1.0122) model_time 0.9340 (1.0074) loss 0.6957 (0.8436) grad_norm 8.1679 (8.5051/1.5848) mem 68106MB [2022-12-20 00:05:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][180/1519] eta 0:22:34 lr 0.000020 time 0.9303 (1.0118) model_time 0.9298 (1.0073) loss 1.0963 (0.8417) grad_norm 7.7697 (8.5645/1.8220) mem 68106MB [2022-12-20 00:06:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][190/1519] eta 0:22:23 lr 0.000020 time 0.9312 (1.0112) model_time 0.9310 (1.0069) loss 0.8821 (0.8396) grad_norm 7.2385 (8.6512/1.8800) mem 68106MB [2022-12-20 00:06:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][200/1519] eta 0:22:13 lr 0.000020 time 0.9389 (1.0106) model_time 0.9387 (1.0065) loss 0.7391 (0.8412) grad_norm 8.5027 (8.6508/1.8354) mem 68106MB [2022-12-20 00:06:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][210/1519] eta 0:22:03 lr 0.000020 time 0.9364 (1.0111) model_time 0.9363 (1.0072) loss 0.8789 (0.8417) grad_norm 7.8567 (8.6607/1.8733) mem 68106MB [2022-12-20 00:06:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][220/1519] eta 0:21:52 lr 0.000020 time 0.9362 (1.0106) model_time 0.9360 (1.0068) loss 0.9077 (0.8416) grad_norm 12.4566 (8.6593/1.8796) mem 68106MB [2022-12-20 00:06:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][230/1519] eta 0:21:42 lr 0.000020 time 1.0381 (1.0105) model_time 1.0380 (1.0069) loss 1.1923 (0.8446) grad_norm 6.9647 (8.6603/1.9129) mem 68106MB [2022-12-20 00:06:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][240/1519] eta 0:21:32 lr 0.000020 time 0.9349 (1.0103) model_time 0.9347 (1.0068) loss 1.2032 (0.8483) grad_norm 9.3634 (8.6520/1.8922) mem 68106MB [2022-12-20 00:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][250/1519] eta 0:21:21 lr 0.000020 time 0.9517 (1.0101) model_time 0.9515 (1.0067) loss 0.7662 (0.8475) grad_norm 10.4111 (8.6687/1.8825) mem 68106MB [2022-12-20 00:07:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][260/1519] eta 0:21:11 lr 0.000020 time 0.9300 (1.0098) model_time 0.9299 (1.0066) loss 0.7114 (0.8479) grad_norm 10.5071 (8.6526/1.8620) mem 68106MB [2022-12-20 00:07:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][270/1519] eta 0:21:01 lr 0.000020 time 0.9316 (1.0097) model_time 0.9312 (1.0066) loss 0.9231 (0.8478) grad_norm 8.5801 (8.6285/1.8731) mem 68106MB [2022-12-20 00:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][280/1519] eta 0:20:50 lr 0.000020 time 0.9381 (1.0095) model_time 0.9379 (1.0064) loss 0.7891 (0.8474) grad_norm 7.8592 (8.6576/1.8716) mem 68106MB [2022-12-20 00:07:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][290/1519] eta 0:20:40 lr 0.000020 time 1.0325 (1.0095) model_time 1.0323 (1.0066) loss 0.7676 (0.8458) grad_norm 8.9316 (8.6291/1.8591) mem 68106MB [2022-12-20 00:07:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][300/1519] eta 0:20:30 lr 0.000020 time 0.9092 (1.0093) model_time 0.9088 (1.0064) loss 0.8022 (0.8455) grad_norm 8.1210 (8.6106/1.8386) mem 68106MB [2022-12-20 00:08:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][310/1519] eta 0:20:20 lr 0.000020 time 0.9312 (1.0093) model_time 0.9310 (1.0065) loss 0.7233 (0.8481) grad_norm 5.4703 (8.5861/1.8388) mem 68106MB [2022-12-20 00:08:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][320/1519] eta 0:20:09 lr 0.000020 time 0.9304 (1.0090) model_time 0.9302 (1.0063) loss 0.6795 (0.8480) grad_norm 9.3133 (8.5749/1.8260) mem 68106MB [2022-12-20 00:08:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][330/1519] eta 0:19:59 lr 0.000020 time 0.9320 (1.0087) model_time 0.9318 (1.0060) loss 1.2122 (0.8491) grad_norm 6.8963 (8.5647/1.8255) mem 68106MB [2022-12-20 00:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][340/1519] eta 0:19:48 lr 0.000020 time 0.9297 (1.0084) model_time 0.9295 (1.0058) loss 0.8267 (0.8469) grad_norm 9.5717 (8.5787/1.8340) mem 68106MB [2022-12-20 00:08:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][350/1519] eta 0:19:38 lr 0.000020 time 0.9278 (1.0081) model_time 0.9277 (1.0055) loss 0.7706 (0.8464) grad_norm 8.3755 (8.5927/1.8125) mem 68106MB [2022-12-20 00:08:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][360/1519] eta 0:19:28 lr 0.000020 time 0.9349 (1.0081) model_time 0.9347 (1.0056) loss 0.8285 (0.8494) grad_norm 9.3108 (8.6030/1.7956) mem 68106MB [2022-12-20 00:09:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][370/1519] eta 0:19:18 lr 0.000020 time 0.9276 (1.0079) model_time 0.9273 (1.0055) loss 0.7493 (0.8474) grad_norm 9.4160 (8.5869/1.7926) mem 68106MB [2022-12-20 00:09:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][380/1519] eta 0:19:07 lr 0.000020 time 0.9318 (1.0078) model_time 0.9317 (1.0054) loss 0.8080 (0.8457) grad_norm 8.2171 (8.6075/1.7909) mem 68106MB [2022-12-20 00:09:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][390/1519] eta 0:18:57 lr 0.000020 time 0.9288 (1.0079) model_time 0.9286 (1.0055) loss 0.8798 (0.8445) grad_norm 7.7338 (8.5948/1.7894) mem 68106MB [2022-12-20 00:09:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][400/1519] eta 0:18:47 lr 0.000020 time 0.9314 (1.0076) model_time 0.9313 (1.0053) loss 0.7951 (0.8449) grad_norm 6.9407 (8.6225/1.8396) mem 68106MB [2022-12-20 00:09:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][410/1519] eta 0:18:37 lr 0.000020 time 1.0310 (1.0077) model_time 1.0308 (1.0055) loss 0.7704 (0.8454) grad_norm 7.9874 (8.6157/1.8248) mem 68106MB [2022-12-20 00:09:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][420/1519] eta 0:18:27 lr 0.000020 time 0.9301 (1.0075) model_time 0.9299 (1.0053) loss 0.7651 (0.8454) grad_norm 9.3601 (8.6157/1.8235) mem 68106MB [2022-12-20 00:10:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][430/1519] eta 0:18:17 lr 0.000020 time 0.9283 (1.0074) model_time 0.9280 (1.0052) loss 0.8213 (0.8456) grad_norm 11.5464 (8.6494/1.8433) mem 68106MB [2022-12-20 00:10:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][440/1519] eta 0:18:06 lr 0.000020 time 0.9326 (1.0072) model_time 0.9324 (1.0051) loss 0.9432 (0.8457) grad_norm 7.6717 (8.6266/1.8340) mem 68106MB [2022-12-20 00:10:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][450/1519] eta 0:17:56 lr 0.000020 time 0.9365 (1.0073) model_time 0.9364 (1.0052) loss 0.7391 (0.8448) grad_norm 7.2245 (8.6088/1.8263) mem 68106MB [2022-12-20 00:10:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][460/1519] eta 0:17:46 lr 0.000020 time 0.9409 (1.0075) model_time 0.9407 (1.0055) loss 0.6698 (0.8445) grad_norm 9.3836 (8.6290/1.8476) mem 68106MB [2022-12-20 00:10:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][470/1519] eta 0:17:36 lr 0.000019 time 0.9551 (1.0074) model_time 0.9549 (1.0054) loss 0.7424 (0.8439) grad_norm 6.9338 (8.6017/1.8400) mem 68106MB [2022-12-20 00:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][480/1519] eta 0:17:26 lr 0.000019 time 0.9328 (1.0076) model_time 0.9326 (1.0056) loss 1.1507 (0.8452) grad_norm 5.9577 (8.6019/1.8412) mem 68106MB [2022-12-20 00:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][490/1519] eta 0:17:16 lr 0.000019 time 0.9360 (1.0076) model_time 0.9358 (1.0057) loss 0.8318 (0.8448) grad_norm 6.0665 (8.6085/1.8626) mem 68106MB [2022-12-20 00:11:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][500/1519] eta 0:17:06 lr 0.000019 time 0.9313 (1.0075) model_time 0.9312 (1.0057) loss 0.7126 (0.8445) grad_norm 9.2682 (8.6129/1.8496) mem 68106MB [2022-12-20 00:11:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][510/1519] eta 0:16:56 lr 0.000019 time 0.9361 (1.0074) model_time 0.9360 (1.0055) loss 0.8072 (0.8437) grad_norm 7.5454 (8.5909/1.8392) mem 68106MB [2022-12-20 00:11:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][520/1519] eta 0:16:46 lr 0.000019 time 0.9336 (1.0072) model_time 0.9335 (1.0053) loss 0.8785 (0.8439) grad_norm 8.2775 (8.6031/1.8563) mem 68106MB [2022-12-20 00:11:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][530/1519] eta 0:16:35 lr 0.000019 time 0.9356 (1.0070) model_time 0.9355 (1.0052) loss 0.7765 (0.8449) grad_norm 11.3071 (8.6034/1.8625) mem 68106MB [2022-12-20 00:11:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][540/1519] eta 0:16:25 lr 0.000019 time 0.9398 (1.0070) model_time 0.9397 (1.0052) loss 0.7701 (0.8448) grad_norm 8.7791 (8.5894/1.8557) mem 68106MB [2022-12-20 00:12:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][550/1519] eta 0:16:15 lr 0.000019 time 0.9289 (1.0069) model_time 0.9287 (1.0051) loss 0.9534 (0.8450) grad_norm 11.3651 (8.6027/1.8630) mem 68106MB [2022-12-20 00:12:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][560/1519] eta 0:16:05 lr 0.000019 time 0.9311 (1.0069) model_time 0.9310 (1.0051) loss 0.8496 (0.8443) grad_norm 10.3338 (8.6405/1.8838) mem 68106MB [2022-12-20 00:12:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][570/1519] eta 0:15:55 lr 0.000019 time 0.9307 (1.0069) model_time 0.9306 (1.0052) loss 1.0419 (0.8441) grad_norm 7.8849 (8.6506/1.8836) mem 68106MB [2022-12-20 00:12:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][580/1519] eta 0:15:45 lr 0.000019 time 0.9414 (1.0068) model_time 0.9413 (1.0051) loss 0.9182 (0.8440) grad_norm 7.5673 (8.6268/1.8772) mem 68106MB [2022-12-20 00:12:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][590/1519] eta 0:15:35 lr 0.000019 time 1.0289 (1.0068) model_time 1.0288 (1.0051) loss 1.4099 (0.8447) grad_norm 8.6686 (8.6136/1.8663) mem 68106MB [2022-12-20 00:12:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][600/1519] eta 0:15:25 lr 0.000019 time 0.9301 (1.0066) model_time 0.9299 (1.0050) loss 0.7260 (0.8439) grad_norm 8.6050 (8.6069/1.8609) mem 68106MB [2022-12-20 00:13:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][610/1519] eta 0:15:14 lr 0.000019 time 0.9306 (1.0066) model_time 0.9304 (1.0050) loss 0.9189 (0.8431) grad_norm 10.8267 (8.6027/1.8733) mem 68106MB [2022-12-20 00:13:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][620/1519] eta 0:15:04 lr 0.000019 time 0.9318 (1.0066) model_time 0.9316 (1.0050) loss 0.9706 (0.8434) grad_norm 11.4670 (8.6170/1.9067) mem 68106MB [2022-12-20 00:13:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][630/1519] eta 0:14:54 lr 0.000019 time 0.9256 (1.0065) model_time 0.9255 (1.0049) loss 0.6949 (0.8440) grad_norm 11.7113 (8.6126/1.9225) mem 68106MB [2022-12-20 00:13:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][640/1519] eta 0:14:44 lr 0.000019 time 0.9290 (1.0064) model_time 0.9288 (1.0048) loss 1.0079 (0.8439) grad_norm 7.3608 (8.6201/1.9311) mem 68106MB [2022-12-20 00:13:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][650/1519] eta 0:14:34 lr 0.000019 time 0.9284 (1.0062) model_time 0.9282 (1.0047) loss 0.8048 (0.8445) grad_norm 12.8126 (8.6328/1.9726) mem 68106MB [2022-12-20 00:13:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][660/1519] eta 0:14:24 lr 0.000019 time 0.9418 (1.0061) model_time 0.9417 (1.0046) loss 0.8766 (0.8444) grad_norm 7.1509 (8.6487/1.9703) mem 68106MB [2022-12-20 00:14:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][670/1519] eta 0:14:14 lr 0.000019 time 0.9313 (1.0061) model_time 0.9312 (1.0046) loss 1.0542 (0.8453) grad_norm 8.5587 (8.6495/1.9607) mem 68106MB [2022-12-20 00:14:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][680/1519] eta 0:14:04 lr 0.000019 time 0.9341 (1.0060) model_time 0.9339 (1.0046) loss 0.7528 (0.8452) grad_norm 8.6495 (8.6455/1.9623) mem 68106MB [2022-12-20 00:14:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][690/1519] eta 0:13:53 lr 0.000019 time 0.9392 (1.0060) model_time 0.9390 (1.0045) loss 0.8125 (0.8443) grad_norm 7.9698 (8.6379/1.9620) mem 68106MB [2022-12-20 00:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][700/1519] eta 0:13:44 lr 0.000019 time 0.9287 (1.0063) model_time 0.9286 (1.0049) loss 1.0073 (0.8448) grad_norm 7.7605 (8.6242/1.9668) mem 68106MB [2022-12-20 00:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][710/1519] eta 0:13:34 lr 0.000019 time 0.9321 (1.0063) model_time 0.9319 (1.0048) loss 0.7256 (0.8451) grad_norm 9.2686 (8.6073/1.9596) mem 68106MB [2022-12-20 00:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][720/1519] eta 0:13:24 lr 0.000019 time 0.9368 (1.0063) model_time 0.9367 (1.0049) loss 1.0840 (0.8460) grad_norm 13.5105 (8.6368/1.9885) mem 68106MB [2022-12-20 00:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][730/1519] eta 0:13:14 lr 0.000019 time 0.9347 (1.0065) model_time 0.9346 (1.0051) loss 0.8558 (0.8467) grad_norm 6.9180 (8.6165/1.9795) mem 68106MB [2022-12-20 00:15:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][740/1519] eta 0:13:04 lr 0.000019 time 0.9328 (1.0065) model_time 0.9325 (1.0051) loss 0.8313 (0.8467) grad_norm 7.0620 (8.6317/1.9698) mem 68106MB [2022-12-20 00:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][750/1519] eta 0:12:54 lr 0.000019 time 0.9310 (1.0066) model_time 0.9309 (1.0053) loss 1.1007 (0.8469) grad_norm 6.3684 (8.6224/1.9803) mem 68106MB [2022-12-20 00:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][760/1519] eta 0:12:44 lr 0.000019 time 0.9319 (1.0067) model_time 0.9318 (1.0053) loss 0.9304 (0.8484) grad_norm 7.1405 (8.6298/1.9823) mem 68106MB [2022-12-20 00:15:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][770/1519] eta 0:12:34 lr 0.000019 time 1.1907 (1.0069) model_time 1.1905 (1.0055) loss 0.7287 (0.8488) grad_norm 8.3384 (8.6169/1.9785) mem 68106MB [2022-12-20 00:15:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][780/1519] eta 0:12:23 lr 0.000019 time 0.9449 (1.0067) model_time 0.9447 (1.0054) loss 0.7547 (0.8485) grad_norm 6.6497 (8.5843/1.9143) mem 68106MB [2022-12-20 00:16:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][790/1519] eta 0:12:14 lr 0.000019 time 0.9393 (1.0070) model_time 0.9392 (1.0056) loss 1.0283 (0.8492) grad_norm 21.0404 (8.6112/2.0290) mem 68106MB [2022-12-20 00:16:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][800/1519] eta 0:12:04 lr 0.000019 time 0.8853 (1.0070) model_time 0.8851 (1.0057) loss 0.7416 (0.8492) grad_norm 8.8395 (8.6305/2.0786) mem 68106MB [2022-12-20 00:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][810/1519] eta 0:11:53 lr 0.000019 time 0.9347 (1.0069) model_time 0.9345 (1.0056) loss 0.7023 (0.8493) grad_norm 6.5114 (8.6166/2.0584) mem 68106MB [2022-12-20 00:16:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][820/1519] eta 0:11:43 lr 0.000019 time 0.9669 (1.0069) model_time 0.9667 (1.0056) loss 0.7697 (0.8498) grad_norm 8.4246 (8.6281/2.0542) mem 68106MB [2022-12-20 00:16:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][830/1519] eta 0:11:33 lr 0.000019 time 0.9422 (1.0068) model_time 0.9421 (1.0055) loss 0.7411 (0.8495) grad_norm 7.7138 (8.6344/2.0516) mem 68106MB [2022-12-20 00:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][840/1519] eta 0:11:23 lr 0.000019 time 0.9339 (1.0067) model_time 0.9337 (1.0054) loss 0.6566 (0.8495) grad_norm 8.7734 (8.6299/2.0508) mem 68106MB [2022-12-20 00:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][850/1519] eta 0:11:13 lr 0.000019 time 0.9215 (1.0067) model_time 0.9214 (1.0054) loss 0.8943 (0.8498) grad_norm 10.6075 (8.6368/2.0512) mem 68106MB [2022-12-20 00:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][860/1519] eta 0:11:03 lr 0.000019 time 0.9365 (1.0066) model_time 0.9364 (1.0053) loss 0.9504 (0.8499) grad_norm 5.6250 (8.6483/2.0744) mem 68106MB [2022-12-20 00:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][870/1519] eta 0:10:53 lr 0.000019 time 0.9298 (1.0065) model_time 0.9296 (1.0053) loss 0.7303 (0.8495) grad_norm 9.8701 (8.6605/2.0580) mem 68106MB [2022-12-20 00:17:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][880/1519] eta 0:10:43 lr 0.000019 time 0.9387 (1.0065) model_time 0.9385 (1.0052) loss 0.7494 (0.8484) grad_norm 11.1731 (8.6648/2.0528) mem 68106MB [2022-12-20 00:17:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][890/1519] eta 0:10:33 lr 0.000019 time 0.9405 (1.0064) model_time 0.9403 (1.0052) loss 0.9859 (0.8492) grad_norm 8.4590 (8.6730/2.0539) mem 68106MB [2022-12-20 00:17:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][900/1519] eta 0:10:22 lr 0.000019 time 0.9365 (1.0064) model_time 0.9363 (1.0052) loss 0.9327 (0.8492) grad_norm 10.7025 (8.6865/2.0582) mem 68106MB [2022-12-20 00:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][910/1519] eta 0:10:12 lr 0.000019 time 0.9449 (1.0065) model_time 0.9447 (1.0053) loss 0.6825 (0.8489) grad_norm 9.0880 (8.6788/2.0549) mem 68106MB [2022-12-20 00:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][920/1519] eta 0:10:02 lr 0.000019 time 0.9288 (1.0064) model_time 0.9287 (1.0052) loss 1.1760 (0.8506) grad_norm 6.3935 (8.7191/2.1326) mem 68106MB [2022-12-20 00:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][930/1519] eta 0:09:52 lr 0.000019 time 0.9361 (1.0064) model_time 0.9359 (1.0053) loss 0.7843 (0.8502) grad_norm 12.0273 (8.7449/2.1317) mem 68106MB [2022-12-20 00:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][940/1519] eta 0:09:42 lr 0.000019 time 0.9166 (1.0064) model_time 0.9165 (1.0053) loss 0.9276 (0.8506) grad_norm 7.3534 (8.7544/2.1598) mem 68106MB [2022-12-20 00:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][950/1519] eta 0:09:32 lr 0.000019 time 0.9921 (1.0064) model_time 0.9919 (1.0053) loss 0.8889 (0.8506) grad_norm 8.6397 (8.7430/2.1601) mem 68106MB [2022-12-20 00:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][960/1519] eta 0:09:22 lr 0.000019 time 0.9286 (1.0064) model_time 0.9285 (1.0052) loss 0.9749 (0.8507) grad_norm 10.4209 (8.7493/2.1667) mem 68106MB [2022-12-20 00:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][970/1519] eta 0:09:12 lr 0.000019 time 0.9270 (1.0063) model_time 0.9268 (1.0051) loss 0.8678 (0.8502) grad_norm 8.6920 (8.7583/2.1676) mem 68106MB [2022-12-20 00:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][980/1519] eta 0:09:02 lr 0.000019 time 0.9325 (1.0063) model_time 0.9324 (1.0051) loss 0.7555 (0.8495) grad_norm 9.0658 (8.7503/2.1700) mem 68106MB [2022-12-20 00:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][990/1519] eta 0:08:52 lr 0.000019 time 0.9351 (1.0062) model_time 0.9350 (1.0051) loss 0.7557 (0.8491) grad_norm 11.0609 (8.7692/2.1907) mem 68106MB [2022-12-20 00:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1000/1519] eta 0:08:42 lr 0.000019 time 0.9378 (1.0062) model_time 0.9377 (1.0050) loss 0.6877 (0.8491) grad_norm 7.7290 (8.7372/2.1577) mem 68106MB [2022-12-20 00:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1010/1519] eta 0:08:32 lr 0.000019 time 0.9866 (1.0061) model_time 0.9863 (1.0050) loss 0.6956 (0.8495) grad_norm 9.8014 (8.7371/2.1562) mem 68106MB [2022-12-20 00:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1020/1519] eta 0:08:22 lr 0.000019 time 0.9311 (1.0061) model_time 0.9309 (1.0050) loss 0.9564 (0.8499) grad_norm 7.3639 (8.7352/2.1481) mem 68106MB [2022-12-20 00:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1030/1519] eta 0:08:11 lr 0.000019 time 0.9252 (1.0060) model_time 0.9250 (1.0049) loss 0.6679 (0.8499) grad_norm 7.7466 (8.7205/2.1364) mem 68106MB [2022-12-20 00:20:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1040/1519] eta 0:08:01 lr 0.000019 time 0.9165 (1.0061) model_time 0.9164 (1.0050) loss 0.7176 (0.8497) grad_norm 8.3905 (8.7347/2.1312) mem 68106MB [2022-12-20 00:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1050/1519] eta 0:07:51 lr 0.000019 time 0.9137 (1.0061) model_time 0.9135 (1.0050) loss 0.8203 (0.8496) grad_norm 6.8539 (8.7407/2.1288) mem 68106MB [2022-12-20 00:20:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1060/1519] eta 0:07:41 lr 0.000019 time 0.9341 (1.0061) model_time 0.9340 (1.0050) loss 0.6910 (0.8493) grad_norm 8.0701 (8.7237/2.1062) mem 68106MB [2022-12-20 00:20:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1070/1519] eta 0:07:31 lr 0.000019 time 0.9353 (1.0061) model_time 0.9350 (1.0050) loss 0.6749 (0.8488) grad_norm 7.2518 (8.7485/2.1147) mem 68106MB [2022-12-20 00:20:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1080/1519] eta 0:07:21 lr 0.000019 time 0.9332 (1.0060) model_time 0.9331 (1.0049) loss 1.2693 (0.8488) grad_norm 10.7136 (8.7526/2.1173) mem 68106MB [2022-12-20 00:21:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1090/1519] eta 0:07:11 lr 0.000019 time 0.9341 (1.0060) model_time 0.9339 (1.0049) loss 0.7823 (0.8486) grad_norm 9.8095 (8.7572/2.0988) mem 68106MB [2022-12-20 00:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1100/1519] eta 0:07:01 lr 0.000019 time 0.9283 (1.0061) model_time 0.9282 (1.0050) loss 1.2247 (0.8485) grad_norm 14.4264 (8.7568/2.1298) mem 68106MB [2022-12-20 00:21:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1110/1519] eta 0:06:51 lr 0.000019 time 0.9392 (1.0061) model_time 0.9390 (1.0050) loss 0.8227 (0.8486) grad_norm 9.2881 (8.7882/2.1271) mem 68106MB [2022-12-20 00:21:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1120/1519] eta 0:06:41 lr 0.000019 time 0.9281 (1.0061) model_time 0.9276 (1.0050) loss 0.7084 (0.8487) grad_norm 7.2659 (8.7784/2.1170) mem 68106MB [2022-12-20 00:21:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1130/1519] eta 0:06:31 lr 0.000019 time 0.9415 (1.0060) model_time 0.9414 (1.0050) loss 1.0771 (0.8486) grad_norm 8.9030 (8.7932/2.1158) mem 68106MB [2022-12-20 00:21:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1140/1519] eta 0:06:21 lr 0.000019 time 0.9324 (1.0060) model_time 0.9323 (1.0049) loss 0.9623 (0.8489) grad_norm 9.6922 (8.7952/2.1178) mem 68106MB [2022-12-20 00:22:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1150/1519] eta 0:06:11 lr 0.000019 time 0.9328 (1.0059) model_time 0.9326 (1.0049) loss 0.7199 (0.8484) grad_norm 6.7746 (8.7729/2.1066) mem 68106MB [2022-12-20 00:22:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1160/1519] eta 0:06:01 lr 0.000019 time 0.9336 (1.0058) model_time 0.9334 (1.0048) loss 0.8141 (0.8483) grad_norm 6.8918 (8.7266/2.0850) mem 68106MB [2022-12-20 00:22:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1170/1519] eta 0:05:51 lr 0.000019 time 0.9316 (1.0058) model_time 0.9314 (1.0048) loss 0.7827 (0.8486) grad_norm 9.1464 (8.7370/2.0921) mem 68106MB [2022-12-20 00:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1180/1519] eta 0:05:40 lr 0.000019 time 0.9282 (1.0058) model_time 0.9280 (1.0048) loss 0.9454 (0.8481) grad_norm 7.5166 (8.7504/2.0869) mem 68106MB [2022-12-20 00:22:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1190/1519] eta 0:05:30 lr 0.000019 time 0.9368 (1.0057) model_time 0.9367 (1.0047) loss 0.6843 (0.8482) grad_norm 8.1154 (8.7738/2.1073) mem 68106MB [2022-12-20 00:22:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1200/1519] eta 0:05:20 lr 0.000019 time 0.9348 (1.0057) model_time 0.9347 (1.0047) loss 0.6914 (0.8478) grad_norm 9.2863 (8.7733/2.1032) mem 68106MB [2022-12-20 00:23:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1210/1519] eta 0:05:10 lr 0.000019 time 0.9184 (1.0057) model_time 0.9183 (1.0047) loss 1.2235 (0.8480) grad_norm 6.2655 (8.7603/2.0980) mem 68106MB [2022-12-20 00:23:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1220/1519] eta 0:05:00 lr 0.000019 time 0.9312 (1.0057) model_time 0.9311 (1.0047) loss 1.0425 (0.8478) grad_norm 7.8805 (8.7419/2.0745) mem 68106MB [2022-12-20 00:23:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1230/1519] eta 0:04:50 lr 0.000019 time 0.9393 (1.0057) model_time 0.9391 (1.0047) loss 0.7747 (0.8484) grad_norm 8.6202 (8.7686/2.0671) mem 68106MB [2022-12-20 00:23:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1240/1519] eta 0:04:40 lr 0.000019 time 0.9325 (1.0057) model_time 0.9324 (1.0048) loss 0.9252 (0.8484) grad_norm 8.4103 (8.7581/2.0599) mem 68106MB [2022-12-20 00:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1250/1519] eta 0:04:30 lr 0.000019 time 0.9367 (1.0057) model_time 0.9366 (1.0047) loss 0.6852 (0.8484) grad_norm 9.4181 (8.7363/2.0263) mem 68106MB [2022-12-20 00:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1260/1519] eta 0:04:20 lr 0.000019 time 0.9317 (1.0057) model_time 0.9315 (1.0047) loss 0.9132 (0.8484) grad_norm 5.6044 (8.7355/2.0371) mem 68106MB [2022-12-20 00:24:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1270/1519] eta 0:04:10 lr 0.000019 time 0.9376 (1.0058) model_time 0.9374 (1.0048) loss 1.1831 (0.8483) grad_norm 10.8374 (8.7662/2.0482) mem 68106MB [2022-12-20 00:24:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1280/1519] eta 0:04:00 lr 0.000019 time 0.9301 (1.0059) model_time 0.9299 (1.0050) loss 0.9809 (0.8482) grad_norm 5.8340 (8.7895/2.0728) mem 68106MB [2022-12-20 00:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1290/1519] eta 0:03:50 lr 0.000019 time 0.9323 (1.0059) model_time 0.9321 (1.0050) loss 1.1066 (0.8479) grad_norm 9.4844 (8.7967/2.0751) mem 68106MB [2022-12-20 00:24:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1300/1519] eta 0:03:40 lr 0.000019 time 0.9283 (1.0060) model_time 0.9282 (1.0050) loss 0.7545 (0.8478) grad_norm 9.3883 (8.8313/2.0839) mem 68106MB [2022-12-20 00:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1310/1519] eta 0:03:30 lr 0.000019 time 0.9364 (1.0059) model_time 0.9362 (1.0049) loss 0.8662 (0.8476) grad_norm 6.6343 (8.8423/2.0934) mem 68106MB [2022-12-20 00:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1320/1519] eta 0:03:20 lr 0.000019 time 0.9235 (1.0059) model_time 0.9233 (1.0049) loss 0.9774 (0.8474) grad_norm 7.6236 (8.8167/2.0664) mem 68106MB [2022-12-20 00:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1330/1519] eta 0:03:10 lr 0.000019 time 0.9283 (1.0058) model_time 0.9282 (1.0049) loss 0.8176 (0.8476) grad_norm 7.9740 (8.8280/2.0572) mem 68106MB [2022-12-20 00:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1340/1519] eta 0:03:00 lr 0.000019 time 0.9298 (1.0058) model_time 0.9297 (1.0049) loss 0.8048 (0.8471) grad_norm 6.8020 (8.8136/2.0610) mem 68106MB [2022-12-20 00:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1350/1519] eta 0:02:49 lr 0.000019 time 0.9348 (1.0058) model_time 0.9347 (1.0049) loss 0.9438 (0.8473) grad_norm 6.2417 (8.7944/2.0678) mem 68106MB [2022-12-20 00:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1360/1519] eta 0:02:39 lr 0.000019 time 0.9353 (1.0058) model_time 0.9352 (1.0049) loss 0.8210 (0.8476) grad_norm 8.4510 (8.7957/2.0696) mem 68106MB [2022-12-20 00:25:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1370/1519] eta 0:02:29 lr 0.000019 time 0.9677 (1.0059) model_time 0.9675 (1.0049) loss 0.9035 (0.8477) grad_norm 8.2932 (8.7758/2.0395) mem 68106MB [2022-12-20 00:25:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1380/1519] eta 0:02:19 lr 0.000019 time 0.9917 (1.0058) model_time 0.9916 (1.0049) loss 1.0862 (0.8480) grad_norm 7.9157 (8.7932/2.0332) mem 68106MB [2022-12-20 00:26:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1390/1519] eta 0:02:09 lr 0.000019 time 0.9317 (1.0058) model_time 0.9316 (1.0049) loss 0.6778 (0.8476) grad_norm 15.8936 (8.7674/1.9386) mem 68106MB [2022-12-20 00:26:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1400/1519] eta 0:01:59 lr 0.000019 time 0.9297 (1.0057) model_time 0.9296 (1.0048) loss 1.0760 (0.8481) grad_norm 9.8624 (8.7621/1.8898) mem 68106MB [2022-12-20 00:26:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1410/1519] eta 0:01:49 lr 0.000019 time 0.9345 (1.0057) model_time 0.9343 (1.0048) loss 1.1038 (0.8482) grad_norm 8.6190 (8.7767/1.8901) mem 68106MB [2022-12-20 00:26:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1420/1519] eta 0:01:39 lr 0.000019 time 0.9829 (1.0058) model_time 0.9827 (1.0049) loss 0.6671 (0.8482) grad_norm 7.3926 (8.7542/1.8851) mem 68106MB [2022-12-20 00:26:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1430/1519] eta 0:01:29 lr 0.000019 time 0.9284 (1.0057) model_time 0.9282 (1.0048) loss 1.1312 (0.8485) grad_norm 7.7425 (8.7513/1.8824) mem 68106MB [2022-12-20 00:26:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1440/1519] eta 0:01:19 lr 0.000019 time 0.9380 (1.0056) model_time 0.9378 (1.0047) loss 1.1790 (0.8484) grad_norm 5.9302 (8.7411/1.9011) mem 68106MB [2022-12-20 00:27:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1450/1519] eta 0:01:09 lr 0.000019 time 0.9267 (1.0056) model_time 0.9265 (1.0047) loss 0.6838 (0.8486) grad_norm 9.5988 (8.7339/1.9105) mem 68106MB [2022-12-20 00:27:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1460/1519] eta 0:00:59 lr 0.000019 time 0.9293 (1.0056) model_time 0.9291 (1.0047) loss 0.7282 (0.8485) grad_norm 7.8032 (8.7464/1.8978) mem 68106MB [2022-12-20 00:27:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1470/1519] eta 0:00:49 lr 0.000019 time 0.9318 (1.0055) model_time 0.9316 (1.0046) loss 0.7061 (0.8481) grad_norm 6.7256 (8.7242/1.9051) mem 68106MB [2022-12-20 00:27:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1480/1519] eta 0:00:39 lr 0.000019 time 0.9325 (1.0055) model_time 0.9324 (1.0046) loss 0.6874 (0.8480) grad_norm 7.3033 (8.7072/1.9108) mem 68106MB [2022-12-20 00:27:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1490/1519] eta 0:00:29 lr 0.000019 time 0.9314 (1.0055) model_time 0.9312 (1.0046) loss 0.7499 (0.8481) grad_norm 9.1552 (8.7036/1.9034) mem 68106MB [2022-12-20 00:27:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1500/1519] eta 0:00:19 lr 0.000019 time 0.9291 (1.0054) model_time 0.9290 (1.0046) loss 0.8143 (0.8485) grad_norm 7.8764 (8.6909/1.9043) mem 68106MB [2022-12-20 00:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [45/100][1510/1519] eta 0:00:09 lr 0.000019 time 0.9404 (1.0056) model_time 0.9403 (1.0047) loss 0.7233 (0.8481) grad_norm 8.2645 (8.6920/1.9019) mem 68106MB [2022-12-20 00:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 45 training takes 0:25:27 [2022-12-20 00:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_45.pth saving...... [2022-12-20 00:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_45.pth saved !!! [2022-12-20 00:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.688 (0.688) Loss 0.5250 (0.5250) Acc@1 90.625 (90.625) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 00:28:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.332) Loss 0.5135 (0.4869) Acc@1 91.667 (92.361) Acc@5 97.569 (98.453) Mem 68106MB [2022-12-20 00:28:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.316) Loss 0.4493 (0.4866) Acc@1 92.361 (92.212) Acc@5 99.306 (98.363) Mem 68106MB [2022-12-20 00:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.310) Loss 0.6099 (0.4924) Acc@1 88.889 (91.879) Acc@5 97.569 (98.297) Mem 68106MB [2022-12-20 00:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.307) Loss 0.4639 (0.4844) Acc@1 92.708 (91.997) Acc@5 98.264 (98.374) Mem 68106MB [2022-12-20 00:28:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.4808 (0.4820) Acc@1 91.319 (92.041) Acc@5 99.653 (98.414) Mem 68106MB [2022-12-20 00:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.5051 (0.4818) Acc@1 90.625 (92.105) Acc@5 97.917 (98.418) Mem 68106MB [2022-12-20 00:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.303) Loss 0.5096 (0.4823) Acc@1 92.361 (92.107) Acc@5 98.611 (98.430) Mem 68106MB [2022-12-20 00:29:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.300 (0.302) Loss 0.4156 (0.4799) Acc@1 93.403 (92.147) Acc@5 97.917 (98.440) Mem 68106MB [2022-12-20 00:29:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:45] * Acc@1 92.101 Acc@5 98.445 [2022-12-20 00:29:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.1% [2022-12-20 00:29:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 00:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 00:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.10% [2022-12-20 00:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][0/1519] eta 0:35:03 lr 0.000019 time 1.3850 (1.3850) model_time 0.9710 (0.9710) loss 0.6956 (0.6956) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 00:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][10/1519] eta 0:26:08 lr 0.000019 time 0.9193 (1.0395) model_time 0.9192 (1.0015) loss 0.8238 (0.7425) grad_norm 10.4562 (8.3538/1.2262) mem 68106MB [2022-12-20 00:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][20/1519] eta 0:25:38 lr 0.000019 time 0.9230 (1.0264) model_time 0.9229 (1.0063) loss 0.9627 (0.8153) grad_norm 8.3135 (8.1191/1.4841) mem 68106MB [2022-12-20 00:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][30/1519] eta 0:25:16 lr 0.000019 time 0.9258 (1.0187) model_time 0.9257 (1.0050) loss 0.7915 (0.8124) grad_norm 6.7178 (7.7964/1.4969) mem 68106MB [2022-12-20 00:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][40/1519] eta 0:24:58 lr 0.000019 time 0.9257 (1.0133) model_time 0.9255 (1.0029) loss 0.8211 (0.8166) grad_norm 7.7157 (8.0176/1.6603) mem 68106MB [2022-12-20 00:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][50/1519] eta 0:24:46 lr 0.000019 time 1.0315 (1.0119) model_time 1.0314 (1.0034) loss 0.8029 (0.8182) grad_norm 9.1216 (8.0734/1.5549) mem 68106MB [2022-12-20 00:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][60/1519] eta 0:24:34 lr 0.000019 time 0.9167 (1.0107) model_time 0.9166 (1.0036) loss 0.8032 (0.8148) grad_norm 10.7535 (8.1557/1.5411) mem 68106MB [2022-12-20 00:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][70/1519] eta 0:24:22 lr 0.000019 time 0.9170 (1.0095) model_time 0.9169 (1.0033) loss 0.8253 (0.8258) grad_norm 10.1761 (8.3321/1.5366) mem 68106MB [2022-12-20 00:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][80/1519] eta 0:24:13 lr 0.000019 time 0.9221 (1.0099) model_time 0.9220 (1.0045) loss 0.7509 (0.8243) grad_norm 7.9898 (8.2976/1.4827) mem 68106MB [2022-12-20 00:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][90/1519] eta 0:24:03 lr 0.000019 time 0.9301 (1.0101) model_time 0.9299 (1.0052) loss 1.3163 (0.8278) grad_norm 11.5549 (8.2874/1.5314) mem 68106MB [2022-12-20 00:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][100/1519] eta 0:23:55 lr 0.000019 time 0.9281 (1.0114) model_time 0.9280 (1.0070) loss 0.8764 (0.8301) grad_norm 6.2251 (8.1924/1.5442) mem 68106MB [2022-12-20 00:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][110/1519] eta 0:23:43 lr 0.000019 time 0.9316 (1.0103) model_time 0.9315 (1.0063) loss 0.7228 (0.8262) grad_norm 10.3466 (8.2939/1.6785) mem 68106MB [2022-12-20 00:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][120/1519] eta 0:23:32 lr 0.000019 time 0.9258 (1.0095) model_time 0.9256 (1.0057) loss 0.6963 (0.8222) grad_norm 6.7482 (8.2731/1.6595) mem 68106MB [2022-12-20 00:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][130/1519] eta 0:23:20 lr 0.000019 time 0.9228 (1.0085) model_time 0.9226 (1.0050) loss 0.8779 (0.8242) grad_norm 10.4877 (8.3819/1.7562) mem 68106MB [2022-12-20 00:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][140/1519] eta 0:23:10 lr 0.000019 time 0.9326 (1.0085) model_time 0.9318 (1.0052) loss 1.0369 (0.8286) grad_norm 7.6798 (8.3636/1.7082) mem 68106MB [2022-12-20 00:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][150/1519] eta 0:23:00 lr 0.000019 time 0.9302 (1.0085) model_time 0.9300 (1.0054) loss 0.6637 (0.8338) grad_norm 7.9294 (8.3368/1.6877) mem 68106MB [2022-12-20 00:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][160/1519] eta 0:22:49 lr 0.000019 time 0.9387 (1.0080) model_time 0.9385 (1.0051) loss 0.7264 (0.8346) grad_norm 7.5212 (8.3794/1.6862) mem 68106MB [2022-12-20 00:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][170/1519] eta 0:22:39 lr 0.000019 time 0.9399 (1.0079) model_time 0.9398 (1.0052) loss 1.0616 (0.8342) grad_norm 6.9183 (8.4322/1.8360) mem 68106MB [2022-12-20 00:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][180/1519] eta 0:22:28 lr 0.000019 time 0.9298 (1.0074) model_time 0.9296 (1.0048) loss 0.7756 (0.8378) grad_norm 9.9612 (8.4142/1.8064) mem 68106MB [2022-12-20 00:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][190/1519] eta 0:22:18 lr 0.000019 time 0.9266 (1.0071) model_time 0.9265 (1.0046) loss 0.7650 (0.8367) grad_norm 9.1202 (8.4353/1.7911) mem 68106MB [2022-12-20 00:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][200/1519] eta 0:22:07 lr 0.000019 time 0.9291 (1.0067) model_time 0.9289 (1.0043) loss 0.6779 (0.8313) grad_norm 9.2557 (8.4785/1.7825) mem 68106MB [2022-12-20 00:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][210/1519] eta 0:21:57 lr 0.000019 time 0.9828 (1.0068) model_time 0.9827 (1.0044) loss 0.8638 (0.8320) grad_norm 9.2639 (8.4830/1.7460) mem 68106MB [2022-12-20 00:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][220/1519] eta 0:21:47 lr 0.000019 time 0.9313 (1.0064) model_time 0.9312 (1.0042) loss 1.0971 (0.8307) grad_norm 6.8697 (8.4607/1.7641) mem 68106MB [2022-12-20 00:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][230/1519] eta 0:21:37 lr 0.000019 time 0.9380 (1.0064) model_time 0.9379 (1.0042) loss 0.6823 (0.8299) grad_norm 5.4396 (8.5026/1.8447) mem 68106MB [2022-12-20 00:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][240/1519] eta 0:21:27 lr 0.000019 time 0.9259 (1.0065) model_time 0.9255 (1.0044) loss 0.6985 (0.8289) grad_norm 8.3319 (8.4496/1.8286) mem 68106MB [2022-12-20 00:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][250/1519] eta 0:21:17 lr 0.000019 time 0.9354 (1.0064) model_time 0.9353 (1.0044) loss 1.0086 (0.8282) grad_norm 6.1558 (8.4354/1.8199) mem 68106MB [2022-12-20 00:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][260/1519] eta 0:21:06 lr 0.000019 time 0.9274 (1.0061) model_time 0.9272 (1.0042) loss 0.7421 (0.8255) grad_norm 9.6324 (8.3966/1.8130) mem 68106MB [2022-12-20 00:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][270/1519] eta 0:20:56 lr 0.000019 time 0.9696 (1.0061) model_time 0.9694 (1.0042) loss 0.7773 (0.8268) grad_norm 9.7664 (8.4033/1.8183) mem 68106MB [2022-12-20 00:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][280/1519] eta 0:20:46 lr 0.000019 time 0.9246 (1.0058) model_time 0.9245 (1.0040) loss 0.7007 (0.8285) grad_norm 6.3057 (8.3745/1.8240) mem 68106MB [2022-12-20 00:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][290/1519] eta 0:20:35 lr 0.000019 time 0.9265 (1.0055) model_time 0.9264 (1.0037) loss 0.7094 (0.8269) grad_norm 8.9576 (8.3237/1.8253) mem 68106MB [2022-12-20 00:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][300/1519] eta 0:20:25 lr 0.000019 time 0.9360 (1.0056) model_time 0.9358 (1.0038) loss 0.8347 (0.8264) grad_norm 7.0852 (8.3120/1.8103) mem 68106MB [2022-12-20 00:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][310/1519] eta 0:20:15 lr 0.000019 time 0.9269 (1.0054) model_time 0.9267 (1.0037) loss 0.8624 (0.8267) grad_norm 7.0494 (8.2858/1.7906) mem 68106MB [2022-12-20 00:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][320/1519] eta 0:20:05 lr 0.000019 time 0.9249 (1.0054) model_time 0.9248 (1.0038) loss 1.2785 (0.8290) grad_norm 8.1478 (8.2824/1.7725) mem 68106MB [2022-12-20 00:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][330/1519] eta 0:19:55 lr 0.000019 time 0.9500 (1.0056) model_time 0.9497 (1.0040) loss 0.6686 (0.8302) grad_norm 7.1371 (8.2764/1.7612) mem 68106MB [2022-12-20 00:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][340/1519] eta 0:19:45 lr 0.000019 time 0.9120 (1.0055) model_time 0.9118 (1.0039) loss 0.7886 (0.8296) grad_norm 10.1990 (8.2975/1.7525) mem 68106MB [2022-12-20 00:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][350/1519] eta 0:19:35 lr 0.000019 time 0.9288 (1.0053) model_time 0.9287 (1.0037) loss 0.6872 (0.8312) grad_norm 10.2815 (8.3202/1.7411) mem 68106MB [2022-12-20 00:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][360/1519] eta 0:19:24 lr 0.000019 time 0.9314 (1.0051) model_time 0.9312 (1.0036) loss 0.7496 (0.8300) grad_norm 8.5001 (8.3247/1.7316) mem 68106MB [2022-12-20 00:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][370/1519] eta 0:19:14 lr 0.000019 time 0.9333 (1.0051) model_time 0.9332 (1.0036) loss 1.1083 (0.8312) grad_norm 7.2059 (8.3304/1.7188) mem 68106MB [2022-12-20 00:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][380/1519] eta 0:19:04 lr 0.000019 time 0.9277 (1.0050) model_time 0.9275 (1.0035) loss 0.8726 (0.8319) grad_norm 6.6824 (8.3008/1.7166) mem 68106MB [2022-12-20 00:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][390/1519] eta 0:18:54 lr 0.000019 time 0.9353 (1.0050) model_time 0.9351 (1.0035) loss 0.9092 (0.8331) grad_norm 9.0037 (8.3017/1.6971) mem 68106MB [2022-12-20 00:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][400/1519] eta 0:18:44 lr 0.000019 time 0.9361 (1.0050) model_time 0.9360 (1.0036) loss 0.6919 (0.8338) grad_norm 7.0626 (8.2885/1.6891) mem 68106MB [2022-12-20 00:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][410/1519] eta 0:18:35 lr 0.000019 time 0.9048 (1.0055) model_time 0.9047 (1.0041) loss 1.0281 (0.8355) grad_norm 8.6540 (8.2886/1.6699) mem 68106MB [2022-12-20 00:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][420/1519] eta 0:18:24 lr 0.000019 time 0.9390 (1.0053) model_time 0.9389 (1.0040) loss 0.7344 (0.8342) grad_norm 9.2465 (8.3001/1.6694) mem 68106MB [2022-12-20 00:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][430/1519] eta 0:18:14 lr 0.000019 time 0.9268 (1.0052) model_time 0.9266 (1.0039) loss 0.7797 (0.8338) grad_norm 8.9111 (8.2897/1.6761) mem 68106MB [2022-12-20 00:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][440/1519] eta 0:18:04 lr 0.000019 time 0.9236 (1.0050) model_time 0.9230 (1.0037) loss 0.9623 (0.8353) grad_norm 7.8228 (8.2807/1.6678) mem 68106MB [2022-12-20 00:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][450/1519] eta 0:17:54 lr 0.000019 time 0.9294 (1.0051) model_time 0.9293 (1.0038) loss 0.6724 (0.8361) grad_norm 10.8096 (8.2902/1.6616) mem 68106MB [2022-12-20 00:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][460/1519] eta 0:17:44 lr 0.000019 time 0.9374 (1.0051) model_time 0.9372 (1.0038) loss 0.8992 (0.8355) grad_norm 8.8901 (8.2687/1.6667) mem 68106MB [2022-12-20 00:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][470/1519] eta 0:17:34 lr 0.000019 time 0.9340 (1.0049) model_time 0.9339 (1.0037) loss 0.8613 (0.8353) grad_norm 6.0092 (8.2447/1.6618) mem 68106MB [2022-12-20 00:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][480/1519] eta 0:17:24 lr 0.000019 time 0.9142 (1.0054) model_time 0.9140 (1.0041) loss 0.8631 (0.8352) grad_norm 6.4017 (8.2445/1.6562) mem 68106MB [2022-12-20 00:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][490/1519] eta 0:17:14 lr 0.000019 time 0.9321 (1.0052) model_time 0.9319 (1.0040) loss 1.1362 (0.8365) grad_norm 9.3778 (8.2471/1.6466) mem 68106MB [2022-12-20 00:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][500/1519] eta 0:17:04 lr 0.000019 time 0.9301 (1.0052) model_time 0.9299 (1.0040) loss 0.8777 (0.8370) grad_norm 7.4409 (8.2370/1.6351) mem 68106MB [2022-12-20 00:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][510/1519] eta 0:16:54 lr 0.000019 time 0.9170 (1.0051) model_time 0.9168 (1.0039) loss 0.7003 (0.8380) grad_norm 7.2912 (8.2314/1.6202) mem 68106MB [2022-12-20 00:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][520/1519] eta 0:16:44 lr 0.000019 time 0.9290 (1.0050) model_time 0.9288 (1.0039) loss 0.8233 (0.8391) grad_norm 8.0043 (8.2327/1.6130) mem 68106MB [2022-12-20 00:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][530/1519] eta 0:16:33 lr 0.000019 time 0.9312 (1.0050) model_time 0.9309 (1.0039) loss 0.9382 (0.8404) grad_norm 8.3040 (8.2527/1.6385) mem 68106MB [2022-12-20 00:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][540/1519] eta 0:16:23 lr 0.000019 time 0.9320 (1.0049) model_time 0.9318 (1.0038) loss 0.8064 (0.8406) grad_norm 5.8596 (8.2263/1.6420) mem 68106MB [2022-12-20 00:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][550/1519] eta 0:16:13 lr 0.000019 time 0.9242 (1.0051) model_time 0.9240 (1.0040) loss 0.9207 (0.8409) grad_norm 8.8433 (8.2242/1.6365) mem 68106MB [2022-12-20 00:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][560/1519] eta 0:16:03 lr 0.000019 time 0.9354 (1.0051) model_time 0.9352 (1.0040) loss 1.0594 (0.8412) grad_norm 9.0025 (8.2363/1.6290) mem 68106MB [2022-12-20 00:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][570/1519] eta 0:15:53 lr 0.000019 time 0.9347 (1.0051) model_time 0.9345 (1.0040) loss 0.7704 (0.8424) grad_norm 11.2922 (8.2609/1.6509) mem 68106MB [2022-12-20 00:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][580/1519] eta 0:15:43 lr 0.000019 time 0.9346 (1.0050) model_time 0.9344 (1.0039) loss 0.7094 (0.8414) grad_norm 9.5021 (8.2748/1.6504) mem 68106MB [2022-12-20 00:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][590/1519] eta 0:15:33 lr 0.000019 time 0.9316 (1.0050) model_time 0.9315 (1.0039) loss 0.7648 (0.8414) grad_norm 6.7604 (8.3013/1.6694) mem 68106MB [2022-12-20 00:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][600/1519] eta 0:15:23 lr 0.000019 time 0.9324 (1.0049) model_time 0.9322 (1.0038) loss 0.6868 (0.8414) grad_norm 10.6305 (8.3329/1.6895) mem 68106MB [2022-12-20 00:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][610/1519] eta 0:15:13 lr 0.000019 time 0.9390 (1.0053) model_time 0.9389 (1.0042) loss 1.0251 (0.8424) grad_norm 6.7067 (8.3449/1.6995) mem 68106MB [2022-12-20 00:39:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][620/1519] eta 0:15:03 lr 0.000019 time 0.9294 (1.0051) model_time 0.9293 (1.0041) loss 1.0089 (0.8428) grad_norm 9.3894 (8.3622/1.7214) mem 68106MB [2022-12-20 00:40:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][630/1519] eta 0:14:53 lr 0.000019 time 0.9323 (1.0052) model_time 0.9322 (1.0042) loss 0.7673 (0.8422) grad_norm 8.9853 (8.3891/1.7079) mem 68106MB [2022-12-20 00:40:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][640/1519] eta 0:14:43 lr 0.000019 time 1.0272 (1.0052) model_time 1.0271 (1.0042) loss 0.7943 (0.8416) grad_norm 9.1893 (8.3812/1.7044) mem 68106MB [2022-12-20 00:40:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][650/1519] eta 0:14:33 lr 0.000019 time 0.9236 (1.0051) model_time 0.9235 (1.0041) loss 0.6951 (0.8417) grad_norm 5.8823 (8.3701/1.7082) mem 68106MB [2022-12-20 00:40:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][660/1519] eta 0:14:23 lr 0.000019 time 0.9344 (1.0051) model_time 0.9343 (1.0041) loss 0.8437 (0.8423) grad_norm 7.8009 (8.3499/1.7047) mem 68106MB [2022-12-20 00:40:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][670/1519] eta 0:14:13 lr 0.000019 time 0.9315 (1.0051) model_time 0.9314 (1.0041) loss 1.0093 (0.8421) grad_norm 8.6715 (8.3652/1.7290) mem 68106MB [2022-12-20 00:40:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][680/1519] eta 0:14:03 lr 0.000019 time 0.9301 (1.0050) model_time 0.9300 (1.0040) loss 0.7430 (0.8418) grad_norm 7.7638 (8.3847/1.7327) mem 68106MB [2022-12-20 00:41:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][690/1519] eta 0:13:53 lr 0.000019 time 0.9346 (1.0049) model_time 0.9345 (1.0039) loss 0.8278 (0.8422) grad_norm 7.0314 (8.3941/1.7346) mem 68106MB [2022-12-20 00:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][700/1519] eta 0:13:43 lr 0.000019 time 0.9404 (1.0050) model_time 0.9403 (1.0040) loss 0.7879 (0.8424) grad_norm 6.6273 (8.4114/1.7300) mem 68106MB [2022-12-20 00:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][710/1519] eta 0:13:33 lr 0.000019 time 0.9962 (1.0051) model_time 0.9961 (1.0041) loss 0.7464 (0.8432) grad_norm 8.6105 (8.4012/1.7113) mem 68106MB [2022-12-20 00:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][720/1519] eta 0:13:23 lr 0.000019 time 1.1054 (1.0055) model_time 1.1052 (1.0046) loss 0.9843 (0.8460) grad_norm 11.5764 (8.4107/1.7186) mem 68106MB [2022-12-20 00:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][730/1519] eta 0:13:13 lr 0.000019 time 0.9296 (1.0056) model_time 0.9294 (1.0046) loss 0.6924 (0.8456) grad_norm 9.8919 (8.4052/1.6946) mem 68106MB [2022-12-20 00:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][740/1519] eta 0:13:03 lr 0.000019 time 0.9401 (1.0055) model_time 0.9399 (1.0045) loss 0.6953 (0.8441) grad_norm 7.0599 (8.4071/1.7016) mem 68106MB [2022-12-20 00:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][750/1519] eta 0:12:53 lr 0.000019 time 0.9393 (1.0054) model_time 0.9391 (1.0045) loss 0.9322 (0.8448) grad_norm 8.0799 (8.4297/1.7095) mem 68106MB [2022-12-20 00:42:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][760/1519] eta 0:12:43 lr 0.000019 time 0.9405 (1.0053) model_time 0.9404 (1.0044) loss 0.7973 (0.8444) grad_norm 6.9933 (8.4402/1.7319) mem 68106MB [2022-12-20 00:42:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][770/1519] eta 0:12:32 lr 0.000019 time 0.9247 (1.0053) model_time 0.9246 (1.0044) loss 0.7321 (0.8444) grad_norm 12.0704 (8.4216/1.6970) mem 68106MB [2022-12-20 00:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][780/1519] eta 0:12:22 lr 0.000019 time 0.9388 (1.0053) model_time 0.9385 (1.0044) loss 1.0491 (0.8451) grad_norm 6.8780 (8.4128/1.7026) mem 68106MB [2022-12-20 00:42:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][790/1519] eta 0:12:12 lr 0.000019 time 0.9302 (1.0053) model_time 0.9301 (1.0044) loss 0.7022 (0.8442) grad_norm 8.3249 (8.3909/1.7002) mem 68106MB [2022-12-20 00:42:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][800/1519] eta 0:12:02 lr 0.000019 time 0.9301 (1.0053) model_time 0.9299 (1.0044) loss 0.8761 (0.8438) grad_norm 6.4575 (8.3868/1.7047) mem 68106MB [2022-12-20 00:43:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][810/1519] eta 0:11:52 lr 0.000019 time 0.9489 (1.0052) model_time 0.9487 (1.0044) loss 1.0656 (0.8448) grad_norm 6.5104 (8.3692/1.7116) mem 68106MB [2022-12-20 00:43:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][820/1519] eta 0:11:42 lr 0.000019 time 1.0469 (1.0053) model_time 1.0468 (1.0044) loss 0.9917 (0.8460) grad_norm 9.3055 (8.3675/1.7022) mem 68106MB [2022-12-20 00:43:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][830/1519] eta 0:11:32 lr 0.000019 time 0.9417 (1.0053) model_time 0.9415 (1.0044) loss 0.8331 (0.8460) grad_norm 12.2400 (8.3650/1.6871) mem 68106MB [2022-12-20 00:43:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][840/1519] eta 0:11:22 lr 0.000019 time 0.9349 (1.0052) model_time 0.9348 (1.0043) loss 0.6920 (0.8465) grad_norm 9.3923 (8.3884/1.6911) mem 68106MB [2022-12-20 00:43:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][850/1519] eta 0:11:12 lr 0.000019 time 0.9372 (1.0051) model_time 0.9370 (1.0043) loss 0.7173 (0.8454) grad_norm 6.6790 (8.3736/1.6875) mem 68106MB [2022-12-20 00:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][860/1519] eta 0:11:02 lr 0.000019 time 0.9390 (1.0052) model_time 0.9389 (1.0044) loss 0.9161 (0.8462) grad_norm 11.8726 (8.4339/1.7209) mem 68106MB [2022-12-20 00:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][870/1519] eta 0:10:52 lr 0.000019 time 0.9262 (1.0053) model_time 0.9260 (1.0044) loss 0.7773 (0.8457) grad_norm 8.1007 (8.4180/1.7085) mem 68106MB [2022-12-20 00:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][880/1519] eta 0:10:42 lr 0.000019 time 0.9337 (1.0053) model_time 0.9336 (1.0045) loss 0.8613 (0.8461) grad_norm 8.3061 (8.4293/1.7415) mem 68106MB [2022-12-20 00:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][890/1519] eta 0:10:32 lr 0.000019 time 0.9329 (1.0053) model_time 0.9328 (1.0044) loss 0.9635 (0.8462) grad_norm 6.7250 (8.4512/1.7401) mem 68106MB [2022-12-20 00:44:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][900/1519] eta 0:10:22 lr 0.000019 time 0.9732 (1.0052) model_time 0.9730 (1.0044) loss 0.9914 (0.8465) grad_norm 14.3853 (8.4725/1.7775) mem 68106MB [2022-12-20 00:44:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][910/1519] eta 0:10:12 lr 0.000019 time 0.9285 (1.0051) model_time 0.9284 (1.0043) loss 0.7712 (0.8462) grad_norm 7.0242 (8.4847/1.7778) mem 68106MB [2022-12-20 00:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][920/1519] eta 0:10:02 lr 0.000019 time 0.9314 (1.0051) model_time 0.9312 (1.0043) loss 0.9405 (0.8467) grad_norm 8.5105 (8.5025/1.7800) mem 68106MB [2022-12-20 00:45:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][930/1519] eta 0:09:51 lr 0.000019 time 0.9376 (1.0050) model_time 0.9375 (1.0042) loss 0.7038 (0.8465) grad_norm 9.2000 (8.5146/1.7752) mem 68106MB [2022-12-20 00:45:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][940/1519] eta 0:09:41 lr 0.000019 time 0.9973 (1.0050) model_time 0.9972 (1.0042) loss 0.8373 (0.8467) grad_norm 8.0370 (8.4943/1.7704) mem 68106MB [2022-12-20 00:45:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][950/1519] eta 0:09:31 lr 0.000019 time 0.9323 (1.0051) model_time 0.9322 (1.0043) loss 0.6888 (0.8468) grad_norm 8.7454 (8.4956/1.7809) mem 68106MB [2022-12-20 00:45:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][960/1519] eta 0:09:21 lr 0.000019 time 0.9287 (1.0050) model_time 0.9286 (1.0042) loss 0.6839 (0.8472) grad_norm 9.8606 (8.5102/1.7846) mem 68106MB [2022-12-20 00:45:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][970/1519] eta 0:09:11 lr 0.000019 time 0.9321 (1.0051) model_time 0.9320 (1.0043) loss 0.6942 (0.8481) grad_norm 6.1342 (8.4996/1.7878) mem 68106MB [2022-12-20 00:45:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][980/1519] eta 0:09:01 lr 0.000019 time 0.9343 (1.0050) model_time 0.9341 (1.0042) loss 0.7366 (0.8479) grad_norm 8.2762 (8.5190/1.7794) mem 68106MB [2022-12-20 00:46:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][990/1519] eta 0:08:51 lr 0.000019 time 0.9299 (1.0049) model_time 0.9297 (1.0042) loss 1.0841 (0.8484) grad_norm 5.8030 (8.4931/1.7957) mem 68106MB [2022-12-20 00:46:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1000/1519] eta 0:08:41 lr 0.000019 time 0.9271 (1.0049) model_time 0.9269 (1.0041) loss 0.7416 (0.8489) grad_norm 8.4980 (8.5073/1.7923) mem 68106MB [2022-12-20 00:46:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1010/1519] eta 0:08:31 lr 0.000019 time 0.9319 (1.0049) model_time 0.9318 (1.0041) loss 0.7867 (0.8481) grad_norm 13.8874 (8.5446/1.8324) mem 68106MB [2022-12-20 00:46:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1020/1519] eta 0:08:21 lr 0.000019 time 0.9280 (1.0049) model_time 0.9279 (1.0041) loss 0.7079 (0.8482) grad_norm 7.5517 (8.5251/1.8254) mem 68106MB [2022-12-20 00:46:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1030/1519] eta 0:08:11 lr 0.000019 time 0.9311 (1.0050) model_time 0.9309 (1.0042) loss 0.7347 (0.8489) grad_norm 11.6006 (8.5356/1.8189) mem 68106MB [2022-12-20 00:46:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1040/1519] eta 0:08:01 lr 0.000019 time 0.9263 (1.0051) model_time 0.9261 (1.0043) loss 0.7460 (0.8490) grad_norm 7.4878 (8.5606/1.8563) mem 68106MB [2022-12-20 00:47:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1050/1519] eta 0:07:51 lr 0.000019 time 0.9328 (1.0050) model_time 0.9327 (1.0042) loss 0.9362 (0.8504) grad_norm 7.5369 (8.5483/1.8537) mem 68106MB [2022-12-20 00:47:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1060/1519] eta 0:07:41 lr 0.000019 time 0.9347 (1.0049) model_time 0.9342 (1.0042) loss 0.9304 (0.8504) grad_norm 6.6152 (8.5616/1.8434) mem 68106MB [2022-12-20 00:47:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1070/1519] eta 0:07:31 lr 0.000019 time 0.9383 (1.0049) model_time 0.9382 (1.0041) loss 0.9345 (0.8506) grad_norm 9.8429 (8.5897/1.8477) mem 68106MB [2022-12-20 00:47:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1080/1519] eta 0:07:21 lr 0.000019 time 0.9294 (1.0049) model_time 0.9293 (1.0041) loss 0.9313 (0.8503) grad_norm 10.6503 (8.6051/1.8440) mem 68106MB [2022-12-20 00:47:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1090/1519] eta 0:07:11 lr 0.000019 time 0.9312 (1.0048) model_time 0.9306 (1.0041) loss 0.7687 (0.8500) grad_norm 6.7880 (8.6110/1.8493) mem 68106MB [2022-12-20 00:47:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1100/1519] eta 0:07:01 lr 0.000019 time 0.9339 (1.0048) model_time 0.9338 (1.0040) loss 0.6804 (0.8499) grad_norm 6.7751 (8.6071/1.8529) mem 68106MB [2022-12-20 00:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1110/1519] eta 0:06:51 lr 0.000019 time 0.9296 (1.0050) model_time 0.9295 (1.0043) loss 0.7866 (0.8498) grad_norm 7.7656 (8.6187/1.8568) mem 68106MB [2022-12-20 00:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1120/1519] eta 0:06:41 lr 0.000019 time 0.9956 (1.0050) model_time 0.9955 (1.0043) loss 0.8175 (0.8500) grad_norm 6.9638 (8.6162/1.8566) mem 68106MB [2022-12-20 00:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1130/1519] eta 0:06:30 lr 0.000019 time 0.9319 (1.0050) model_time 0.9317 (1.0042) loss 0.8644 (0.8496) grad_norm 11.6234 (8.6135/1.8540) mem 68106MB [2022-12-20 00:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1140/1519] eta 0:06:20 lr 0.000019 time 0.9323 (1.0050) model_time 0.9322 (1.0043) loss 0.7414 (0.8493) grad_norm 8.7845 (8.6449/1.8401) mem 68106MB [2022-12-20 00:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1150/1519] eta 0:06:10 lr 0.000019 time 0.9344 (1.0052) model_time 0.9342 (1.0045) loss 0.7042 (0.8493) grad_norm 8.5663 (8.6484/1.8350) mem 68106MB [2022-12-20 00:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1160/1519] eta 0:06:00 lr 0.000019 time 0.9312 (1.0051) model_time 0.9311 (1.0044) loss 0.8259 (0.8490) grad_norm 9.8464 (8.6447/1.8377) mem 68106MB [2022-12-20 00:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1170/1519] eta 0:05:50 lr 0.000019 time 0.9312 (1.0052) model_time 0.9310 (1.0044) loss 0.7545 (0.8489) grad_norm 9.1025 (8.6159/1.8156) mem 68106MB [2022-12-20 00:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1180/1519] eta 0:05:40 lr 0.000019 time 0.9361 (1.0052) model_time 0.9359 (1.0045) loss 1.3535 (0.8493) grad_norm 6.7504 (8.5944/1.8125) mem 68106MB [2022-12-20 00:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1190/1519] eta 0:05:30 lr 0.000019 time 0.9771 (1.0053) model_time 0.9769 (1.0045) loss 0.7388 (0.8493) grad_norm 7.5103 (8.5597/1.7945) mem 68106MB [2022-12-20 00:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1200/1519] eta 0:05:20 lr 0.000019 time 0.9311 (1.0052) model_time 0.9310 (1.0045) loss 0.7658 (0.8489) grad_norm 8.0343 (8.5355/1.7695) mem 68106MB [2022-12-20 00:49:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1210/1519] eta 0:05:10 lr 0.000019 time 0.9325 (1.0054) model_time 0.9324 (1.0047) loss 0.7087 (0.8484) grad_norm 11.2484 (8.5303/1.7660) mem 68106MB [2022-12-20 00:50:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1220/1519] eta 0:05:00 lr 0.000019 time 0.9338 (1.0056) model_time 0.9336 (1.0049) loss 1.0191 (0.8485) grad_norm 6.7367 (8.4990/1.7435) mem 68106MB [2022-12-20 00:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1230/1519] eta 0:04:50 lr 0.000019 time 0.9311 (1.0056) model_time 0.9309 (1.0049) loss 0.6897 (0.8485) grad_norm 5.8757 (8.5010/1.7560) mem 68106MB [2022-12-20 00:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1240/1519] eta 0:04:40 lr 0.000019 time 0.9298 (1.0055) model_time 0.9296 (1.0048) loss 1.0796 (0.8489) grad_norm 8.1576 (8.4933/1.7555) mem 68106MB [2022-12-20 00:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1250/1519] eta 0:04:30 lr 0.000019 time 0.9303 (1.0054) model_time 0.9301 (1.0047) loss 0.6818 (0.8485) grad_norm 7.3576 (8.4807/1.7593) mem 68106MB [2022-12-20 00:50:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1260/1519] eta 0:04:20 lr 0.000019 time 0.9123 (1.0055) model_time 0.9122 (1.0048) loss 0.7138 (0.8487) grad_norm 7.0526 (8.4966/1.7599) mem 68106MB [2022-12-20 00:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1270/1519] eta 0:04:10 lr 0.000019 time 0.9367 (1.0055) model_time 0.9364 (1.0048) loss 0.8296 (0.8489) grad_norm 7.9041 (8.4623/1.7306) mem 68106MB [2022-12-20 00:51:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1280/1519] eta 0:04:00 lr 0.000019 time 0.9392 (1.0054) model_time 0.9391 (1.0047) loss 1.3202 (0.8499) grad_norm 9.1558 (8.4574/1.7327) mem 68106MB [2022-12-20 00:51:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1290/1519] eta 0:03:50 lr 0.000019 time 0.9511 (1.0055) model_time 0.9510 (1.0048) loss 0.6775 (0.8496) grad_norm 6.6705 (8.4427/1.7492) mem 68106MB [2022-12-20 00:51:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1300/1519] eta 0:03:40 lr 0.000019 time 0.9828 (1.0055) model_time 0.9826 (1.0048) loss 0.7746 (0.8495) grad_norm 9.0236 (8.4472/1.7492) mem 68106MB [2022-12-20 00:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1310/1519] eta 0:03:30 lr 0.000019 time 0.9337 (1.0054) model_time 0.9336 (1.0047) loss 0.9899 (0.8499) grad_norm 7.2146 (8.4267/1.7410) mem 68106MB [2022-12-20 00:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1320/1519] eta 0:03:20 lr 0.000019 time 1.1940 (1.0056) model_time 1.1938 (1.0049) loss 0.8418 (0.8500) grad_norm 5.4757 (8.4183/1.7353) mem 68106MB [2022-12-20 00:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1330/1519] eta 0:03:10 lr 0.000019 time 0.9294 (1.0056) model_time 0.9292 (1.0049) loss 0.8163 (0.8501) grad_norm 8.5097 (8.4040/1.7330) mem 68106MB [2022-12-20 00:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1340/1519] eta 0:02:59 lr 0.000019 time 0.9251 (1.0055) model_time 0.9249 (1.0049) loss 0.7343 (0.8499) grad_norm 10.4035 (8.4267/1.7734) mem 68106MB [2022-12-20 00:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1350/1519] eta 0:02:49 lr 0.000019 time 0.9405 (1.0056) model_time 0.9404 (1.0049) loss 0.6708 (0.8501) grad_norm 8.0870 (8.4032/1.7620) mem 68106MB [2022-12-20 00:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1360/1519] eta 0:02:39 lr 0.000019 time 0.9207 (1.0056) model_time 0.9205 (1.0049) loss 0.9221 (0.8499) grad_norm 8.4941 (8.3678/1.7328) mem 68106MB [2022-12-20 00:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1370/1519] eta 0:02:29 lr 0.000019 time 0.9331 (1.0055) model_time 0.9329 (1.0049) loss 0.9051 (0.8499) grad_norm 9.5314 (8.4057/1.7509) mem 68106MB [2022-12-20 00:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1380/1519] eta 0:02:19 lr 0.000019 time 0.9324 (1.0055) model_time 0.9323 (1.0049) loss 0.7031 (0.8494) grad_norm 10.2225 (8.4131/1.7503) mem 68106MB [2022-12-20 00:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1390/1519] eta 0:02:09 lr 0.000019 time 0.9201 (1.0055) model_time 0.9199 (1.0049) loss 1.1760 (0.8497) grad_norm 7.0213 (8.4435/1.7553) mem 68106MB [2022-12-20 00:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1400/1519] eta 0:01:59 lr 0.000019 time 0.9223 (1.0055) model_time 0.9222 (1.0048) loss 1.0408 (0.8498) grad_norm 8.3166 (8.4374/1.7496) mem 68106MB [2022-12-20 00:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1410/1519] eta 0:01:49 lr 0.000019 time 0.9296 (1.0054) model_time 0.9294 (1.0048) loss 0.7436 (0.8494) grad_norm 7.4611 (8.4526/1.7489) mem 68106MB [2022-12-20 00:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1420/1519] eta 0:01:39 lr 0.000019 time 0.9267 (1.0056) model_time 0.9265 (1.0049) loss 0.9157 (0.8495) grad_norm 9.1985 (8.4582/1.7450) mem 68106MB [2022-12-20 00:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1430/1519] eta 0:01:29 lr 0.000019 time 0.9342 (1.0056) model_time 0.9341 (1.0049) loss 1.0680 (0.8499) grad_norm 7.6156 (8.4318/1.7178) mem 68106MB [2022-12-20 00:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1440/1519] eta 0:01:19 lr 0.000019 time 0.9372 (1.0056) model_time 0.9371 (1.0049) loss 0.6912 (0.8497) grad_norm 7.8947 (8.4283/1.7154) mem 68106MB [2022-12-20 00:53:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1450/1519] eta 0:01:09 lr 0.000019 time 0.9287 (1.0056) model_time 0.9285 (1.0050) loss 0.8624 (0.8502) grad_norm 6.4729 (8.4475/1.7161) mem 68106MB [2022-12-20 00:54:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1460/1519] eta 0:00:59 lr 0.000019 time 0.9318 (1.0056) model_time 0.9317 (1.0050) loss 0.8650 (0.8501) grad_norm 7.5138 (8.4066/1.6873) mem 68106MB [2022-12-20 00:54:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1470/1519] eta 0:00:49 lr 0.000019 time 0.9273 (1.0056) model_time 0.9266 (1.0049) loss 0.9193 (0.8500) grad_norm 9.8394 (8.4229/1.6851) mem 68106MB [2022-12-20 00:54:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1480/1519] eta 0:00:39 lr 0.000019 time 1.0335 (1.0056) model_time 1.0334 (1.0050) loss 1.1838 (0.8498) grad_norm 7.0046 (8.4340/1.6433) mem 68106MB [2022-12-20 00:54:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1490/1519] eta 0:00:29 lr 0.000019 time 0.9226 (1.0056) model_time 0.9223 (1.0049) loss 0.8394 (0.8499) grad_norm 13.8319 (8.4468/1.6600) mem 68106MB [2022-12-20 00:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1500/1519] eta 0:00:19 lr 0.000019 time 0.9320 (1.0056) model_time 0.9318 (1.0049) loss 0.9384 (0.8498) grad_norm 10.3862 (8.4216/1.6243) mem 68106MB [2022-12-20 00:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [46/100][1510/1519] eta 0:00:09 lr 0.000019 time 0.9171 (1.0057) model_time 0.9170 (1.0051) loss 0.7146 (0.8500) grad_norm 8.1513 (8.4053/1.6346) mem 68106MB [2022-12-20 00:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 46 training takes 0:25:27 [2022-12-20 00:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_46.pth saving...... [2022-12-20 00:55:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_46.pth saved !!! [2022-12-20 00:55:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.5132 (0.5132) Acc@1 91.319 (91.319) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 00:55:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.331) Loss 0.5244 (0.4898) Acc@1 92.014 (92.077) Acc@5 98.264 (98.516) Mem 68106MB [2022-12-20 00:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.314) Loss 0.4497 (0.4846) Acc@1 92.014 (92.130) Acc@5 98.958 (98.396) Mem 68106MB [2022-12-20 00:55:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.313) Loss 0.6170 (0.4915) Acc@1 89.236 (91.902) Acc@5 97.917 (98.376) Mem 68106MB [2022-12-20 00:55:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.310) Loss 0.4848 (0.4831) Acc@1 92.014 (92.056) Acc@5 98.264 (98.442) Mem 68106MB [2022-12-20 00:55:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.308) Loss 0.4675 (0.4806) Acc@1 90.972 (92.082) Acc@5 98.958 (98.495) Mem 68106MB [2022-12-20 00:55:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.308 (0.307) Loss 0.4840 (0.4805) Acc@1 90.972 (92.054) Acc@5 98.264 (98.486) Mem 68106MB [2022-12-20 00:55:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.305) Loss 0.5132 (0.4810) Acc@1 92.361 (91.999) Acc@5 98.264 (98.499) Mem 68106MB [2022-12-20 00:55:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.301 (0.304) Loss 0.4161 (0.4791) Acc@1 93.056 (92.040) Acc@5 98.611 (98.530) Mem 68106MB [2022-12-20 00:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:46] * Acc@1 91.978 Acc@5 98.535 [2022-12-20 00:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.0% [2022-12-20 00:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.10% [2022-12-20 00:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][0/1519] eta 0:45:13 lr 0.000019 time 1.7865 (1.7865) model_time 0.9823 (0.9823) loss 0.6728 (0.6728) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 00:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][10/1519] eta 0:27:09 lr 0.000019 time 0.9355 (1.0799) model_time 0.9354 (1.0063) loss 0.8467 (0.7578) grad_norm 11.6615 (7.5970/2.1767) mem 68106MB [2022-12-20 00:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][20/1519] eta 0:26:01 lr 0.000019 time 0.9309 (1.0415) model_time 0.9307 (1.0028) loss 0.7645 (0.7941) grad_norm 7.1135 (8.2652/2.2177) mem 68106MB [2022-12-20 00:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][30/1519] eta 0:25:34 lr 0.000019 time 0.9232 (1.0309) model_time 0.9230 (1.0045) loss 0.7937 (0.8148) grad_norm 7.7423 (8.6579/2.1066) mem 68106MB [2022-12-20 00:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][40/1519] eta 0:25:12 lr 0.000019 time 0.9391 (1.0228) model_time 0.9389 (1.0028) loss 0.8227 (0.8156) grad_norm 7.2936 (8.3790/1.9370) mem 68106MB [2022-12-20 00:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][50/1519] eta 0:24:55 lr 0.000019 time 0.9212 (1.0178) model_time 0.9211 (1.0017) loss 1.0200 (0.8160) grad_norm 7.4227 (8.2712/1.8275) mem 68106MB [2022-12-20 00:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][60/1519] eta 0:24:40 lr 0.000019 time 0.9255 (1.0147) model_time 0.9253 (1.0011) loss 0.6953 (0.8055) grad_norm 6.4781 (8.6931/2.5675) mem 68106MB [2022-12-20 00:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][70/1519] eta 0:24:29 lr 0.000019 time 0.9351 (1.0139) model_time 0.9349 (1.0022) loss 0.7629 (0.8068) grad_norm 7.4500 (8.6405/2.3928) mem 68106MB [2022-12-20 00:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][80/1519] eta 0:24:18 lr 0.000019 time 0.9446 (1.0138) model_time 0.9444 (1.0035) loss 0.8845 (0.8188) grad_norm 7.1685 (8.5559/2.3212) mem 68106MB [2022-12-20 00:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][90/1519] eta 0:24:07 lr 0.000019 time 0.9215 (1.0128) model_time 0.9214 (1.0036) loss 0.8187 (0.8179) grad_norm 9.7472 (8.5855/2.3746) mem 68106MB [2022-12-20 00:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][100/1519] eta 0:23:56 lr 0.000019 time 0.9272 (1.0120) model_time 0.9270 (1.0037) loss 0.9162 (0.8322) grad_norm 5.7130 (8.5776/2.3955) mem 68106MB [2022-12-20 00:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][110/1519] eta 0:23:45 lr 0.000019 time 0.9348 (1.0114) model_time 0.9346 (1.0037) loss 0.7517 (0.8346) grad_norm 6.9133 (8.6125/2.3281) mem 68106MB [2022-12-20 00:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][120/1519] eta 0:23:34 lr 0.000019 time 0.9209 (1.0111) model_time 0.9207 (1.0041) loss 1.0420 (0.8358) grad_norm 7.3729 (8.5648/2.2569) mem 68106MB [2022-12-20 00:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][130/1519] eta 0:23:23 lr 0.000019 time 0.9768 (1.0105) model_time 0.9767 (1.0040) loss 0.6920 (0.8354) grad_norm 8.2035 (8.5795/2.1831) mem 68106MB [2022-12-20 00:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][140/1519] eta 0:23:13 lr 0.000019 time 0.9302 (1.0105) model_time 0.9299 (1.0044) loss 0.7521 (0.8362) grad_norm 9.7709 (8.5702/2.1320) mem 68106MB [2022-12-20 00:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][150/1519] eta 0:23:02 lr 0.000019 time 0.9290 (1.0101) model_time 0.9289 (1.0044) loss 0.9713 (0.8346) grad_norm 8.5206 (8.5303/2.0765) mem 68106MB [2022-12-20 00:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][160/1519] eta 0:22:51 lr 0.000019 time 0.9257 (1.0094) model_time 0.9255 (1.0040) loss 0.6717 (0.8376) grad_norm 6.4484 (8.4291/2.0676) mem 68106MB [2022-12-20 00:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][170/1519] eta 0:22:41 lr 0.000019 time 0.9223 (1.0091) model_time 0.9222 (1.0040) loss 0.8119 (0.8373) grad_norm 9.6523 (8.4247/2.0140) mem 68106MB [2022-12-20 00:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][180/1519] eta 0:22:30 lr 0.000019 time 0.9296 (1.0085) model_time 0.9294 (1.0037) loss 0.8068 (0.8406) grad_norm 8.5200 (8.4681/2.0296) mem 68106MB [2022-12-20 00:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][190/1519] eta 0:22:19 lr 0.000019 time 0.9170 (1.0081) model_time 0.9168 (1.0035) loss 0.7900 (0.8399) grad_norm 7.3795 (8.4742/1.9863) mem 68106MB [2022-12-20 00:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][200/1519] eta 0:22:08 lr 0.000019 time 0.9201 (1.0076) model_time 0.9200 (1.0032) loss 1.2541 (0.8378) grad_norm 8.4982 (8.5200/1.9609) mem 68106MB [2022-12-20 00:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][210/1519] eta 0:21:58 lr 0.000019 time 0.9340 (1.0072) model_time 0.9339 (1.0030) loss 0.7366 (0.8420) grad_norm 7.9163 (8.4860/1.9363) mem 68106MB [2022-12-20 00:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][220/1519] eta 0:21:48 lr 0.000019 time 0.9306 (1.0071) model_time 0.9305 (1.0031) loss 0.7902 (0.8390) grad_norm 12.3345 (8.5629/1.9950) mem 68106MB [2022-12-20 00:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][230/1519] eta 0:21:37 lr 0.000019 time 0.9227 (1.0070) model_time 0.9225 (1.0031) loss 0.7974 (0.8391) grad_norm 5.4378 (8.5717/2.0268) mem 68106MB [2022-12-20 00:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][240/1519] eta 0:21:27 lr 0.000019 time 0.9171 (1.0065) model_time 0.9170 (1.0028) loss 0.9067 (0.8408) grad_norm 6.8960 (8.5280/2.0028) mem 68106MB [2022-12-20 01:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][250/1519] eta 0:21:18 lr 0.000019 time 1.1715 (1.0076) model_time 1.1713 (1.0040) loss 1.0213 (0.8424) grad_norm 9.7406 (8.5722/1.9956) mem 68106MB [2022-12-20 01:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][260/1519] eta 0:21:08 lr 0.000019 time 0.9292 (1.0072) model_time 0.9290 (1.0037) loss 1.0007 (0.8426) grad_norm 6.3479 (8.4988/2.0005) mem 68106MB [2022-12-20 01:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][270/1519] eta 0:20:57 lr 0.000019 time 0.9140 (1.0070) model_time 0.9139 (1.0037) loss 0.7749 (0.8454) grad_norm 8.3154 (8.5284/2.0180) mem 68106MB [2022-12-20 01:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][280/1519] eta 0:20:47 lr 0.000019 time 0.9291 (1.0068) model_time 0.9289 (1.0035) loss 0.9895 (0.8463) grad_norm 7.4877 (8.5151/1.9855) mem 68106MB [2022-12-20 01:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][290/1519] eta 0:20:37 lr 0.000019 time 0.9272 (1.0067) model_time 0.9270 (1.0035) loss 0.6876 (0.8432) grad_norm 9.8804 (8.5327/1.9731) mem 68106MB [2022-12-20 01:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][300/1519] eta 0:20:26 lr 0.000019 time 0.9331 (1.0064) model_time 0.9329 (1.0034) loss 0.9378 (0.8432) grad_norm 9.9933 (8.5880/1.9855) mem 68106MB [2022-12-20 01:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][310/1519] eta 0:20:17 lr 0.000019 time 0.9182 (1.0072) model_time 0.9180 (1.0042) loss 0.9468 (0.8417) grad_norm 11.0886 (8.5887/2.0065) mem 68106MB [2022-12-20 01:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][320/1519] eta 0:20:08 lr 0.000019 time 0.9332 (1.0079) model_time 0.9330 (1.0050) loss 0.8782 (0.8415) grad_norm 9.4190 (8.5721/1.9819) mem 68106MB [2022-12-20 01:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][330/1519] eta 0:19:58 lr 0.000019 time 0.9306 (1.0078) model_time 0.9305 (1.0049) loss 0.8379 (0.8383) grad_norm 6.5528 (8.5516/1.9740) mem 68106MB [2022-12-20 01:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][340/1519] eta 0:19:47 lr 0.000018 time 0.9291 (1.0074) model_time 0.9289 (1.0047) loss 0.9454 (0.8382) grad_norm 7.5900 (8.5423/1.9644) mem 68106MB [2022-12-20 01:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][350/1519] eta 0:19:37 lr 0.000018 time 0.9265 (1.0074) model_time 0.9263 (1.0047) loss 0.6888 (0.8392) grad_norm 8.8226 (8.5453/1.9369) mem 68106MB [2022-12-20 01:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][360/1519] eta 0:19:27 lr 0.000018 time 0.9311 (1.0071) model_time 0.9310 (1.0045) loss 1.1074 (0.8398) grad_norm 7.8081 (8.5466/1.9229) mem 68106MB [2022-12-20 01:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][370/1519] eta 0:19:16 lr 0.000018 time 0.9305 (1.0069) model_time 0.9303 (1.0044) loss 0.8723 (0.8386) grad_norm 7.8321 (8.5608/1.9173) mem 68106MB [2022-12-20 01:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][380/1519] eta 0:19:06 lr 0.000018 time 0.9392 (1.0069) model_time 0.9389 (1.0044) loss 0.6789 (0.8410) grad_norm 11.7136 (8.5675/1.9111) mem 68106MB [2022-12-20 01:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][390/1519] eta 0:18:57 lr 0.000018 time 1.1904 (1.0075) model_time 1.1902 (1.0051) loss 0.6748 (0.8408) grad_norm 10.7325 (8.5894/1.9008) mem 68106MB [2022-12-20 01:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][400/1519] eta 0:18:47 lr 0.000018 time 0.9163 (1.0075) model_time 0.9162 (1.0051) loss 0.7843 (0.8405) grad_norm 9.5787 (8.5907/1.8879) mem 68106MB [2022-12-20 01:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][410/1519] eta 0:18:37 lr 0.000018 time 0.9253 (1.0075) model_time 0.9251 (1.0052) loss 1.1978 (0.8402) grad_norm 8.0506 (8.5903/1.8715) mem 68106MB [2022-12-20 01:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][420/1519] eta 0:18:27 lr 0.000018 time 0.9306 (1.0074) model_time 0.9304 (1.0050) loss 0.7053 (0.8392) grad_norm 7.1399 (8.5691/1.8747) mem 68106MB [2022-12-20 01:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][430/1519] eta 0:18:17 lr 0.000018 time 1.0039 (1.0077) model_time 1.0037 (1.0054) loss 0.7369 (0.8390) grad_norm 11.1857 (8.5618/1.8679) mem 68106MB [2022-12-20 01:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][440/1519] eta 0:18:07 lr 0.000018 time 0.9258 (1.0075) model_time 0.9256 (1.0053) loss 0.8028 (0.8382) grad_norm 9.0881 (8.5542/1.8513) mem 68106MB [2022-12-20 01:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][450/1519] eta 0:17:57 lr 0.000018 time 0.9220 (1.0076) model_time 0.9218 (1.0055) loss 0.9116 (0.8387) grad_norm 7.7957 (8.6243/1.9855) mem 68106MB [2022-12-20 01:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][460/1519] eta 0:17:46 lr 0.000018 time 0.9210 (1.0074) model_time 0.9209 (1.0053) loss 0.6685 (0.8392) grad_norm 7.0253 (8.6293/1.9982) mem 68106MB [2022-12-20 01:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][470/1519] eta 0:17:36 lr 0.000018 time 0.9321 (1.0072) model_time 0.9320 (1.0050) loss 0.7501 (0.8400) grad_norm 8.3461 (8.6181/1.9878) mem 68106MB [2022-12-20 01:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][480/1519] eta 0:17:26 lr 0.000018 time 0.9283 (1.0070) model_time 0.9282 (1.0049) loss 0.7032 (0.8395) grad_norm 7.6263 (8.6038/1.9840) mem 68106MB [2022-12-20 01:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][490/1519] eta 0:17:16 lr 0.000018 time 0.9167 (1.0071) model_time 0.9166 (1.0050) loss 1.0114 (0.8398) grad_norm 7.7077 (8.5847/1.9778) mem 68106MB [2022-12-20 01:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][500/1519] eta 0:17:06 lr 0.000018 time 0.9260 (1.0070) model_time 0.9259 (1.0049) loss 0.6755 (0.8388) grad_norm 7.2006 (8.5626/1.9682) mem 68106MB [2022-12-20 01:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][510/1519] eta 0:16:56 lr 0.000018 time 0.9312 (1.0070) model_time 0.9311 (1.0050) loss 0.6891 (0.8385) grad_norm 7.6238 (8.5699/1.9582) mem 68106MB [2022-12-20 01:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][520/1519] eta 0:16:45 lr 0.000018 time 0.9165 (1.0068) model_time 0.9163 (1.0049) loss 1.1082 (0.8387) grad_norm 7.8002 (8.5434/1.9492) mem 68106MB [2022-12-20 01:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][530/1519] eta 0:16:35 lr 0.000018 time 0.9234 (1.0067) model_time 0.9233 (1.0048) loss 0.7892 (0.8412) grad_norm 9.0787 (8.5285/1.9396) mem 68106MB [2022-12-20 01:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][540/1519] eta 0:16:25 lr 0.000018 time 0.9324 (1.0066) model_time 0.9322 (1.0047) loss 0.7860 (0.8410) grad_norm 8.0416 (8.5133/1.9258) mem 68106MB [2022-12-20 01:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][550/1519] eta 0:16:15 lr 0.000018 time 0.9319 (1.0064) model_time 0.9317 (1.0045) loss 0.8369 (0.8401) grad_norm 5.1342 (8.4878/1.9267) mem 68106MB [2022-12-20 01:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][560/1519] eta 0:16:05 lr 0.000018 time 0.9266 (1.0065) model_time 0.9265 (1.0047) loss 0.9846 (0.8397) grad_norm 7.8697 (8.4843/1.9125) mem 68106MB [2022-12-20 01:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][570/1519] eta 0:15:55 lr 0.000018 time 0.9842 (1.0065) model_time 0.9840 (1.0046) loss 0.9998 (0.8402) grad_norm 8.9718 (8.4917/1.9154) mem 68106MB [2022-12-20 01:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][580/1519] eta 0:15:44 lr 0.000018 time 0.9346 (1.0064) model_time 0.9345 (1.0046) loss 0.8096 (0.8413) grad_norm 15.1114 (8.5045/1.9417) mem 68106MB [2022-12-20 01:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][590/1519] eta 0:15:34 lr 0.000018 time 0.9238 (1.0063) model_time 0.9236 (1.0046) loss 0.9405 (0.8416) grad_norm 6.4762 (8.5059/1.9413) mem 68106MB [2022-12-20 01:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][600/1519] eta 0:15:24 lr 0.000018 time 0.9206 (1.0063) model_time 0.9204 (1.0046) loss 0.9630 (0.8408) grad_norm 6.7396 (8.4912/1.9303) mem 68106MB [2022-12-20 01:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][610/1519] eta 0:15:14 lr 0.000018 time 0.9263 (1.0062) model_time 0.9262 (1.0045) loss 0.7398 (0.8399) grad_norm 8.1879 (8.4994/1.9180) mem 68106MB [2022-12-20 01:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][620/1519] eta 0:15:04 lr 0.000018 time 0.9241 (1.0061) model_time 0.9240 (1.0044) loss 0.8278 (0.8403) grad_norm 8.2340 (8.4803/1.9113) mem 68106MB [2022-12-20 01:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][630/1519] eta 0:14:54 lr 0.000018 time 0.9072 (1.0061) model_time 0.9071 (1.0044) loss 0.9252 (0.8410) grad_norm 8.2437 (8.4630/1.9090) mem 68106MB [2022-12-20 01:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][640/1519] eta 0:14:44 lr 0.000018 time 0.9220 (1.0059) model_time 0.9218 (1.0043) loss 0.8251 (0.8410) grad_norm 8.4651 (8.5082/1.9203) mem 68106MB [2022-12-20 01:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][650/1519] eta 0:14:34 lr 0.000018 time 0.9257 (1.0059) model_time 0.9256 (1.0043) loss 0.8149 (0.8407) grad_norm 8.3017 (8.5615/1.9715) mem 68106MB [2022-12-20 01:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][660/1519] eta 0:14:23 lr 0.000018 time 0.9330 (1.0058) model_time 0.9328 (1.0042) loss 0.9732 (0.8404) grad_norm 7.8275 (8.5207/1.8888) mem 68106MB [2022-12-20 01:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][670/1519] eta 0:14:13 lr 0.000018 time 0.9271 (1.0057) model_time 0.9270 (1.0041) loss 1.0743 (0.8409) grad_norm 9.9745 (8.5234/1.8896) mem 68106MB [2022-12-20 01:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][680/1519] eta 0:14:03 lr 0.000018 time 0.9341 (1.0055) model_time 0.9338 (1.0040) loss 0.7929 (0.8422) grad_norm 10.6292 (8.5750/1.9332) mem 68106MB [2022-12-20 01:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][690/1519] eta 0:13:53 lr 0.000018 time 1.0196 (1.0056) model_time 1.0194 (1.0040) loss 0.9936 (0.8425) grad_norm 7.8263 (8.5455/1.9119) mem 68106MB [2022-12-20 01:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][700/1519] eta 0:13:43 lr 0.000018 time 0.9198 (1.0056) model_time 0.9196 (1.0041) loss 0.9217 (0.8416) grad_norm 14.1992 (8.5607/1.9288) mem 68106MB [2022-12-20 01:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][710/1519] eta 0:13:33 lr 0.000018 time 0.8949 (1.0056) model_time 0.8948 (1.0041) loss 0.6908 (0.8416) grad_norm 10.0249 (8.5603/1.9232) mem 68106MB [2022-12-20 01:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][720/1519] eta 0:13:23 lr 0.000018 time 0.9009 (1.0055) model_time 0.9008 (1.0040) loss 0.6766 (0.8415) grad_norm 10.2828 (8.5748/1.9321) mem 68106MB [2022-12-20 01:08:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][730/1519] eta 0:13:13 lr 0.000018 time 0.9371 (1.0055) model_time 0.9369 (1.0040) loss 0.9988 (0.8413) grad_norm 8.9190 (8.5771/1.9316) mem 68106MB [2022-12-20 01:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][740/1519] eta 0:13:03 lr 0.000018 time 0.9361 (1.0056) model_time 0.9359 (1.0041) loss 1.0439 (0.8418) grad_norm 8.1816 (8.5945/1.9359) mem 68106MB [2022-12-20 01:08:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][750/1519] eta 0:12:53 lr 0.000018 time 0.9324 (1.0056) model_time 0.9322 (1.0042) loss 0.7055 (0.8416) grad_norm 6.9959 (8.5826/1.9410) mem 68106MB [2022-12-20 01:08:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][760/1519] eta 0:12:43 lr 0.000018 time 0.9386 (1.0056) model_time 0.9384 (1.0041) loss 0.8297 (0.8419) grad_norm 6.9527 (8.6037/1.9302) mem 68106MB [2022-12-20 01:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][770/1519] eta 0:12:33 lr 0.000018 time 0.9328 (1.0057) model_time 0.9326 (1.0043) loss 0.9690 (0.8412) grad_norm 8.0519 (8.6067/1.9377) mem 68106MB [2022-12-20 01:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][780/1519] eta 0:12:23 lr 0.000018 time 0.9353 (1.0057) model_time 0.9351 (1.0042) loss 0.9670 (0.8418) grad_norm 7.8551 (8.6436/2.1425) mem 68106MB [2022-12-20 01:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][790/1519] eta 0:12:13 lr 0.000018 time 0.9305 (1.0056) model_time 0.9303 (1.0042) loss 1.0481 (0.8421) grad_norm 6.6513 (8.6202/2.1494) mem 68106MB [2022-12-20 01:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][800/1519] eta 0:12:03 lr 0.000018 time 0.9329 (1.0057) model_time 0.9327 (1.0043) loss 1.0095 (0.8433) grad_norm 6.9206 (8.6125/2.1544) mem 68106MB [2022-12-20 01:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][810/1519] eta 0:11:52 lr 0.000018 time 0.9298 (1.0056) model_time 0.9297 (1.0042) loss 0.8234 (0.8427) grad_norm 7.7578 (8.6381/2.1539) mem 68106MB [2022-12-20 01:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][820/1519] eta 0:11:42 lr 0.000018 time 0.9265 (1.0057) model_time 0.9263 (1.0043) loss 0.6904 (0.8428) grad_norm 8.6485 (8.6125/2.1228) mem 68106MB [2022-12-20 01:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][830/1519] eta 0:11:32 lr 0.000018 time 0.9305 (1.0056) model_time 0.9303 (1.0042) loss 0.8674 (0.8427) grad_norm 6.6524 (8.5935/2.1082) mem 68106MB [2022-12-20 01:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][840/1519] eta 0:11:22 lr 0.000018 time 0.9308 (1.0055) model_time 0.9306 (1.0042) loss 0.7112 (0.8422) grad_norm 9.2478 (8.6256/2.1201) mem 68106MB [2022-12-20 01:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][850/1519] eta 0:11:12 lr 0.000018 time 0.9266 (1.0054) model_time 0.9264 (1.0041) loss 0.6938 (0.8419) grad_norm 7.9874 (8.6081/2.1168) mem 68106MB [2022-12-20 01:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][860/1519] eta 0:11:02 lr 0.000018 time 0.9305 (1.0053) model_time 0.9303 (1.0040) loss 0.6936 (0.8406) grad_norm 12.0996 (8.6557/2.1105) mem 68106MB [2022-12-20 01:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][870/1519] eta 0:10:52 lr 0.000018 time 0.9338 (1.0052) model_time 0.9336 (1.0039) loss 0.7991 (0.8406) grad_norm 12.9520 (8.6604/2.1051) mem 68106MB [2022-12-20 01:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][880/1519] eta 0:10:42 lr 0.000018 time 0.9284 (1.0051) model_time 0.9283 (1.0038) loss 0.7952 (0.8411) grad_norm 6.3400 (8.6317/2.1260) mem 68106MB [2022-12-20 01:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][890/1519] eta 0:10:32 lr 0.000018 time 0.9321 (1.0051) model_time 0.9320 (1.0038) loss 0.8905 (0.8416) grad_norm 7.9755 (8.6267/2.1258) mem 68106MB [2022-12-20 01:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][900/1519] eta 0:10:22 lr 0.000018 time 0.9117 (1.0051) model_time 0.9116 (1.0039) loss 0.7165 (0.8421) grad_norm 8.5179 (8.5921/2.1068) mem 68106MB [2022-12-20 01:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][910/1519] eta 0:10:12 lr 0.000018 time 0.9400 (1.0052) model_time 0.9398 (1.0039) loss 0.6967 (0.8425) grad_norm 6.9292 (8.5850/2.0864) mem 68106MB [2022-12-20 01:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][920/1519] eta 0:10:02 lr 0.000018 time 0.9321 (1.0051) model_time 0.9319 (1.0038) loss 0.6809 (0.8433) grad_norm 6.5359 (8.6114/2.1305) mem 68106MB [2022-12-20 01:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][930/1519] eta 0:09:51 lr 0.000018 time 0.9276 (1.0050) model_time 0.9274 (1.0038) loss 0.8044 (0.8428) grad_norm 10.0215 (8.6528/2.1422) mem 68106MB [2022-12-20 01:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][940/1519] eta 0:09:41 lr 0.000018 time 0.9317 (1.0049) model_time 0.9316 (1.0037) loss 0.7910 (0.8426) grad_norm 10.3079 (8.6797/2.1403) mem 68106MB [2022-12-20 01:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][950/1519] eta 0:09:31 lr 0.000018 time 0.9410 (1.0049) model_time 0.9409 (1.0036) loss 0.7798 (0.8423) grad_norm 9.3275 (8.6862/2.1497) mem 68106MB [2022-12-20 01:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][960/1519] eta 0:09:21 lr 0.000018 time 0.9323 (1.0049) model_time 0.9321 (1.0036) loss 0.6971 (0.8420) grad_norm 7.5697 (8.6941/2.1556) mem 68106MB [2022-12-20 01:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][970/1519] eta 0:09:11 lr 0.000018 time 0.9318 (1.0049) model_time 0.9316 (1.0037) loss 1.2176 (0.8427) grad_norm 6.6474 (8.6756/2.1629) mem 68106MB [2022-12-20 01:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][980/1519] eta 0:09:01 lr 0.000018 time 0.9325 (1.0048) model_time 0.9324 (1.0036) loss 0.7450 (0.8426) grad_norm 6.8653 (8.6680/2.1716) mem 68106MB [2022-12-20 01:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][990/1519] eta 0:08:51 lr 0.000018 time 0.9296 (1.0048) model_time 0.9294 (1.0036) loss 1.0227 (0.8428) grad_norm 7.6490 (8.6283/2.1789) mem 68106MB [2022-12-20 01:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1000/1519] eta 0:08:41 lr 0.000018 time 0.9331 (1.0047) model_time 0.9330 (1.0035) loss 0.6921 (0.8430) grad_norm 10.1847 (8.6391/2.1941) mem 68106MB [2022-12-20 01:12:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1010/1519] eta 0:08:31 lr 0.000018 time 0.9906 (1.0047) model_time 0.9905 (1.0036) loss 0.7914 (0.8439) grad_norm 6.2531 (8.6291/2.1965) mem 68106MB [2022-12-20 01:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1020/1519] eta 0:08:21 lr 0.000018 time 0.9313 (1.0047) model_time 0.9311 (1.0035) loss 1.0488 (0.8441) grad_norm 7.0332 (8.6272/2.1902) mem 68106MB [2022-12-20 01:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1030/1519] eta 0:08:11 lr 0.000018 time 0.9308 (1.0047) model_time 0.9307 (1.0035) loss 0.7081 (0.8447) grad_norm 7.4737 (8.6377/2.1976) mem 68106MB [2022-12-20 01:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1040/1519] eta 0:08:01 lr 0.000018 time 0.9330 (1.0046) model_time 0.9329 (1.0035) loss 0.8000 (0.8444) grad_norm 6.8369 (8.6322/2.2059) mem 68106MB [2022-12-20 01:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1050/1519] eta 0:07:51 lr 0.000018 time 0.9351 (1.0047) model_time 0.9349 (1.0036) loss 0.7184 (0.8442) grad_norm 10.9349 (8.5955/2.1126) mem 68106MB [2022-12-20 01:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1060/1519] eta 0:07:41 lr 0.000018 time 0.9289 (1.0047) model_time 0.9287 (1.0036) loss 0.9886 (0.8452) grad_norm 5.3101 (8.5711/2.1031) mem 68106MB [2022-12-20 01:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1070/1519] eta 0:07:31 lr 0.000018 time 0.9253 (1.0047) model_time 0.9251 (1.0036) loss 0.6865 (0.8450) grad_norm 5.9055 (8.5625/2.1022) mem 68106MB [2022-12-20 01:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1080/1519] eta 0:07:21 lr 0.000018 time 0.9370 (1.0048) model_time 0.9369 (1.0036) loss 0.8662 (0.8455) grad_norm 9.6418 (8.5739/2.0927) mem 68106MB [2022-12-20 01:14:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1090/1519] eta 0:07:11 lr 0.000018 time 0.9377 (1.0048) model_time 0.9376 (1.0037) loss 0.8027 (0.8458) grad_norm 7.8872 (8.6138/2.1044) mem 68106MB [2022-12-20 01:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1100/1519] eta 0:07:00 lr 0.000018 time 0.9355 (1.0048) model_time 0.9354 (1.0036) loss 0.7938 (0.8461) grad_norm 8.0817 (8.6294/2.1228) mem 68106MB [2022-12-20 01:14:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1110/1519] eta 0:06:50 lr 0.000018 time 0.9329 (1.0047) model_time 0.9327 (1.0036) loss 0.9099 (0.8458) grad_norm 7.5655 (8.6355/2.1256) mem 68106MB [2022-12-20 01:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1120/1519] eta 0:06:40 lr 0.000018 time 0.9314 (1.0048) model_time 0.9313 (1.0037) loss 0.7281 (0.8459) grad_norm 7.4173 (8.6514/2.1241) mem 68106MB [2022-12-20 01:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1130/1519] eta 0:06:30 lr 0.000018 time 0.9431 (1.0048) model_time 0.9430 (1.0037) loss 1.3009 (0.8459) grad_norm 7.5116 (8.6513/2.1292) mem 68106MB [2022-12-20 01:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1140/1519] eta 0:06:20 lr 0.000018 time 0.9347 (1.0047) model_time 0.9346 (1.0036) loss 1.0598 (0.8465) grad_norm 8.9022 (8.6772/2.1345) mem 68106MB [2022-12-20 01:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1150/1519] eta 0:06:10 lr 0.000018 time 0.9345 (1.0047) model_time 0.9344 (1.0036) loss 0.8760 (0.8458) grad_norm 6.5333 (8.7003/2.1265) mem 68106MB [2022-12-20 01:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1160/1519] eta 0:06:00 lr 0.000018 time 0.9446 (1.0046) model_time 0.9445 (1.0036) loss 0.8639 (0.8463) grad_norm 9.9010 (8.7160/2.1412) mem 68106MB [2022-12-20 01:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1170/1519] eta 0:05:50 lr 0.000018 time 0.9275 (1.0046) model_time 0.9273 (1.0035) loss 0.6983 (0.8467) grad_norm 10.0769 (8.7246/2.1478) mem 68106MB [2022-12-20 01:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1180/1519] eta 0:05:40 lr 0.000018 time 0.9335 (1.0045) model_time 0.9333 (1.0035) loss 0.8368 (0.8469) grad_norm 11.6094 (8.7250/2.1191) mem 68106MB [2022-12-20 01:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1190/1519] eta 0:05:30 lr 0.000018 time 0.9908 (1.0048) model_time 0.9907 (1.0037) loss 0.9791 (0.8470) grad_norm 7.1134 (8.7228/2.1203) mem 68106MB [2022-12-20 01:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1200/1519] eta 0:05:20 lr 0.000018 time 0.9367 (1.0047) model_time 0.9366 (1.0037) loss 0.6798 (0.8468) grad_norm 6.9778 (8.7314/2.1265) mem 68106MB [2022-12-20 01:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1210/1519] eta 0:05:10 lr 0.000018 time 0.9319 (1.0046) model_time 0.9316 (1.0035) loss 0.6760 (0.8467) grad_norm 10.9725 (8.7499/2.1267) mem 68106MB [2022-12-20 01:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1220/1519] eta 0:05:00 lr 0.000018 time 0.9413 (1.0046) model_time 0.9411 (1.0035) loss 0.7843 (0.8469) grad_norm 6.9474 (8.7628/2.1212) mem 68106MB [2022-12-20 01:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1230/1519] eta 0:04:50 lr 0.000018 time 0.9307 (1.0046) model_time 0.9305 (1.0035) loss 0.7580 (0.8467) grad_norm 7.9158 (8.7558/2.1171) mem 68106MB [2022-12-20 01:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1240/1519] eta 0:04:40 lr 0.000018 time 0.9345 (1.0046) model_time 0.9344 (1.0035) loss 0.8009 (0.8462) grad_norm 6.7907 (8.7272/2.1172) mem 68106MB [2022-12-20 01:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1250/1519] eta 0:04:30 lr 0.000018 time 0.9315 (1.0046) model_time 0.9314 (1.0035) loss 0.8109 (0.8462) grad_norm 9.2668 (8.6751/2.0731) mem 68106MB [2022-12-20 01:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1260/1519] eta 0:04:20 lr 0.000018 time 0.9298 (1.0046) model_time 0.9297 (1.0036) loss 0.6814 (0.8463) grad_norm 8.8096 (8.6723/2.0638) mem 68106MB [2022-12-20 01:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1270/1519] eta 0:04:10 lr 0.000018 time 0.9298 (1.0048) model_time 0.9297 (1.0037) loss 0.8116 (0.8462) grad_norm 8.1890 (8.6703/2.0684) mem 68106MB [2022-12-20 01:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1280/1519] eta 0:04:00 lr 0.000018 time 0.9334 (1.0047) model_time 0.9332 (1.0037) loss 0.8416 (0.8468) grad_norm 7.4104 (8.6118/2.0270) mem 68106MB [2022-12-20 01:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1290/1519] eta 0:03:50 lr 0.000018 time 0.9300 (1.0047) model_time 0.9299 (1.0037) loss 0.8098 (0.8475) grad_norm 10.1096 (8.6329/2.0268) mem 68106MB [2022-12-20 01:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1300/1519] eta 0:03:40 lr 0.000018 time 0.9412 (1.0047) model_time 0.9410 (1.0037) loss 0.6790 (0.8471) grad_norm 7.7902 (8.6289/2.0131) mem 68106MB [2022-12-20 01:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1310/1519] eta 0:03:29 lr 0.000018 time 0.9307 (1.0047) model_time 0.9306 (1.0037) loss 0.8611 (0.8467) grad_norm 7.5610 (8.6044/1.9908) mem 68106MB [2022-12-20 01:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1320/1519] eta 0:03:19 lr 0.000018 time 0.9336 (1.0046) model_time 0.9335 (1.0036) loss 0.6793 (0.8467) grad_norm 8.1612 (8.5943/1.9789) mem 68106MB [2022-12-20 01:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1330/1519] eta 0:03:09 lr 0.000018 time 0.9322 (1.0046) model_time 0.9321 (1.0036) loss 0.6699 (0.8471) grad_norm 5.7147 (8.6006/2.0102) mem 68106MB [2022-12-20 01:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1340/1519] eta 0:02:59 lr 0.000018 time 0.9320 (1.0046) model_time 0.9318 (1.0036) loss 0.7537 (0.8470) grad_norm 10.9573 (8.5940/2.0080) mem 68106MB [2022-12-20 01:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1350/1519] eta 0:02:49 lr 0.000018 time 0.9368 (1.0046) model_time 0.9366 (1.0036) loss 1.0704 (0.8473) grad_norm 9.2492 (8.6152/2.0020) mem 68106MB [2022-12-20 01:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1360/1519] eta 0:02:39 lr 0.000018 time 0.9273 (1.0045) model_time 0.9271 (1.0036) loss 1.3459 (0.8472) grad_norm 7.7332 (8.6079/2.0045) mem 68106MB [2022-12-20 01:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1370/1519] eta 0:02:29 lr 0.000018 time 0.9369 (1.0046) model_time 0.9368 (1.0036) loss 0.7158 (0.8467) grad_norm 9.5189 (8.6194/1.9946) mem 68106MB [2022-12-20 01:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1380/1519] eta 0:02:19 lr 0.000018 time 0.9335 (1.0045) model_time 0.9334 (1.0036) loss 0.7749 (0.8465) grad_norm 12.0532 (8.5798/1.7738) mem 68106MB [2022-12-20 01:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1390/1519] eta 0:02:09 lr 0.000018 time 0.9243 (1.0045) model_time 0.9240 (1.0036) loss 0.9055 (0.8465) grad_norm 8.7252 (8.6074/1.7827) mem 68106MB [2022-12-20 01:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1400/1519] eta 0:01:59 lr 0.000018 time 0.9268 (1.0045) model_time 0.9266 (1.0035) loss 0.7361 (0.8463) grad_norm 6.6889 (8.5984/1.7854) mem 68106MB [2022-12-20 01:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1410/1519] eta 0:01:49 lr 0.000018 time 0.9307 (1.0045) model_time 0.9306 (1.0035) loss 0.7822 (0.8467) grad_norm 6.8849 (8.5903/1.7879) mem 68106MB [2022-12-20 01:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1420/1519] eta 0:01:39 lr 0.000018 time 0.9256 (1.0045) model_time 0.9252 (1.0035) loss 0.8188 (0.8468) grad_norm 9.7460 (8.5868/1.7949) mem 68106MB [2022-12-20 01:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1430/1519] eta 0:01:29 lr 0.000018 time 0.9294 (1.0045) model_time 0.9291 (1.0036) loss 1.0553 (0.8474) grad_norm 6.3580 (8.5807/1.7925) mem 68106MB [2022-12-20 01:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1440/1519] eta 0:01:19 lr 0.000018 time 0.9279 (1.0045) model_time 0.9278 (1.0036) loss 1.3327 (0.8473) grad_norm 6.7413 (8.5766/1.7795) mem 68106MB [2022-12-20 01:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1450/1519] eta 0:01:09 lr 0.000018 time 0.9234 (1.0046) model_time 0.9232 (1.0036) loss 0.7230 (0.8466) grad_norm 8.4254 (8.5574/1.7738) mem 68106MB [2022-12-20 01:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1460/1519] eta 0:00:59 lr 0.000018 time 0.9053 (1.0045) model_time 0.9052 (1.0036) loss 0.8201 (0.8464) grad_norm 14.0656 (8.5789/1.8015) mem 68106MB [2022-12-20 01:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1470/1519] eta 0:00:49 lr 0.000018 time 0.9333 (1.0045) model_time 0.9332 (1.0036) loss 0.9584 (0.8464) grad_norm 6.3695 (8.5525/1.7965) mem 68106MB [2022-12-20 01:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1480/1519] eta 0:00:39 lr 0.000018 time 0.9367 (1.0045) model_time 0.9365 (1.0035) loss 0.7392 (0.8467) grad_norm 8.0533 (8.5501/1.7690) mem 68106MB [2022-12-20 01:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1490/1519] eta 0:00:29 lr 0.000018 time 0.9218 (1.0044) model_time 0.9217 (1.0035) loss 1.0624 (0.8469) grad_norm 9.6048 (8.5767/1.7713) mem 68106MB [2022-12-20 01:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1500/1519] eta 0:00:19 lr 0.000018 time 0.9304 (1.0044) model_time 0.9302 (1.0035) loss 0.8106 (0.8467) grad_norm 12.4739 (8.6073/1.7977) mem 68106MB [2022-12-20 01:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [47/100][1510/1519] eta 0:00:09 lr 0.000018 time 0.9246 (1.0045) model_time 0.9245 (1.0035) loss 0.8005 (0.8465) grad_norm 9.1698 (8.6004/1.7979) mem 68106MB [2022-12-20 01:21:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 47 training takes 0:25:25 [2022-12-20 01:21:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_47.pth saving...... [2022-12-20 01:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_47.pth saved !!! [2022-12-20 01:21:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.677 (0.677) Loss 0.5313 (0.5313) Acc@1 90.625 (90.625) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 01:21:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.331) Loss 0.5150 (0.4933) Acc@1 91.667 (92.045) Acc@5 97.917 (98.327) Mem 68106MB [2022-12-20 01:21:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4628 (0.4919) Acc@1 92.361 (91.882) Acc@5 99.306 (98.330) Mem 68106MB [2022-12-20 01:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.6032 (0.4967) Acc@1 89.583 (91.756) Acc@5 97.917 (98.275) Mem 68106MB [2022-12-20 01:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4763 (0.4889) Acc@1 92.361 (91.921) Acc@5 98.264 (98.349) Mem 68106MB [2022-12-20 01:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.4683 (0.4861) Acc@1 91.319 (91.959) Acc@5 99.306 (98.414) Mem 68106MB [2022-12-20 01:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.5006 (0.4844) Acc@1 90.972 (91.985) Acc@5 97.569 (98.389) Mem 68106MB [2022-12-20 01:22:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.302 (0.304) Loss 0.5136 (0.4848) Acc@1 93.056 (91.931) Acc@5 98.611 (98.406) Mem 68106MB [2022-12-20 01:22:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.303) Loss 0.4282 (0.4832) Acc@1 93.403 (91.932) Acc@5 98.611 (98.453) Mem 68106MB [2022-12-20 01:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:47] * Acc@1 91.880 Acc@5 98.453 [2022-12-20 01:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 91.9% [2022-12-20 01:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.10% [2022-12-20 01:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][0/1519] eta 0:46:48 lr 0.000018 time 1.8489 (1.8489) model_time 1.0409 (1.0409) loss 0.8485 (0.8485) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 01:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][10/1519] eta 0:27:21 lr 0.000018 time 0.9371 (1.0877) model_time 0.9369 (1.0138) loss 0.9569 (0.8718) grad_norm 7.9336 (8.6867/0.5577) mem 68106MB [2022-12-20 01:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][20/1519] eta 0:26:09 lr 0.000018 time 0.9669 (1.0471) model_time 0.9668 (1.0082) loss 0.6829 (0.8860) grad_norm 6.7016 (8.3166/1.1312) mem 68106MB [2022-12-20 01:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][30/1519] eta 0:25:48 lr 0.000018 time 0.9496 (1.0400) model_time 0.9495 (1.0135) loss 0.6974 (0.8750) grad_norm 9.9380 (8.5798/1.8762) mem 68106MB [2022-12-20 01:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][40/1519] eta 0:25:23 lr 0.000018 time 0.9363 (1.0300) model_time 0.9361 (1.0098) loss 0.8691 (0.8810) grad_norm 10.6145 (8.7117/1.7979) mem 68106MB [2022-12-20 01:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][50/1519] eta 0:25:07 lr 0.000018 time 0.9236 (1.0262) model_time 0.9234 (1.0099) loss 0.7728 (0.8672) grad_norm 8.6879 (8.6140/1.7222) mem 68106MB [2022-12-20 01:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][60/1519] eta 0:24:51 lr 0.000018 time 0.9319 (1.0222) model_time 0.9316 (1.0086) loss 0.7456 (0.8578) grad_norm 6.4682 (8.4266/1.7249) mem 68106MB [2022-12-20 01:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][70/1519] eta 0:24:36 lr 0.000018 time 0.9270 (1.0189) model_time 0.9268 (1.0071) loss 0.7613 (0.8516) grad_norm 5.9573 (8.3630/1.7637) mem 68106MB [2022-12-20 01:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][80/1519] eta 0:24:23 lr 0.000018 time 0.9675 (1.0169) model_time 0.9674 (1.0065) loss 0.7563 (0.8536) grad_norm 7.5120 (8.2625/1.7121) mem 68106MB [2022-12-20 01:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][90/1519] eta 0:24:13 lr 0.000018 time 0.9339 (1.0169) model_time 0.9337 (1.0076) loss 0.7406 (0.8443) grad_norm 9.1373 (8.2692/1.6324) mem 68106MB [2022-12-20 01:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][100/1519] eta 0:24:01 lr 0.000018 time 0.9692 (1.0158) model_time 0.9690 (1.0074) loss 0.9798 (0.8429) grad_norm 7.8805 (8.2456/1.6112) mem 68106MB [2022-12-20 01:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][110/1519] eta 0:23:49 lr 0.000018 time 0.9305 (1.0142) model_time 0.9303 (1.0065) loss 0.7230 (0.8424) grad_norm 10.3296 (8.2669/1.5677) mem 68106MB [2022-12-20 01:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][120/1519] eta 0:23:37 lr 0.000018 time 0.9353 (1.0130) model_time 0.9352 (1.0059) loss 0.9195 (0.8449) grad_norm 8.3610 (8.3187/1.5639) mem 68106MB [2022-12-20 01:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][130/1519] eta 0:23:28 lr 0.000018 time 0.9285 (1.0137) model_time 0.9284 (1.0071) loss 0.8946 (0.8536) grad_norm 6.6040 (8.2412/1.5618) mem 68106MB [2022-12-20 01:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][140/1519] eta 0:23:16 lr 0.000018 time 0.9301 (1.0126) model_time 0.9298 (1.0064) loss 0.7419 (0.8518) grad_norm 10.4859 (8.3017/1.6035) mem 68106MB [2022-12-20 01:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][150/1519] eta 0:23:05 lr 0.000018 time 0.9356 (1.0119) model_time 0.9355 (1.0061) loss 0.9060 (0.8488) grad_norm 6.2513 (8.2477/1.5902) mem 68106MB [2022-12-20 01:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][160/1519] eta 0:22:54 lr 0.000018 time 0.9282 (1.0116) model_time 0.9280 (1.0061) loss 0.6968 (0.8434) grad_norm 7.5985 (8.2720/1.5957) mem 68106MB [2022-12-20 01:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][170/1519] eta 0:22:43 lr 0.000018 time 0.9182 (1.0110) model_time 0.9180 (1.0059) loss 0.7598 (0.8419) grad_norm 8.2363 (8.3209/1.6573) mem 68106MB [2022-12-20 01:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][180/1519] eta 0:22:32 lr 0.000018 time 0.9359 (1.0103) model_time 0.9358 (1.0055) loss 0.6861 (0.8411) grad_norm 8.4965 (8.3074/1.6195) mem 68106MB [2022-12-20 01:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][190/1519] eta 0:22:22 lr 0.000018 time 0.9352 (1.0099) model_time 0.9351 (1.0053) loss 0.9370 (0.8444) grad_norm 11.3582 (8.4175/1.7978) mem 68106MB [2022-12-20 01:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][200/1519] eta 0:22:11 lr 0.000018 time 0.9265 (1.0095) model_time 0.9264 (1.0051) loss 0.8261 (0.8436) grad_norm 7.3857 (8.3909/1.7677) mem 68106MB [2022-12-20 01:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][210/1519] eta 0:22:00 lr 0.000018 time 0.9319 (1.0090) model_time 0.9317 (1.0048) loss 0.7031 (0.8428) grad_norm 8.7827 (8.3816/1.7454) mem 68106MB [2022-12-20 01:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][220/1519] eta 0:21:50 lr 0.000018 time 0.9281 (1.0085) model_time 0.9279 (1.0044) loss 0.7317 (0.8404) grad_norm 10.4346 (8.4281/1.9127) mem 68106MB [2022-12-20 01:26:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][230/1519] eta 0:21:39 lr 0.000018 time 0.9312 (1.0081) model_time 0.9311 (1.0043) loss 1.1798 (0.8452) grad_norm 12.3715 (8.4503/1.9191) mem 68106MB [2022-12-20 01:26:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][240/1519] eta 0:21:29 lr 0.000018 time 0.9763 (1.0080) model_time 0.9762 (1.0042) loss 1.0196 (0.8460) grad_norm 7.3746 (8.4621/1.9540) mem 68106MB [2022-12-20 01:26:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][250/1519] eta 0:21:18 lr 0.000018 time 0.9346 (1.0077) model_time 0.9343 (1.0041) loss 0.6975 (0.8439) grad_norm 10.0798 (8.4535/1.9811) mem 68106MB [2022-12-20 01:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][260/1519] eta 0:21:08 lr 0.000018 time 0.9212 (1.0075) model_time 0.9211 (1.0040) loss 0.9229 (0.8441) grad_norm 7.8708 (8.4898/1.9830) mem 68106MB [2022-12-20 01:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][270/1519] eta 0:20:57 lr 0.000018 time 0.9293 (1.0072) model_time 0.9291 (1.0038) loss 0.8821 (0.8432) grad_norm 7.9899 (8.4804/1.9860) mem 68106MB [2022-12-20 01:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][280/1519] eta 0:20:47 lr 0.000018 time 0.9365 (1.0069) model_time 0.9364 (1.0036) loss 0.7082 (0.8425) grad_norm 6.8928 (8.5092/1.9886) mem 68106MB [2022-12-20 01:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][290/1519] eta 0:20:37 lr 0.000018 time 0.9364 (1.0066) model_time 0.9362 (1.0035) loss 0.9462 (0.8427) grad_norm 5.4762 (8.5327/2.0181) mem 68106MB [2022-12-20 01:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][300/1519] eta 0:20:27 lr 0.000018 time 0.9328 (1.0069) model_time 0.9326 (1.0038) loss 0.6744 (0.8414) grad_norm 8.3682 (8.6292/2.2542) mem 68106MB [2022-12-20 01:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][310/1519] eta 0:20:17 lr 0.000018 time 0.9339 (1.0066) model_time 0.9338 (1.0036) loss 0.6882 (0.8433) grad_norm 9.0414 (8.6259/2.2201) mem 68106MB [2022-12-20 01:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][320/1519] eta 0:20:06 lr 0.000018 time 0.9280 (1.0066) model_time 0.9279 (1.0037) loss 0.9852 (0.8435) grad_norm 8.5993 (8.6180/2.1948) mem 68106MB [2022-12-20 01:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][330/1519] eta 0:19:56 lr 0.000018 time 0.9348 (1.0064) model_time 0.9346 (1.0035) loss 0.7014 (0.8429) grad_norm 5.6898 (8.6309/2.2287) mem 68106MB [2022-12-20 01:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][340/1519] eta 0:19:46 lr 0.000018 time 0.9847 (1.0064) model_time 0.9846 (1.0036) loss 0.6777 (0.8423) grad_norm 8.5990 (8.6240/2.1984) mem 68106MB [2022-12-20 01:28:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][350/1519] eta 0:19:36 lr 0.000018 time 0.9324 (1.0061) model_time 0.9322 (1.0034) loss 0.8675 (0.8424) grad_norm 6.4883 (8.5952/2.1772) mem 68106MB [2022-12-20 01:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][360/1519] eta 0:19:26 lr 0.000018 time 1.0224 (1.0063) model_time 1.0221 (1.0036) loss 0.8310 (0.8421) grad_norm 8.6158 (8.5753/2.1526) mem 68106MB [2022-12-20 01:28:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][370/1519] eta 0:19:16 lr 0.000018 time 0.9240 (1.0061) model_time 0.9238 (1.0035) loss 0.6719 (0.8447) grad_norm 8.3857 (8.5885/2.1405) mem 68106MB [2022-12-20 01:28:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][380/1519] eta 0:19:06 lr 0.000018 time 0.9340 (1.0062) model_time 0.9339 (1.0037) loss 0.9946 (0.8441) grad_norm 10.2969 (8.5983/2.1429) mem 68106MB [2022-12-20 01:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][390/1519] eta 0:18:55 lr 0.000018 time 0.9346 (1.0061) model_time 0.9342 (1.0037) loss 1.1587 (0.8443) grad_norm 7.9870 (8.5852/2.1181) mem 68106MB [2022-12-20 01:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][400/1519] eta 0:18:46 lr 0.000018 time 0.8890 (1.0063) model_time 0.8889 (1.0039) loss 0.8600 (0.8444) grad_norm 7.8401 (8.5554/2.1015) mem 68106MB [2022-12-20 01:29:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][410/1519] eta 0:18:35 lr 0.000018 time 0.9273 (1.0063) model_time 0.9271 (1.0039) loss 0.7050 (0.8436) grad_norm 7.0560 (8.5456/2.0951) mem 68106MB [2022-12-20 01:29:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][420/1519] eta 0:18:25 lr 0.000018 time 0.9283 (1.0062) model_time 0.9282 (1.0039) loss 0.8884 (0.8445) grad_norm 6.8712 (8.5458/2.1065) mem 68106MB [2022-12-20 01:29:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][430/1519] eta 0:18:15 lr 0.000018 time 0.9266 (1.0061) model_time 0.9265 (1.0038) loss 0.9470 (0.8453) grad_norm 9.5711 (8.5582/2.0883) mem 68106MB [2022-12-20 01:29:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][440/1519] eta 0:18:05 lr 0.000018 time 0.9161 (1.0062) model_time 0.9159 (1.0039) loss 0.6951 (0.8448) grad_norm 13.5453 (8.5599/2.1033) mem 68106MB [2022-12-20 01:29:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][450/1519] eta 0:17:55 lr 0.000018 time 0.9273 (1.0060) model_time 0.9272 (1.0038) loss 0.7387 (0.8456) grad_norm 6.2341 (8.5447/2.0961) mem 68106MB [2022-12-20 01:29:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][460/1519] eta 0:17:45 lr 0.000018 time 0.9236 (1.0059) model_time 0.9234 (1.0037) loss 0.8152 (0.8452) grad_norm 6.8954 (8.5422/2.0814) mem 68106MB [2022-12-20 01:30:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][470/1519] eta 0:17:35 lr 0.000018 time 0.9213 (1.0059) model_time 0.9211 (1.0038) loss 0.8953 (0.8464) grad_norm 7.6139 (8.5526/2.0692) mem 68106MB [2022-12-20 01:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][480/1519] eta 0:17:25 lr 0.000018 time 0.9189 (1.0060) model_time 0.9187 (1.0039) loss 0.7919 (0.8475) grad_norm 10.7338 (8.5550/2.0555) mem 68106MB [2022-12-20 01:30:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][490/1519] eta 0:17:15 lr 0.000018 time 0.9222 (1.0059) model_time 0.9220 (1.0038) loss 0.8872 (0.8474) grad_norm 7.1640 (8.5508/2.0531) mem 68106MB [2022-12-20 01:30:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][500/1519] eta 0:17:04 lr 0.000018 time 0.9276 (1.0057) model_time 0.9275 (1.0037) loss 0.7279 (0.8476) grad_norm 11.2531 (8.5648/2.0417) mem 68106MB [2022-12-20 01:30:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][510/1519] eta 0:16:54 lr 0.000018 time 0.9306 (1.0055) model_time 0.9304 (1.0035) loss 0.9775 (0.8478) grad_norm 10.8371 (8.5655/2.0286) mem 68106MB [2022-12-20 01:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][520/1519] eta 0:16:44 lr 0.000018 time 0.9765 (1.0056) model_time 0.9763 (1.0036) loss 0.8050 (0.8483) grad_norm 7.8419 (8.5999/2.0524) mem 68106MB [2022-12-20 01:31:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][530/1519] eta 0:16:34 lr 0.000018 time 0.9294 (1.0054) model_time 0.9293 (1.0035) loss 0.8933 (0.8498) grad_norm 8.1200 (8.6445/2.1007) mem 68106MB [2022-12-20 01:31:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][540/1519] eta 0:16:24 lr 0.000018 time 1.0362 (1.0056) model_time 1.0360 (1.0037) loss 0.7425 (0.8483) grad_norm 7.2972 (8.6166/2.0927) mem 68106MB [2022-12-20 01:31:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][550/1519] eta 0:16:14 lr 0.000018 time 0.9285 (1.0055) model_time 0.9284 (1.0036) loss 1.2850 (0.8487) grad_norm 6.9248 (8.6256/2.0830) mem 68106MB [2022-12-20 01:31:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][560/1519] eta 0:16:04 lr 0.000018 time 0.9395 (1.0055) model_time 0.9393 (1.0036) loss 0.9298 (0.8494) grad_norm 6.9760 (8.6190/2.0692) mem 68106MB [2022-12-20 01:31:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][570/1519] eta 0:15:54 lr 0.000018 time 0.9415 (1.0054) model_time 0.9414 (1.0035) loss 0.8598 (0.8501) grad_norm 9.5906 (8.6108/2.0555) mem 68106MB [2022-12-20 01:31:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][580/1519] eta 0:15:43 lr 0.000018 time 0.9300 (1.0053) model_time 0.9298 (1.0035) loss 0.6947 (0.8488) grad_norm 8.4953 (8.5997/2.0437) mem 68106MB [2022-12-20 01:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][590/1519] eta 0:15:33 lr 0.000018 time 0.9309 (1.0052) model_time 0.9307 (1.0034) loss 1.2331 (0.8495) grad_norm 9.6395 (8.6085/2.0318) mem 68106MB [2022-12-20 01:32:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][600/1519] eta 0:15:23 lr 0.000018 time 0.9311 (1.0051) model_time 0.9308 (1.0034) loss 0.6858 (0.8490) grad_norm 6.6713 (8.6030/2.0217) mem 68106MB [2022-12-20 01:32:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][610/1519] eta 0:15:13 lr 0.000018 time 0.9256 (1.0051) model_time 0.9253 (1.0033) loss 0.8551 (0.8491) grad_norm 7.9519 (8.5943/2.0445) mem 68106MB [2022-12-20 01:32:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][620/1519] eta 0:15:03 lr 0.000018 time 0.9249 (1.0050) model_time 0.9247 (1.0033) loss 0.8304 (0.8496) grad_norm 8.5239 (8.6062/2.0388) mem 68106MB [2022-12-20 01:32:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][630/1519] eta 0:14:53 lr 0.000018 time 1.0097 (1.0051) model_time 1.0096 (1.0034) loss 0.8463 (0.8512) grad_norm 7.8734 (8.5807/2.0132) mem 68106MB [2022-12-20 01:32:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][640/1519] eta 0:14:43 lr 0.000018 time 0.9287 (1.0050) model_time 0.9286 (1.0033) loss 0.7725 (0.8508) grad_norm 6.9644 (8.5501/2.0153) mem 68106MB [2022-12-20 01:33:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][650/1519] eta 0:14:33 lr 0.000018 time 0.9349 (1.0049) model_time 0.9347 (1.0033) loss 0.6772 (0.8500) grad_norm 10.4288 (8.5633/2.0181) mem 68106MB [2022-12-20 01:33:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][660/1519] eta 0:14:23 lr 0.000018 time 0.9330 (1.0048) model_time 0.9327 (1.0032) loss 1.0252 (0.8505) grad_norm 9.2836 (8.5941/2.0130) mem 68106MB [2022-12-20 01:33:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][670/1519] eta 0:14:12 lr 0.000018 time 0.9312 (1.0047) model_time 0.9311 (1.0031) loss 1.1038 (0.8519) grad_norm 7.3718 (8.6034/2.0113) mem 68106MB [2022-12-20 01:33:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][680/1519] eta 0:14:02 lr 0.000018 time 0.9174 (1.0048) model_time 0.9172 (1.0032) loss 1.1105 (0.8524) grad_norm 6.2651 (8.6095/2.0162) mem 68106MB [2022-12-20 01:33:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][690/1519] eta 0:13:52 lr 0.000018 time 0.9293 (1.0048) model_time 0.9292 (1.0032) loss 1.1156 (0.8527) grad_norm 7.8630 (8.6114/2.0275) mem 68106MB [2022-12-20 01:33:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][700/1519] eta 0:13:42 lr 0.000018 time 0.9277 (1.0047) model_time 0.9275 (1.0032) loss 0.7647 (0.8518) grad_norm 7.5038 (8.6139/2.0328) mem 68106MB [2022-12-20 01:34:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][710/1519] eta 0:13:32 lr 0.000018 time 0.9339 (1.0046) model_time 0.9337 (1.0031) loss 1.0349 (0.8517) grad_norm 12.3534 (8.6234/2.0540) mem 68106MB [2022-12-20 01:34:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][720/1519] eta 0:13:23 lr 0.000018 time 0.9863 (1.0050) model_time 0.9862 (1.0035) loss 0.9099 (0.8511) grad_norm 6.6859 (8.6081/2.0511) mem 68106MB [2022-12-20 01:34:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][730/1519] eta 0:13:12 lr 0.000018 time 0.9214 (1.0049) model_time 0.9213 (1.0034) loss 0.7531 (0.8511) grad_norm 8.7631 (8.6221/2.0468) mem 68106MB [2022-12-20 01:34:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][740/1519] eta 0:13:02 lr 0.000018 time 0.9305 (1.0050) model_time 0.9304 (1.0035) loss 0.6999 (0.8506) grad_norm 6.2848 (8.6069/2.0432) mem 68106MB [2022-12-20 01:34:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][750/1519] eta 0:12:52 lr 0.000018 time 0.9271 (1.0049) model_time 0.9269 (1.0034) loss 0.7392 (0.8510) grad_norm 8.1030 (8.5995/2.0477) mem 68106MB [2022-12-20 01:34:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][760/1519] eta 0:12:42 lr 0.000018 time 0.9291 (1.0050) model_time 0.9290 (1.0035) loss 0.9400 (0.8509) grad_norm 13.2314 (8.6184/2.0803) mem 68106MB [2022-12-20 01:35:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][770/1519] eta 0:12:32 lr 0.000018 time 0.9337 (1.0049) model_time 0.9335 (1.0035) loss 0.7600 (0.8511) grad_norm 9.3519 (8.6125/2.0675) mem 68106MB [2022-12-20 01:35:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][780/1519] eta 0:12:22 lr 0.000018 time 0.9401 (1.0049) model_time 0.9400 (1.0035) loss 0.9136 (0.8509) grad_norm 6.4161 (8.6069/2.0729) mem 68106MB [2022-12-20 01:35:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][790/1519] eta 0:12:12 lr 0.000018 time 0.9148 (1.0050) model_time 0.9147 (1.0036) loss 0.7927 (0.8503) grad_norm 8.1850 (8.5806/2.0227) mem 68106MB [2022-12-20 01:35:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][800/1519] eta 0:12:02 lr 0.000018 time 0.9269 (1.0051) model_time 0.9267 (1.0036) loss 0.7327 (0.8510) grad_norm 6.7634 (8.5776/2.0248) mem 68106MB [2022-12-20 01:35:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][810/1519] eta 0:11:52 lr 0.000018 time 1.2042 (1.0053) model_time 1.2041 (1.0039) loss 0.9758 (0.8511) grad_norm 6.2064 (8.5655/2.0310) mem 68106MB [2022-12-20 01:35:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][820/1519] eta 0:11:42 lr 0.000018 time 0.9364 (1.0052) model_time 0.9363 (1.0039) loss 1.0580 (0.8504) grad_norm 7.8455 (8.5872/2.0224) mem 68106MB [2022-12-20 01:36:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][830/1519] eta 0:11:32 lr 0.000018 time 0.9404 (1.0052) model_time 0.9403 (1.0038) loss 0.9080 (0.8500) grad_norm 6.0946 (8.5836/2.0167) mem 68106MB [2022-12-20 01:36:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][840/1519] eta 0:11:22 lr 0.000018 time 0.9407 (1.0052) model_time 0.9406 (1.0039) loss 0.9551 (0.8508) grad_norm 6.9850 (8.5614/1.9982) mem 68106MB [2022-12-20 01:36:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][850/1519] eta 0:11:12 lr 0.000018 time 0.9352 (1.0054) model_time 0.9349 (1.0040) loss 0.6963 (0.8504) grad_norm 6.8255 (8.5539/1.9816) mem 68106MB [2022-12-20 01:36:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][860/1519] eta 0:11:02 lr 0.000018 time 0.9307 (1.0055) model_time 0.9306 (1.0041) loss 1.0800 (0.8504) grad_norm 7.3357 (8.5235/1.9728) mem 68106MB [2022-12-20 01:36:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][870/1519] eta 0:10:52 lr 0.000018 time 0.9324 (1.0054) model_time 0.9323 (1.0040) loss 0.7884 (0.8509) grad_norm 7.0346 (8.5149/1.9630) mem 68106MB [2022-12-20 01:36:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][880/1519] eta 0:10:42 lr 0.000018 time 0.9331 (1.0053) model_time 0.9329 (1.0040) loss 0.9317 (0.8515) grad_norm 8.6796 (8.5004/1.9526) mem 68106MB [2022-12-20 01:37:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][890/1519] eta 0:10:32 lr 0.000018 time 0.9340 (1.0053) model_time 0.9338 (1.0040) loss 1.0803 (0.8529) grad_norm 6.9494 (8.4926/1.9376) mem 68106MB [2022-12-20 01:37:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][900/1519] eta 0:10:22 lr 0.000018 time 0.9197 (1.0052) model_time 0.9196 (1.0039) loss 0.7073 (0.8530) grad_norm 7.4421 (8.4576/1.8072) mem 68106MB [2022-12-20 01:37:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][910/1519] eta 0:10:12 lr 0.000018 time 0.9207 (1.0052) model_time 0.9206 (1.0039) loss 0.8546 (0.8528) grad_norm 7.9256 (8.4480/1.8119) mem 68106MB [2022-12-20 01:37:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][920/1519] eta 0:10:02 lr 0.000018 time 1.2057 (1.0055) model_time 1.2056 (1.0042) loss 1.5048 (0.8539) grad_norm 11.8715 (8.4538/1.8178) mem 68106MB [2022-12-20 01:37:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][930/1519] eta 0:09:52 lr 0.000018 time 0.9602 (1.0055) model_time 0.9601 (1.0042) loss 0.6694 (0.8531) grad_norm 7.4190 (8.4349/1.7818) mem 68106MB [2022-12-20 01:37:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][940/1519] eta 0:09:42 lr 0.000018 time 0.9206 (1.0054) model_time 0.9205 (1.0041) loss 1.0663 (0.8532) grad_norm 9.8836 (8.4333/1.7858) mem 68106MB [2022-12-20 01:38:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][950/1519] eta 0:09:32 lr 0.000018 time 0.9332 (1.0053) model_time 0.9331 (1.0041) loss 0.7304 (0.8529) grad_norm 10.8055 (8.4400/1.7916) mem 68106MB [2022-12-20 01:38:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][960/1519] eta 0:09:21 lr 0.000018 time 0.9236 (1.0053) model_time 0.9235 (1.0041) loss 0.7394 (0.8528) grad_norm 8.0208 (8.4520/1.7897) mem 68106MB [2022-12-20 01:38:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][970/1519] eta 0:09:11 lr 0.000018 time 0.9300 (1.0053) model_time 0.9298 (1.0041) loss 1.0180 (0.8525) grad_norm 7.8488 (8.4234/1.7829) mem 68106MB [2022-12-20 01:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][980/1519] eta 0:09:01 lr 0.000018 time 0.9213 (1.0053) model_time 0.9211 (1.0040) loss 1.0191 (0.8526) grad_norm 7.1630 (8.4094/1.7605) mem 68106MB [2022-12-20 01:38:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][990/1519] eta 0:08:51 lr 0.000018 time 0.9361 (1.0052) model_time 0.9360 (1.0040) loss 0.8658 (0.8523) grad_norm 10.9917 (8.4142/1.7707) mem 68106MB [2022-12-20 01:38:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1000/1519] eta 0:08:41 lr 0.000018 time 0.9991 (1.0052) model_time 0.9987 (1.0040) loss 0.9785 (0.8524) grad_norm 7.6903 (8.4265/1.7680) mem 68106MB [2022-12-20 01:39:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1010/1519] eta 0:08:31 lr 0.000018 time 0.9246 (1.0052) model_time 0.9244 (1.0040) loss 0.8254 (0.8530) grad_norm 7.9155 (8.4229/1.7617) mem 68106MB [2022-12-20 01:39:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1020/1519] eta 0:08:21 lr 0.000018 time 0.9249 (1.0052) model_time 0.9245 (1.0039) loss 0.6995 (0.8528) grad_norm 9.2649 (8.4303/1.7410) mem 68106MB [2022-12-20 01:39:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1030/1519] eta 0:08:11 lr 0.000018 time 0.9259 (1.0053) model_time 0.9257 (1.0041) loss 0.7082 (0.8520) grad_norm 5.6985 (8.3971/1.7572) mem 68106MB [2022-12-20 01:39:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1040/1519] eta 0:08:01 lr 0.000018 time 0.9264 (1.0052) model_time 0.9263 (1.0040) loss 0.6827 (0.8522) grad_norm 7.0394 (8.3751/1.7297) mem 68106MB [2022-12-20 01:39:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1050/1519] eta 0:07:51 lr 0.000018 time 0.9220 (1.0054) model_time 0.9219 (1.0042) loss 0.7569 (0.8519) grad_norm 9.2650 (8.3743/1.7314) mem 68106MB [2022-12-20 01:39:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1060/1519] eta 0:07:41 lr 0.000018 time 0.9318 (1.0054) model_time 0.9316 (1.0042) loss 0.8181 (0.8524) grad_norm 8.1392 (8.3678/1.7279) mem 68106MB [2022-12-20 01:40:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1070/1519] eta 0:07:31 lr 0.000018 time 0.9299 (1.0054) model_time 0.9298 (1.0042) loss 0.6931 (0.8523) grad_norm 7.1994 (8.3810/1.8034) mem 68106MB [2022-12-20 01:40:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1080/1519] eta 0:07:21 lr 0.000018 time 0.9295 (1.0053) model_time 0.9294 (1.0041) loss 0.7482 (0.8531) grad_norm 7.2385 (8.3618/1.8022) mem 68106MB [2022-12-20 01:40:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1090/1519] eta 0:07:11 lr 0.000018 time 0.9254 (1.0052) model_time 0.9251 (1.0041) loss 0.9731 (0.8526) grad_norm 9.3737 (8.3684/1.8000) mem 68106MB [2022-12-20 01:40:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1100/1519] eta 0:07:01 lr 0.000018 time 0.9853 (1.0052) model_time 0.9852 (1.0041) loss 0.6959 (0.8520) grad_norm 8.2847 (8.3414/1.8006) mem 68106MB [2022-12-20 01:40:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1110/1519] eta 0:06:51 lr 0.000018 time 0.9266 (1.0052) model_time 0.9265 (1.0041) loss 1.1423 (0.8529) grad_norm 10.3163 (8.3387/1.8091) mem 68106MB [2022-12-20 01:40:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1120/1519] eta 0:06:41 lr 0.000018 time 0.9326 (1.0051) model_time 0.9325 (1.0040) loss 1.1016 (0.8532) grad_norm 8.6794 (8.3271/1.7883) mem 68106MB [2022-12-20 01:41:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1130/1519] eta 0:06:31 lr 0.000018 time 0.9302 (1.0052) model_time 0.9299 (1.0041) loss 1.0510 (0.8535) grad_norm 8.3872 (8.3005/1.7334) mem 68106MB [2022-12-20 01:41:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1140/1519] eta 0:06:20 lr 0.000018 time 0.9313 (1.0052) model_time 0.9311 (1.0041) loss 1.0292 (0.8533) grad_norm 14.3099 (8.3569/1.7691) mem 68106MB [2022-12-20 01:41:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1150/1519] eta 0:06:10 lr 0.000018 time 0.9269 (1.0052) model_time 0.9267 (1.0041) loss 0.7393 (0.8531) grad_norm 10.3393 (8.3670/1.7682) mem 68106MB [2022-12-20 01:41:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1160/1519] eta 0:06:00 lr 0.000018 time 1.0113 (1.0052) model_time 1.0091 (1.0041) loss 0.7557 (0.8535) grad_norm 10.1279 (8.3690/1.7776) mem 68106MB [2022-12-20 01:41:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1170/1519] eta 0:05:50 lr 0.000018 time 0.9204 (1.0053) model_time 0.9203 (1.0042) loss 1.1336 (0.8535) grad_norm 7.5611 (8.3757/1.7759) mem 68106MB [2022-12-20 01:41:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1180/1519] eta 0:05:40 lr 0.000018 time 0.9349 (1.0052) model_time 0.9348 (1.0041) loss 0.7964 (0.8535) grad_norm 13.4109 (8.3975/1.8010) mem 68106MB [2022-12-20 01:42:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1190/1519] eta 0:05:30 lr 0.000018 time 0.9330 (1.0052) model_time 0.9329 (1.0041) loss 0.9321 (0.8537) grad_norm 8.3533 (8.3872/1.8021) mem 68106MB [2022-12-20 01:42:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1200/1519] eta 0:05:20 lr 0.000018 time 0.9320 (1.0051) model_time 0.9319 (1.0041) loss 0.9279 (0.8538) grad_norm 8.2523 (8.3857/1.7975) mem 68106MB [2022-12-20 01:42:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1210/1519] eta 0:05:10 lr 0.000018 time 0.9292 (1.0054) model_time 0.9289 (1.0044) loss 0.9056 (0.8535) grad_norm 10.4414 (8.4137/1.8048) mem 68106MB [2022-12-20 01:42:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1220/1519] eta 0:05:00 lr 0.000018 time 0.9130 (1.0054) model_time 0.9129 (1.0043) loss 0.9880 (0.8533) grad_norm 8.1730 (8.4029/1.8058) mem 68106MB [2022-12-20 01:42:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1230/1519] eta 0:04:50 lr 0.000018 time 0.9206 (1.0054) model_time 0.9205 (1.0043) loss 0.8309 (0.8526) grad_norm 11.0109 (8.4245/1.8188) mem 68106MB [2022-12-20 01:42:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1240/1519] eta 0:04:40 lr 0.000018 time 0.9334 (1.0053) model_time 0.9333 (1.0043) loss 0.7136 (0.8525) grad_norm 7.7667 (8.4776/1.8711) mem 68106MB [2022-12-20 01:43:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1250/1519] eta 0:04:30 lr 0.000018 time 0.9276 (1.0054) model_time 0.9275 (1.0043) loss 1.1099 (0.8531) grad_norm 9.4927 (8.4695/1.8667) mem 68106MB [2022-12-20 01:43:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1260/1519] eta 0:04:20 lr 0.000018 time 0.9261 (1.0053) model_time 0.9259 (1.0043) loss 0.9002 (0.8524) grad_norm 8.1624 (8.4479/1.8643) mem 68106MB [2022-12-20 01:43:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1270/1519] eta 0:04:10 lr 0.000018 time 0.9319 (1.0053) model_time 0.9317 (1.0043) loss 0.9454 (0.8522) grad_norm 10.0960 (8.4294/1.8625) mem 68106MB [2022-12-20 01:43:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1280/1519] eta 0:04:00 lr 0.000018 time 0.9285 (1.0053) model_time 0.9283 (1.0043) loss 0.9426 (0.8524) grad_norm 8.5802 (8.4365/1.8551) mem 68106MB [2022-12-20 01:43:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1290/1519] eta 0:03:50 lr 0.000018 time 0.9201 (1.0053) model_time 0.9200 (1.0043) loss 1.0056 (0.8523) grad_norm 8.0751 (8.4350/1.8405) mem 68106MB [2022-12-20 01:43:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1300/1519] eta 0:03:40 lr 0.000018 time 0.9303 (1.0053) model_time 0.9301 (1.0042) loss 0.8397 (0.8527) grad_norm 9.0806 (8.4453/1.8273) mem 68106MB [2022-12-20 01:44:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1310/1519] eta 0:03:30 lr 0.000018 time 0.9291 (1.0052) model_time 0.9289 (1.0042) loss 0.8897 (0.8522) grad_norm 7.1201 (8.4666/1.8388) mem 68106MB [2022-12-20 01:44:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1320/1519] eta 0:03:20 lr 0.000018 time 0.9210 (1.0052) model_time 0.9209 (1.0042) loss 0.8570 (0.8524) grad_norm 7.6730 (8.4926/1.8443) mem 68106MB [2022-12-20 01:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1330/1519] eta 0:03:09 lr 0.000018 time 0.9220 (1.0052) model_time 0.9219 (1.0042) loss 1.1779 (0.8528) grad_norm 9.8529 (8.4968/1.8408) mem 68106MB [2022-12-20 01:44:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1340/1519] eta 0:02:59 lr 0.000018 time 0.9242 (1.0052) model_time 0.9241 (1.0042) loss 1.0580 (0.8530) grad_norm 7.8466 (8.5098/1.8418) mem 68106MB [2022-12-20 01:44:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1350/1519] eta 0:02:49 lr 0.000018 time 0.9218 (1.0051) model_time 0.9217 (1.0041) loss 1.2810 (0.8530) grad_norm 7.7713 (8.5378/1.8302) mem 68106MB [2022-12-20 01:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1360/1519] eta 0:02:39 lr 0.000018 time 0.9239 (1.0052) model_time 0.9237 (1.0042) loss 1.0556 (0.8527) grad_norm 7.1008 (8.5313/1.8127) mem 68106MB [2022-12-20 01:45:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1370/1519] eta 0:02:29 lr 0.000018 time 0.9172 (1.0051) model_time 0.9171 (1.0041) loss 0.8998 (0.8529) grad_norm 10.9892 (8.5489/1.8257) mem 68106MB [2022-12-20 01:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1380/1519] eta 0:02:19 lr 0.000018 time 0.9217 (1.0051) model_time 0.9215 (1.0041) loss 0.8006 (0.8527) grad_norm 7.9195 (8.5666/1.8243) mem 68106MB [2022-12-20 01:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1390/1519] eta 0:02:09 lr 0.000018 time 0.9242 (1.0051) model_time 0.9241 (1.0041) loss 0.8489 (0.8528) grad_norm 12.8347 (8.5935/1.8591) mem 68106MB [2022-12-20 01:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1400/1519] eta 0:01:59 lr 0.000018 time 0.9289 (1.0050) model_time 0.9286 (1.0040) loss 0.7030 (0.8537) grad_norm 8.3030 (8.6257/1.8860) mem 68106MB [2022-12-20 01:45:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1410/1519] eta 0:01:49 lr 0.000018 time 0.9219 (1.0050) model_time 0.9217 (1.0040) loss 0.7789 (0.8539) grad_norm 7.6882 (8.6434/1.8845) mem 68106MB [2022-12-20 01:45:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1420/1519] eta 0:01:39 lr 0.000018 time 0.9345 (1.0050) model_time 0.9343 (1.0041) loss 0.7484 (0.8541) grad_norm 10.9122 (8.6300/1.8444) mem 68106MB [2022-12-20 01:46:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1430/1519] eta 0:01:29 lr 0.000018 time 0.9388 (1.0050) model_time 0.9386 (1.0040) loss 0.7257 (0.8553) grad_norm 7.5524 (8.6244/1.8441) mem 68106MB [2022-12-20 01:46:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1440/1519] eta 0:01:19 lr 0.000018 time 0.9280 (1.0050) model_time 0.9279 (1.0041) loss 1.2278 (0.8552) grad_norm 6.6549 (8.6273/1.8406) mem 68106MB [2022-12-20 01:46:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1450/1519] eta 0:01:09 lr 0.000018 time 0.9739 (1.0050) model_time 0.9737 (1.0041) loss 0.6818 (0.8550) grad_norm 8.4599 (8.6251/1.8342) mem 68106MB [2022-12-20 01:46:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1460/1519] eta 0:00:59 lr 0.000018 time 0.9349 (1.0050) model_time 0.9346 (1.0041) loss 0.8667 (0.8552) grad_norm 7.2415 (8.6303/1.8335) mem 68106MB [2022-12-20 01:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1470/1519] eta 0:00:49 lr 0.000018 time 0.9574 (1.0051) model_time 0.9572 (1.0041) loss 1.2261 (0.8552) grad_norm 6.1690 (8.6461/1.8376) mem 68106MB [2022-12-20 01:46:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1480/1519] eta 0:00:39 lr 0.000018 time 0.9291 (1.0052) model_time 0.9290 (1.0042) loss 0.6911 (0.8551) grad_norm 9.3122 (8.6382/1.8383) mem 68106MB [2022-12-20 01:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1490/1519] eta 0:00:29 lr 0.000018 time 0.9275 (1.0051) model_time 0.9274 (1.0042) loss 0.6925 (0.8554) grad_norm 10.4749 (8.6412/1.8302) mem 68106MB [2022-12-20 01:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1500/1519] eta 0:00:19 lr 0.000018 time 0.9227 (1.0052) model_time 0.9225 (1.0043) loss 0.9216 (0.8555) grad_norm 8.7890 (8.6265/1.8067) mem 68106MB [2022-12-20 01:47:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [48/100][1510/1519] eta 0:00:09 lr 0.000018 time 0.9744 (1.0052) model_time 0.9743 (1.0043) loss 0.6723 (0.8555) grad_norm 8.3233 (8.6432/1.8313) mem 68106MB [2022-12-20 01:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 48 training takes 0:25:26 [2022-12-20 01:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_48.pth saving...... [2022-12-20 01:48:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_48.pth saved !!! [2022-12-20 01:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.685 (0.685) Loss 0.5070 (0.5070) Acc@1 92.014 (92.014) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 01:48:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.333) Loss 0.5161 (0.4936) Acc@1 91.667 (92.424) Acc@5 98.264 (98.611) Mem 68106MB [2022-12-20 01:48:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4579 (0.4918) Acc@1 92.014 (92.278) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 01:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.299 (0.310) Loss 0.6202 (0.4984) Acc@1 89.583 (92.003) Acc@5 97.917 (98.342) Mem 68106MB [2022-12-20 01:48:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.310) Loss 0.4624 (0.4895) Acc@1 92.708 (92.107) Acc@5 98.958 (98.442) Mem 68106MB [2022-12-20 01:48:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.308) Loss 0.4805 (0.4863) Acc@1 90.625 (92.102) Acc@5 99.653 (98.495) Mem 68106MB [2022-12-20 01:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.301 (0.307) Loss 0.5145 (0.4861) Acc@1 90.972 (92.099) Acc@5 98.264 (98.480) Mem 68106MB [2022-12-20 01:48:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.306) Loss 0.5244 (0.4870) Acc@1 92.014 (92.048) Acc@5 98.264 (98.464) Mem 68106MB [2022-12-20 01:48:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.305) Loss 0.4216 (0.4857) Acc@1 92.708 (92.065) Acc@5 98.611 (98.487) Mem 68106MB [2022-12-20 01:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:48] * Acc@1 92.031 Acc@5 98.494 [2022-12-20 01:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.0% [2022-12-20 01:48:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.10% [2022-12-20 01:48:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][0/1519] eta 0:46:31 lr 0.000018 time 1.8378 (1.8378) model_time 1.0398 (1.0398) loss 0.7649 (0.7649) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 01:48:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][10/1519] eta 0:27:14 lr 0.000018 time 0.9787 (1.0830) model_time 0.9786 (1.0101) loss 0.9859 (0.8484) grad_norm 12.6089 (10.5374/2.2906) mem 68106MB [2022-12-20 01:48:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][20/1519] eta 0:26:05 lr 0.000018 time 0.9313 (1.0445) model_time 0.9312 (1.0062) loss 0.7923 (0.8178) grad_norm 11.1887 (10.6652/2.3166) mem 68106MB [2022-12-20 01:48:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][30/1519] eta 0:25:34 lr 0.000018 time 0.9187 (1.0307) model_time 0.9186 (1.0047) loss 0.9030 (0.8407) grad_norm 7.6341 (9.6436/2.4297) mem 68106MB [2022-12-20 01:49:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][40/1519] eta 0:25:14 lr 0.000018 time 0.9265 (1.0242) model_time 0.9263 (1.0044) loss 0.7509 (0.8527) grad_norm 8.0253 (10.2390/3.3043) mem 68106MB [2022-12-20 01:49:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][50/1519] eta 0:24:58 lr 0.000018 time 0.9227 (1.0203) model_time 0.9225 (1.0043) loss 0.6953 (0.8546) grad_norm 6.9165 (9.5178/3.2940) mem 68106MB [2022-12-20 01:49:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][60/1519] eta 0:24:45 lr 0.000018 time 0.9212 (1.0179) model_time 0.9211 (1.0045) loss 0.6950 (0.8534) grad_norm 8.7744 (9.2696/3.0820) mem 68106MB [2022-12-20 01:49:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][70/1519] eta 0:24:32 lr 0.000018 time 0.9286 (1.0162) model_time 0.9284 (1.0047) loss 0.7262 (0.8581) grad_norm 6.4772 (9.0555/2.9237) mem 68106MB [2022-12-20 01:49:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][80/1519] eta 0:24:21 lr 0.000018 time 0.9940 (1.0154) model_time 0.9938 (1.0053) loss 1.0184 (0.8591) grad_norm 8.5172 (9.1008/2.8091) mem 68106MB [2022-12-20 01:49:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][90/1519] eta 0:24:09 lr 0.000018 time 0.9211 (1.0141) model_time 0.9210 (1.0050) loss 0.8188 (0.8535) grad_norm 7.6700 (9.0674/2.6627) mem 68106MB [2022-12-20 01:50:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][100/1519] eta 0:23:56 lr 0.000018 time 0.9212 (1.0126) model_time 0.9209 (1.0044) loss 0.7734 (0.8571) grad_norm 8.7528 (9.0811/2.5906) mem 68106MB [2022-12-20 01:50:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][110/1519] eta 0:23:44 lr 0.000018 time 0.9254 (1.0113) model_time 0.9253 (1.0039) loss 0.9468 (0.8586) grad_norm 8.3868 (8.9732/2.5379) mem 68106MB [2022-12-20 01:50:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][120/1519] eta 0:23:33 lr 0.000018 time 0.9248 (1.0102) model_time 0.9247 (1.0034) loss 0.9201 (0.8601) grad_norm 10.6758 (8.9546/2.5141) mem 68106MB [2022-12-20 01:50:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][130/1519] eta 0:23:23 lr 0.000018 time 0.9813 (1.0105) model_time 0.9811 (1.0042) loss 0.7306 (0.8604) grad_norm 10.1067 (8.9662/2.4460) mem 68106MB [2022-12-20 01:50:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][140/1519] eta 0:23:15 lr 0.000018 time 0.9167 (1.0117) model_time 0.9165 (1.0058) loss 0.8573 (0.8582) grad_norm 9.6651 (9.0253/2.4896) mem 68106MB [2022-12-20 01:50:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][150/1519] eta 0:23:03 lr 0.000018 time 0.9259 (1.0108) model_time 0.9258 (1.0053) loss 0.7083 (0.8539) grad_norm 12.9731 (9.0311/2.4671) mem 68106MB [2022-12-20 01:51:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][160/1519] eta 0:22:53 lr 0.000018 time 0.9252 (1.0108) model_time 0.9251 (1.0056) loss 0.8485 (0.8539) grad_norm 5.2677 (8.9805/2.4376) mem 68106MB [2022-12-20 01:51:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][170/1519] eta 0:22:42 lr 0.000018 time 0.9282 (1.0101) model_time 0.9280 (1.0051) loss 0.7603 (0.8505) grad_norm 9.0146 (8.9399/2.3895) mem 68106MB [2022-12-20 01:51:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][180/1519] eta 0:22:31 lr 0.000018 time 0.9250 (1.0094) model_time 0.9248 (1.0047) loss 0.7438 (0.8492) grad_norm 6.8238 (8.9047/2.3423) mem 68106MB [2022-12-20 01:51:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][190/1519] eta 0:22:20 lr 0.000018 time 0.9209 (1.0088) model_time 0.9208 (1.0043) loss 0.7671 (0.8534) grad_norm 9.1649 (8.8778/2.3052) mem 68106MB [2022-12-20 01:51:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][200/1519] eta 0:22:10 lr 0.000017 time 0.9213 (1.0084) model_time 0.9211 (1.0041) loss 0.9356 (0.8530) grad_norm 7.9818 (8.9184/2.3000) mem 68106MB [2022-12-20 01:52:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][210/1519] eta 0:22:00 lr 0.000017 time 0.9303 (1.0086) model_time 0.9301 (1.0046) loss 0.7936 (0.8569) grad_norm 11.3676 (8.9388/2.2602) mem 68106MB [2022-12-20 01:52:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][220/1519] eta 0:21:49 lr 0.000017 time 0.9283 (1.0081) model_time 0.9281 (1.0041) loss 0.7145 (0.8555) grad_norm 7.0267 (8.9068/2.2360) mem 68106MB [2022-12-20 01:52:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][230/1519] eta 0:21:39 lr 0.000017 time 0.9039 (1.0078) model_time 0.9038 (1.0040) loss 0.8298 (0.8580) grad_norm 7.9001 (8.8458/2.2119) mem 68106MB [2022-12-20 01:52:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][240/1519] eta 0:21:29 lr 0.000017 time 0.9193 (1.0079) model_time 0.9191 (1.0043) loss 0.8596 (0.8579) grad_norm 9.3026 (8.8027/2.1859) mem 68106MB [2022-12-20 01:52:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][250/1519] eta 0:21:19 lr 0.000017 time 0.9220 (1.0081) model_time 0.9219 (1.0047) loss 1.0211 (0.8568) grad_norm 7.5309 (8.7921/2.1469) mem 68106MB [2022-12-20 01:52:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][260/1519] eta 0:21:09 lr 0.000017 time 1.0380 (1.0082) model_time 1.0379 (1.0048) loss 0.9216 (0.8538) grad_norm 8.6464 (8.7668/2.1124) mem 68106MB [2022-12-20 01:53:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][270/1519] eta 0:20:58 lr 0.000017 time 0.9260 (1.0078) model_time 0.9259 (1.0045) loss 0.7846 (0.8519) grad_norm 8.4067 (8.7413/2.0861) mem 68106MB [2022-12-20 01:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][280/1519] eta 0:20:48 lr 0.000017 time 0.9184 (1.0076) model_time 0.9183 (1.0045) loss 0.7242 (0.8497) grad_norm 6.6447 (8.7093/2.0831) mem 68106MB [2022-12-20 01:53:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][290/1519] eta 0:20:38 lr 0.000017 time 0.9203 (1.0076) model_time 0.9202 (1.0045) loss 0.8326 (0.8524) grad_norm 9.4756 (8.7046/2.0768) mem 68106MB [2022-12-20 01:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][300/1519] eta 0:20:28 lr 0.000017 time 0.9241 (1.0075) model_time 0.9240 (1.0046) loss 1.0860 (0.8537) grad_norm 7.3199 (8.6851/2.0573) mem 68106MB [2022-12-20 01:53:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][310/1519] eta 0:20:17 lr 0.000017 time 0.9255 (1.0073) model_time 0.9254 (1.0044) loss 0.9652 (0.8529) grad_norm 7.4713 (8.6550/2.0358) mem 68106MB [2022-12-20 01:53:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][320/1519] eta 0:20:08 lr 0.000017 time 0.9256 (1.0080) model_time 0.9255 (1.0052) loss 0.9534 (0.8568) grad_norm 28.0659 (8.8074/2.5458) mem 68106MB [2022-12-20 01:54:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][330/1519] eta 0:19:58 lr 0.000017 time 0.9303 (1.0081) model_time 0.9302 (1.0054) loss 0.8584 (0.8571) grad_norm 9.1605 (8.7688/2.5368) mem 68106MB [2022-12-20 01:54:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][340/1519] eta 0:19:48 lr 0.000017 time 0.9205 (1.0080) model_time 0.9203 (1.0053) loss 0.6752 (0.8590) grad_norm 8.8815 (8.7667/2.5383) mem 68106MB [2022-12-20 01:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][350/1519] eta 0:19:38 lr 0.000017 time 0.9917 (1.0079) model_time 0.9916 (1.0053) loss 0.7455 (0.8580) grad_norm 10.4059 (8.7952/2.6164) mem 68106MB [2022-12-20 01:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][360/1519] eta 0:19:27 lr 0.000017 time 0.9298 (1.0077) model_time 0.9297 (1.0052) loss 0.8811 (0.8573) grad_norm 7.4177 (8.7927/2.5899) mem 68106MB [2022-12-20 01:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][370/1519] eta 0:19:17 lr 0.000017 time 0.9291 (1.0075) model_time 0.9289 (1.0051) loss 1.0341 (0.8581) grad_norm 7.6775 (8.7892/2.5741) mem 68106MB [2022-12-20 01:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][380/1519] eta 0:19:07 lr 0.000017 time 0.9684 (1.0075) model_time 0.9683 (1.0051) loss 0.6995 (0.8577) grad_norm 9.2255 (8.7742/2.5536) mem 68106MB [2022-12-20 01:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][390/1519] eta 0:18:57 lr 0.000017 time 0.9191 (1.0075) model_time 0.9190 (1.0051) loss 1.2025 (0.8582) grad_norm 5.9269 (8.7247/2.5409) mem 68106MB [2022-12-20 01:55:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][400/1519] eta 0:18:47 lr 0.000017 time 0.9218 (1.0074) model_time 0.9217 (1.0051) loss 0.7495 (0.8564) grad_norm 8.6971 (8.7376/2.5231) mem 68106MB [2022-12-20 01:55:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][410/1519] eta 0:18:37 lr 0.000017 time 0.9316 (1.0072) model_time 0.9315 (1.0050) loss 0.8696 (0.8558) grad_norm 7.0949 (8.7474/2.5096) mem 68106MB [2022-12-20 01:55:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][420/1519] eta 0:18:26 lr 0.000017 time 0.9263 (1.0071) model_time 0.9262 (1.0049) loss 0.8884 (0.8548) grad_norm 6.8121 (8.7047/2.4960) mem 68106MB [2022-12-20 01:55:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][430/1519] eta 0:18:16 lr 0.000017 time 0.9340 (1.0071) model_time 0.9339 (1.0049) loss 0.6850 (0.8544) grad_norm 10.8932 (8.6951/2.4761) mem 68106MB [2022-12-20 01:55:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][440/1519] eta 0:18:07 lr 0.000017 time 1.1797 (1.0074) model_time 1.1795 (1.0053) loss 0.9137 (0.8538) grad_norm 7.4019 (8.6619/2.4625) mem 68106MB [2022-12-20 01:56:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][450/1519] eta 0:17:57 lr 0.000017 time 0.8836 (1.0076) model_time 0.8835 (1.0056) loss 0.8296 (0.8544) grad_norm 10.5689 (8.6428/2.4518) mem 68106MB [2022-12-20 01:56:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][460/1519] eta 0:17:46 lr 0.000017 time 0.9311 (1.0074) model_time 0.9310 (1.0054) loss 0.7039 (0.8521) grad_norm 6.6276 (8.6337/2.4298) mem 68106MB [2022-12-20 01:56:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][470/1519] eta 0:17:36 lr 0.000017 time 0.9017 (1.0075) model_time 0.9015 (1.0055) loss 0.8850 (0.8523) grad_norm 5.9036 (8.6151/2.4201) mem 68106MB [2022-12-20 01:56:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][480/1519] eta 0:17:26 lr 0.000017 time 0.9236 (1.0073) model_time 0.9234 (1.0053) loss 0.7479 (0.8503) grad_norm 7.9000 (8.6139/2.4037) mem 68106MB [2022-12-20 01:56:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][490/1519] eta 0:17:16 lr 0.000017 time 0.9250 (1.0072) model_time 0.9248 (1.0052) loss 0.8147 (0.8488) grad_norm 10.5421 (8.6131/2.3922) mem 68106MB [2022-12-20 01:56:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][500/1519] eta 0:17:06 lr 0.000017 time 0.9252 (1.0073) model_time 0.9250 (1.0053) loss 0.9413 (0.8485) grad_norm 8.6955 (8.6238/2.3807) mem 68106MB [2022-12-20 01:57:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][510/1519] eta 0:16:56 lr 0.000017 time 0.9238 (1.0071) model_time 0.9237 (1.0053) loss 0.8283 (0.8486) grad_norm 8.8625 (8.6314/2.3725) mem 68106MB [2022-12-20 01:57:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][520/1519] eta 0:16:46 lr 0.000017 time 0.9245 (1.0070) model_time 0.9243 (1.0052) loss 0.7880 (0.8489) grad_norm 10.1741 (8.6236/2.3549) mem 68106MB [2022-12-20 01:57:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][530/1519] eta 0:16:35 lr 0.000017 time 0.9813 (1.0070) model_time 0.9812 (1.0052) loss 0.8754 (0.8492) grad_norm 11.1124 (8.6247/2.3470) mem 68106MB [2022-12-20 01:57:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][540/1519] eta 0:16:25 lr 0.000017 time 0.9248 (1.0068) model_time 0.9246 (1.0050) loss 1.0433 (0.8498) grad_norm 10.4690 (8.6333/2.3379) mem 68106MB [2022-12-20 01:57:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][550/1519] eta 0:16:15 lr 0.000017 time 0.9281 (1.0069) model_time 0.9280 (1.0051) loss 0.8000 (0.8492) grad_norm 7.5755 (8.6175/2.3294) mem 68106MB [2022-12-20 01:57:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][560/1519] eta 0:16:05 lr 0.000017 time 0.9243 (1.0070) model_time 0.9242 (1.0053) loss 0.6955 (0.8480) grad_norm 6.0712 (8.6199/2.3367) mem 68106MB [2022-12-20 01:58:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][570/1519] eta 0:15:55 lr 0.000017 time 0.9363 (1.0069) model_time 0.9362 (1.0052) loss 0.8470 (0.8470) grad_norm 10.6831 (8.6529/2.3383) mem 68106MB [2022-12-20 01:58:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][580/1519] eta 0:15:45 lr 0.000017 time 0.9299 (1.0069) model_time 0.9292 (1.0052) loss 0.8588 (0.8466) grad_norm 6.8574 (8.6351/2.3267) mem 68106MB [2022-12-20 01:58:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][590/1519] eta 0:15:35 lr 0.000017 time 0.9743 (1.0069) model_time 0.9741 (1.0052) loss 0.7163 (0.8450) grad_norm 9.9612 (8.6255/2.3144) mem 68106MB [2022-12-20 01:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][600/1519] eta 0:15:25 lr 0.000017 time 0.9195 (1.0067) model_time 0.9194 (1.0050) loss 0.8955 (0.8455) grad_norm 10.8503 (8.6219/2.3050) mem 68106MB [2022-12-20 01:58:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][610/1519] eta 0:15:15 lr 0.000017 time 1.1271 (1.0068) model_time 1.1268 (1.0052) loss 0.6698 (0.8445) grad_norm 7.8974 (8.5971/2.2886) mem 68106MB [2022-12-20 01:58:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][620/1519] eta 0:15:05 lr 0.000017 time 0.9163 (1.0067) model_time 0.9162 (1.0051) loss 0.6655 (0.8435) grad_norm 5.3757 (8.5506/2.2584) mem 68106MB [2022-12-20 01:59:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][630/1519] eta 0:14:55 lr 0.000017 time 1.0267 (1.0068) model_time 1.0265 (1.0052) loss 1.2197 (0.8440) grad_norm 8.7220 (8.5501/2.2604) mem 68106MB [2022-12-20 01:59:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][640/1519] eta 0:14:44 lr 0.000017 time 0.9342 (1.0067) model_time 0.9339 (1.0051) loss 0.7069 (0.8435) grad_norm 7.4844 (8.5005/2.1374) mem 68106MB [2022-12-20 01:59:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][650/1519] eta 0:14:34 lr 0.000017 time 0.9208 (1.0065) model_time 0.9206 (1.0050) loss 0.6639 (0.8432) grad_norm 6.9209 (8.5197/2.1291) mem 68106MB [2022-12-20 01:59:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][660/1519] eta 0:14:24 lr 0.000017 time 0.9272 (1.0064) model_time 0.9271 (1.0049) loss 0.8707 (0.8431) grad_norm 6.7635 (8.5119/2.1305) mem 68106MB [2022-12-20 01:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][670/1519] eta 0:14:14 lr 0.000017 time 0.9154 (1.0063) model_time 0.9153 (1.0048) loss 0.6876 (0.8425) grad_norm 8.5669 (8.5347/2.1274) mem 68106MB [2022-12-20 01:59:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][680/1519] eta 0:14:04 lr 0.000017 time 0.9277 (1.0062) model_time 0.9275 (1.0047) loss 0.6719 (0.8416) grad_norm 7.2742 (8.5132/2.1167) mem 68106MB [2022-12-20 02:00:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][690/1519] eta 0:13:54 lr 0.000017 time 0.9273 (1.0062) model_time 0.9272 (1.0047) loss 0.7546 (0.8418) grad_norm 8.5458 (8.4973/2.1182) mem 68106MB [2022-12-20 02:00:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][700/1519] eta 0:13:44 lr 0.000017 time 0.9070 (1.0062) model_time 0.9068 (1.0048) loss 0.8975 (0.8419) grad_norm 12.7164 (8.5019/2.1244) mem 68106MB [2022-12-20 02:00:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][710/1519] eta 0:13:33 lr 0.000017 time 0.9617 (1.0062) model_time 0.9616 (1.0047) loss 0.6768 (0.8411) grad_norm 10.0546 (8.5147/2.1209) mem 68106MB [2022-12-20 02:00:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][720/1519] eta 0:13:23 lr 0.000017 time 0.9273 (1.0061) model_time 0.9272 (1.0046) loss 0.7647 (0.8409) grad_norm 7.0350 (8.5048/2.1042) mem 68106MB [2022-12-20 02:00:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][730/1519] eta 0:13:13 lr 0.000017 time 0.9218 (1.0062) model_time 0.9216 (1.0048) loss 0.7060 (0.8403) grad_norm 9.1765 (8.4996/2.1091) mem 68106MB [2022-12-20 02:00:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][740/1519] eta 0:13:03 lr 0.000017 time 0.9287 (1.0061) model_time 0.9286 (1.0047) loss 0.6752 (0.8408) grad_norm 13.2368 (8.4937/2.0979) mem 68106MB [2022-12-20 02:01:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][750/1519] eta 0:12:53 lr 0.000017 time 0.9372 (1.0060) model_time 0.9370 (1.0046) loss 0.7399 (0.8406) grad_norm 12.8483 (8.5085/2.0990) mem 68106MB [2022-12-20 02:01:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][760/1519] eta 0:12:43 lr 0.000017 time 0.9256 (1.0060) model_time 0.9253 (1.0046) loss 0.8457 (0.8397) grad_norm 7.3922 (8.5111/2.0934) mem 68106MB [2022-12-20 02:01:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][770/1519] eta 0:12:33 lr 0.000017 time 0.9652 (1.0060) model_time 0.9651 (1.0046) loss 0.9070 (0.8398) grad_norm 5.8695 (8.4973/2.0982) mem 68106MB [2022-12-20 02:01:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][780/1519] eta 0:12:23 lr 0.000017 time 0.9097 (1.0060) model_time 0.9095 (1.0047) loss 0.7846 (0.8406) grad_norm 7.0285 (8.5022/2.1049) mem 68106MB [2022-12-20 02:01:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][790/1519] eta 0:12:13 lr 0.000017 time 0.9185 (1.0060) model_time 0.9184 (1.0047) loss 0.7635 (0.8413) grad_norm 8.6504 (8.5395/2.1546) mem 68106MB [2022-12-20 02:01:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][800/1519] eta 0:12:03 lr 0.000017 time 0.9329 (1.0059) model_time 0.9328 (1.0046) loss 0.7617 (0.8429) grad_norm 11.1583 (8.5461/2.1454) mem 68106MB [2022-12-20 02:02:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][810/1519] eta 0:11:53 lr 0.000017 time 0.9267 (1.0060) model_time 0.9265 (1.0047) loss 0.7046 (0.8423) grad_norm 6.9496 (8.5199/2.1416) mem 68106MB [2022-12-20 02:02:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][820/1519] eta 0:11:43 lr 0.000017 time 0.9281 (1.0059) model_time 0.9279 (1.0046) loss 0.6749 (0.8422) grad_norm 10.9748 (8.5340/2.1474) mem 68106MB [2022-12-20 02:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][830/1519] eta 0:11:33 lr 0.000017 time 0.9199 (1.0059) model_time 0.9198 (1.0046) loss 0.8679 (0.8424) grad_norm 6.4509 (8.5247/2.1516) mem 68106MB [2022-12-20 02:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][840/1519] eta 0:11:22 lr 0.000017 time 0.9200 (1.0059) model_time 0.9199 (1.0046) loss 0.7093 (0.8419) grad_norm 7.8259 (8.5229/2.1492) mem 68106MB [2022-12-20 02:02:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][850/1519] eta 0:11:12 lr 0.000017 time 0.9345 (1.0059) model_time 0.9343 (1.0046) loss 1.1059 (0.8422) grad_norm 8.1934 (8.5281/2.1729) mem 68106MB [2022-12-20 02:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][860/1519] eta 0:11:03 lr 0.000017 time 0.9336 (1.0061) model_time 0.9333 (1.0048) loss 0.9548 (0.8434) grad_norm 13.4768 (8.5618/2.1978) mem 68106MB [2022-12-20 02:03:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][870/1519] eta 0:10:53 lr 0.000017 time 0.9257 (1.0062) model_time 0.9255 (1.0049) loss 0.8045 (0.8430) grad_norm 9.0796 (8.5691/2.2015) mem 68106MB [2022-12-20 02:03:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][880/1519] eta 0:10:42 lr 0.000017 time 0.9265 (1.0062) model_time 0.9263 (1.0049) loss 0.8445 (0.8428) grad_norm 6.1603 (8.5963/2.2182) mem 68106MB [2022-12-20 02:03:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][890/1519] eta 0:10:32 lr 0.000017 time 0.9166 (1.0061) model_time 0.9165 (1.0049) loss 0.6910 (0.8427) grad_norm 10.3005 (8.6115/2.2438) mem 68106MB [2022-12-20 02:03:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][900/1519] eta 0:10:22 lr 0.000017 time 0.9448 (1.0061) model_time 0.9446 (1.0049) loss 0.6738 (0.8434) grad_norm 7.5528 (8.6229/2.2451) mem 68106MB [2022-12-20 02:03:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][910/1519] eta 0:10:12 lr 0.000017 time 0.9237 (1.0061) model_time 0.9236 (1.0048) loss 0.7816 (0.8424) grad_norm 7.1045 (8.6573/2.2766) mem 68106MB [2022-12-20 02:03:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][920/1519] eta 0:10:02 lr 0.000017 time 0.9212 (1.0060) model_time 0.9211 (1.0048) loss 0.9291 (0.8428) grad_norm 14.3447 (8.5804/1.9961) mem 68106MB [2022-12-20 02:04:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][930/1519] eta 0:09:52 lr 0.000017 time 0.9317 (1.0060) model_time 0.9316 (1.0048) loss 0.8192 (0.8434) grad_norm 9.0074 (8.6012/1.9799) mem 68106MB [2022-12-20 02:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][940/1519] eta 0:09:42 lr 0.000017 time 0.9319 (1.0061) model_time 0.9317 (1.0049) loss 0.6875 (0.8438) grad_norm 7.0956 (8.6009/1.9670) mem 68106MB [2022-12-20 02:04:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][950/1519] eta 0:09:32 lr 0.000017 time 0.9259 (1.0061) model_time 0.9258 (1.0050) loss 0.7480 (0.8438) grad_norm 9.3728 (8.5808/1.8816) mem 68106MB [2022-12-20 02:04:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][960/1519] eta 0:09:22 lr 0.000017 time 0.9362 (1.0061) model_time 0.9360 (1.0050) loss 0.8617 (0.8434) grad_norm 11.6121 (8.5813/1.8993) mem 68106MB [2022-12-20 02:04:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][970/1519] eta 0:09:12 lr 0.000017 time 0.9292 (1.0061) model_time 0.9291 (1.0049) loss 0.6776 (0.8434) grad_norm 7.3391 (8.5916/1.9088) mem 68106MB [2022-12-20 02:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][980/1519] eta 0:09:02 lr 0.000017 time 0.9266 (1.0060) model_time 0.9265 (1.0048) loss 0.7817 (0.8428) grad_norm 10.7345 (8.5997/1.9187) mem 68106MB [2022-12-20 02:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][990/1519] eta 0:08:52 lr 0.000017 time 0.9333 (1.0060) model_time 0.9332 (1.0048) loss 0.9644 (0.8425) grad_norm 8.5311 (8.6378/1.9121) mem 68106MB [2022-12-20 02:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1000/1519] eta 0:08:42 lr 0.000017 time 0.9270 (1.0059) model_time 0.9268 (1.0048) loss 0.8648 (0.8422) grad_norm 8.4925 (8.6100/1.9134) mem 68106MB [2022-12-20 02:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1010/1519] eta 0:08:32 lr 0.000017 time 0.9193 (1.0059) model_time 0.9191 (1.0048) loss 0.7033 (0.8422) grad_norm 8.1817 (8.6021/1.9090) mem 68106MB [2022-12-20 02:05:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1020/1519] eta 0:08:21 lr 0.000017 time 0.9195 (1.0059) model_time 0.9193 (1.0047) loss 1.3420 (0.8424) grad_norm 9.5052 (8.6163/1.9034) mem 68106MB [2022-12-20 02:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1030/1519] eta 0:08:11 lr 0.000017 time 0.9276 (1.0058) model_time 0.9275 (1.0047) loss 0.6931 (0.8418) grad_norm 8.4880 (8.5928/1.9142) mem 68106MB [2022-12-20 02:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1040/1519] eta 0:08:01 lr 0.000017 time 0.9864 (1.0058) model_time 0.9863 (1.0048) loss 0.7754 (0.8418) grad_norm 9.1087 (8.6099/1.9074) mem 68106MB [2022-12-20 02:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1050/1519] eta 0:07:51 lr 0.000017 time 0.9238 (1.0059) model_time 0.9237 (1.0048) loss 0.7396 (0.8412) grad_norm 8.2072 (8.6324/1.9025) mem 68106MB [2022-12-20 02:06:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1060/1519] eta 0:07:41 lr 0.000017 time 0.9235 (1.0058) model_time 0.9234 (1.0047) loss 0.7710 (0.8413) grad_norm 8.2468 (8.6337/1.9015) mem 68106MB [2022-12-20 02:06:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1070/1519] eta 0:07:31 lr 0.000017 time 0.9776 (1.0058) model_time 0.9775 (1.0047) loss 0.7854 (0.8414) grad_norm 6.9333 (8.6381/1.8892) mem 68106MB [2022-12-20 02:06:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1080/1519] eta 0:07:21 lr 0.000017 time 0.9325 (1.0057) model_time 0.9324 (1.0047) loss 0.7482 (0.8415) grad_norm 9.7555 (8.6517/1.9046) mem 68106MB [2022-12-20 02:06:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1090/1519] eta 0:07:11 lr 0.000017 time 0.9274 (1.0057) model_time 0.9273 (1.0047) loss 0.7380 (0.8410) grad_norm 15.2180 (8.6813/1.9332) mem 68106MB [2022-12-20 02:06:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1100/1519] eta 0:07:01 lr 0.000017 time 0.9287 (1.0060) model_time 0.9286 (1.0049) loss 0.8631 (0.8419) grad_norm 8.1025 (8.6630/1.9253) mem 68106MB [2022-12-20 02:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1110/1519] eta 0:06:51 lr 0.000017 time 0.9408 (1.0061) model_time 0.9406 (1.0050) loss 1.1720 (0.8423) grad_norm 11.2354 (8.6989/1.9625) mem 68106MB [2022-12-20 02:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1120/1519] eta 0:06:41 lr 0.000017 time 0.9197 (1.0063) model_time 0.9196 (1.0053) loss 1.0538 (0.8430) grad_norm 5.9909 (8.6814/1.9698) mem 68106MB [2022-12-20 02:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1130/1519] eta 0:06:31 lr 0.000017 time 0.9270 (1.0063) model_time 0.9269 (1.0053) loss 0.7775 (0.8437) grad_norm 7.3492 (8.6768/1.9602) mem 68106MB [2022-12-20 02:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1140/1519] eta 0:06:21 lr 0.000017 time 0.9260 (1.0062) model_time 0.9258 (1.0052) loss 0.9947 (0.8443) grad_norm 7.0875 (8.6597/1.9501) mem 68106MB [2022-12-20 02:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1150/1519] eta 0:06:11 lr 0.000017 time 0.9318 (1.0062) model_time 0.9316 (1.0052) loss 0.7046 (0.8442) grad_norm 8.3167 (8.6845/1.9398) mem 68106MB [2022-12-20 02:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1160/1519] eta 0:06:01 lr 0.000017 time 0.9164 (1.0062) model_time 0.9162 (1.0051) loss 0.8384 (0.8444) grad_norm 6.4470 (8.6598/1.9187) mem 68106MB [2022-12-20 02:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1170/1519] eta 0:05:51 lr 0.000017 time 0.9319 (1.0061) model_time 0.9317 (1.0051) loss 0.7949 (0.8444) grad_norm 8.0141 (8.6141/1.8978) mem 68106MB [2022-12-20 02:08:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1180/1519] eta 0:05:41 lr 0.000017 time 0.9128 (1.0061) model_time 0.9127 (1.0051) loss 0.7089 (0.8434) grad_norm 9.0244 (8.6211/1.9005) mem 68106MB [2022-12-20 02:08:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1190/1519] eta 0:05:30 lr 0.000017 time 0.9286 (1.0061) model_time 0.9284 (1.0051) loss 0.6879 (0.8428) grad_norm 6.2940 (8.6192/1.9265) mem 68106MB [2022-12-20 02:08:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1200/1519] eta 0:05:20 lr 0.000017 time 0.9421 (1.0061) model_time 0.9419 (1.0051) loss 0.8546 (0.8425) grad_norm 10.2890 (8.6323/1.9222) mem 68106MB [2022-12-20 02:08:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1210/1519] eta 0:05:10 lr 0.000017 time 0.9305 (1.0061) model_time 0.9303 (1.0051) loss 0.7049 (0.8424) grad_norm 8.0026 (8.5952/1.9237) mem 68106MB [2022-12-20 02:08:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1220/1519] eta 0:05:00 lr 0.000017 time 0.9399 (1.0060) model_time 0.9398 (1.0051) loss 0.6854 (0.8423) grad_norm 8.4310 (8.5870/1.9228) mem 68106MB [2022-12-20 02:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1230/1519] eta 0:04:50 lr 0.000017 time 0.9333 (1.0060) model_time 0.9331 (1.0050) loss 0.7027 (0.8423) grad_norm 7.1452 (8.5805/1.9300) mem 68106MB [2022-12-20 02:09:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1240/1519] eta 0:04:40 lr 0.000017 time 0.9373 (1.0059) model_time 0.9372 (1.0049) loss 0.7791 (0.8420) grad_norm 7.4157 (8.5711/1.9327) mem 68106MB [2022-12-20 02:09:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1250/1519] eta 0:04:30 lr 0.000017 time 0.9823 (1.0060) model_time 0.9821 (1.0050) loss 0.9921 (0.8419) grad_norm 8.4500 (8.5979/1.9456) mem 68106MB [2022-12-20 02:09:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1260/1519] eta 0:04:20 lr 0.000017 time 0.9288 (1.0060) model_time 0.9286 (1.0050) loss 1.1074 (0.8418) grad_norm 6.5966 (8.6100/1.9444) mem 68106MB [2022-12-20 02:09:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1270/1519] eta 0:04:10 lr 0.000017 time 0.9230 (1.0060) model_time 0.9229 (1.0050) loss 0.8979 (0.8415) grad_norm 8.8959 (8.6082/1.9470) mem 68106MB [2022-12-20 02:09:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1280/1519] eta 0:04:00 lr 0.000017 time 0.9199 (1.0060) model_time 0.9197 (1.0050) loss 0.6832 (0.8409) grad_norm 7.2175 (8.6094/1.9503) mem 68106MB [2022-12-20 02:10:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1290/1519] eta 0:03:50 lr 0.000017 time 0.9329 (1.0059) model_time 0.9327 (1.0049) loss 0.7688 (0.8406) grad_norm 8.9833 (8.6221/1.9502) mem 68106MB [2022-12-20 02:10:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1300/1519] eta 0:03:40 lr 0.000017 time 0.9193 (1.0059) model_time 0.9192 (1.0049) loss 0.6949 (0.8398) grad_norm 10.0795 (8.6129/1.9344) mem 68106MB [2022-12-20 02:10:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1310/1519] eta 0:03:30 lr 0.000017 time 0.9139 (1.0058) model_time 0.9138 (1.0049) loss 1.0040 (0.8405) grad_norm 9.4754 (8.6103/1.9316) mem 68106MB [2022-12-20 02:10:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1320/1519] eta 0:03:20 lr 0.000017 time 0.9310 (1.0058) model_time 0.9308 (1.0048) loss 0.7018 (0.8404) grad_norm 7.6266 (8.6032/1.9376) mem 68106MB [2022-12-20 02:10:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1330/1519] eta 0:03:10 lr 0.000017 time 0.9387 (1.0058) model_time 0.9386 (1.0049) loss 0.7911 (0.8412) grad_norm 5.6800 (8.5990/1.9494) mem 68106MB [2022-12-20 02:10:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1340/1519] eta 0:03:00 lr 0.000017 time 0.9218 (1.0057) model_time 0.9217 (1.0048) loss 0.9422 (0.8415) grad_norm 7.8527 (8.5854/1.9293) mem 68106MB [2022-12-20 02:11:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1350/1519] eta 0:02:49 lr 0.000017 time 0.9302 (1.0057) model_time 0.9300 (1.0048) loss 0.7414 (0.8413) grad_norm 8.5768 (8.5484/1.9179) mem 68106MB [2022-12-20 02:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1360/1519] eta 0:02:39 lr 0.000017 time 0.9285 (1.0059) model_time 0.9283 (1.0050) loss 0.8405 (0.8409) grad_norm 11.8633 (8.5577/1.9272) mem 68106MB [2022-12-20 02:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1370/1519] eta 0:02:29 lr 0.000017 time 0.9189 (1.0058) model_time 0.9188 (1.0049) loss 0.7659 (0.8408) grad_norm 9.4104 (8.5829/1.9190) mem 68106MB [2022-12-20 02:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1380/1519] eta 0:02:19 lr 0.000017 time 0.9212 (1.0058) model_time 0.9211 (1.0049) loss 0.9335 (0.8406) grad_norm 7.6268 (8.5865/1.9100) mem 68106MB [2022-12-20 02:11:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1390/1519] eta 0:02:09 lr 0.000017 time 0.9304 (1.0058) model_time 0.9303 (1.0048) loss 0.8216 (0.8405) grad_norm 7.4612 (8.5583/1.8588) mem 68106MB [2022-12-20 02:11:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1400/1519] eta 0:01:59 lr 0.000017 time 0.9220 (1.0057) model_time 0.9219 (1.0048) loss 0.7227 (0.8408) grad_norm 5.6627 (8.5158/1.8554) mem 68106MB [2022-12-20 02:12:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1410/1519] eta 0:01:49 lr 0.000017 time 0.9273 (1.0057) model_time 0.9272 (1.0048) loss 0.8437 (0.8415) grad_norm 8.3857 (8.5423/1.8664) mem 68106MB [2022-12-20 02:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1420/1519] eta 0:01:39 lr 0.000017 time 0.9305 (1.0057) model_time 0.9304 (1.0048) loss 0.8814 (0.8416) grad_norm 9.4546 (8.5439/1.8526) mem 68106MB [2022-12-20 02:12:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1430/1519] eta 0:01:29 lr 0.000017 time 1.0491 (1.0058) model_time 1.0489 (1.0049) loss 0.8968 (0.8415) grad_norm 7.4712 (8.5738/1.8464) mem 68106MB [2022-12-20 02:12:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1440/1519] eta 0:01:19 lr 0.000017 time 0.9256 (1.0057) model_time 0.9254 (1.0048) loss 0.7528 (0.8413) grad_norm 7.1231 (8.5784/1.8446) mem 68106MB [2022-12-20 02:12:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1450/1519] eta 0:01:09 lr 0.000017 time 0.9323 (1.0057) model_time 0.9321 (1.0048) loss 0.7963 (0.8410) grad_norm 8.9652 (8.5760/1.8192) mem 68106MB [2022-12-20 02:12:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1460/1519] eta 0:00:59 lr 0.000017 time 0.9297 (1.0057) model_time 0.9293 (1.0049) loss 0.6795 (0.8407) grad_norm 7.4943 (8.5509/1.7934) mem 68106MB [2022-12-20 02:13:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1470/1519] eta 0:00:49 lr 0.000017 time 0.9236 (1.0057) model_time 0.9234 (1.0048) loss 0.6868 (0.8404) grad_norm 8.0763 (8.5633/1.8108) mem 68106MB [2022-12-20 02:13:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1480/1519] eta 0:00:39 lr 0.000017 time 0.9311 (1.0057) model_time 0.9309 (1.0048) loss 0.9474 (0.8406) grad_norm 11.9836 (8.5604/1.7973) mem 68106MB [2022-12-20 02:13:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1490/1519] eta 0:00:29 lr 0.000017 time 0.9253 (1.0056) model_time 0.9251 (1.0047) loss 0.7219 (0.8409) grad_norm 11.2101 (8.5462/1.7619) mem 68106MB [2022-12-20 02:13:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1500/1519] eta 0:00:19 lr 0.000017 time 0.9273 (1.0056) model_time 0.9271 (1.0047) loss 0.8949 (0.8409) grad_norm 12.3575 (8.5401/1.7717) mem 68106MB [2022-12-20 02:13:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [49/100][1510/1519] eta 0:00:09 lr 0.000017 time 0.9219 (1.0056) model_time 0.9218 (1.0047) loss 0.7163 (0.8407) grad_norm 6.1441 (8.4962/1.7519) mem 68106MB [2022-12-20 02:13:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 49 training takes 0:25:27 [2022-12-20 02:13:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_49.pth saving...... [2022-12-20 02:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_49.pth saved !!! [2022-12-20 02:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.641 (0.641) Loss 0.5114 (0.5114) Acc@1 91.319 (91.319) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 02:14:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.328) Loss 0.5270 (0.4916) Acc@1 92.014 (92.519) Acc@5 98.264 (98.643) Mem 68106MB [2022-12-20 02:14:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.313) Loss 0.4486 (0.4876) Acc@1 92.361 (92.361) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-20 02:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.305 (0.308) Loss 0.6286 (0.4961) Acc@1 89.931 (92.104) Acc@5 96.875 (98.354) Mem 68106MB [2022-12-20 02:14:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.307) Loss 0.4569 (0.4868) Acc@1 93.056 (92.285) Acc@5 98.611 (98.442) Mem 68106MB [2022-12-20 02:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.305) Loss 0.4663 (0.4843) Acc@1 90.972 (92.252) Acc@5 99.306 (98.489) Mem 68106MB [2022-12-20 02:14:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.304) Loss 0.4901 (0.4835) Acc@1 91.319 (92.236) Acc@5 98.264 (98.446) Mem 68106MB [2022-12-20 02:14:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5482 (0.4849) Acc@1 91.319 (92.180) Acc@5 98.264 (98.455) Mem 68106MB [2022-12-20 02:14:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4307 (0.4831) Acc@1 92.361 (92.168) Acc@5 98.264 (98.487) Mem 68106MB [2022-12-20 02:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:49] * Acc@1 92.166 Acc@5 98.490 [2022-12-20 02:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.2% [2022-12-20 02:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 02:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 02:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.17% [2022-12-20 02:15:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][0/1519] eta 0:35:38 lr 0.000017 time 1.4075 (1.4075) model_time 0.9594 (0.9594) loss 0.8155 (0.8155) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 02:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][10/1519] eta 0:26:05 lr 0.000017 time 0.9257 (1.0376) model_time 0.9255 (0.9965) loss 0.7679 (0.7873) grad_norm 7.0175 (7.7600/1.2196) mem 68106MB [2022-12-20 02:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][20/1519] eta 0:25:33 lr 0.000017 time 0.9826 (1.0229) model_time 0.9825 (1.0013) loss 1.1018 (0.8012) grad_norm 8.7714 (8.4001/1.2579) mem 68106MB [2022-12-20 02:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][30/1519] eta 0:25:12 lr 0.000017 time 0.9329 (1.0158) model_time 0.9328 (1.0010) loss 1.4395 (0.8128) grad_norm 6.5076 (8.5756/1.4099) mem 68106MB [2022-12-20 02:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][40/1519] eta 0:24:59 lr 0.000017 time 0.9274 (1.0140) model_time 0.9273 (1.0028) loss 0.7520 (0.8168) grad_norm 10.3665 (8.6543/1.2985) mem 68106MB [2022-12-20 02:16:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][50/1519] eta 0:24:50 lr 0.000017 time 0.9238 (1.0145) model_time 0.9237 (1.0054) loss 1.3878 (0.8468) grad_norm 8.0873 (8.5650/1.2207) mem 68106MB [2022-12-20 02:16:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][60/1519] eta 0:24:38 lr 0.000017 time 0.9223 (1.0135) model_time 0.9221 (1.0058) loss 0.8135 (0.8436) grad_norm 7.9464 (8.6210/1.2090) mem 68106MB [2022-12-20 02:16:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][70/1519] eta 0:24:31 lr 0.000017 time 0.9226 (1.0158) model_time 0.9225 (1.0091) loss 0.7399 (0.8351) grad_norm 7.0196 (8.4029/1.2720) mem 68106MB [2022-12-20 02:16:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][80/1519] eta 0:24:19 lr 0.000017 time 1.0034 (1.0144) model_time 1.0032 (1.0086) loss 0.6928 (0.8369) grad_norm 8.4532 (8.3148/1.2472) mem 68106MB [2022-12-20 02:16:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][90/1519] eta 0:24:08 lr 0.000017 time 0.9263 (1.0136) model_time 0.9262 (1.0084) loss 0.9632 (0.8339) grad_norm 8.7506 (8.2707/1.2180) mem 68106MB [2022-12-20 02:16:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][100/1519] eta 0:23:56 lr 0.000017 time 0.9339 (1.0124) model_time 0.9338 (1.0076) loss 0.9223 (0.8341) grad_norm 8.5187 (8.5077/1.6523) mem 68106MB [2022-12-20 02:17:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][110/1519] eta 0:23:45 lr 0.000017 time 0.9389 (1.0118) model_time 0.9387 (1.0074) loss 1.1534 (0.8364) grad_norm 6.6191 (8.4937/1.6085) mem 68106MB [2022-12-20 02:17:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][120/1519] eta 0:23:35 lr 0.000017 time 0.9246 (1.0118) model_time 0.9244 (1.0077) loss 0.8084 (0.8309) grad_norm 9.8965 (8.5729/1.9179) mem 68106MB [2022-12-20 02:17:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][130/1519] eta 0:23:24 lr 0.000017 time 0.9345 (1.0113) model_time 0.9344 (1.0076) loss 0.7155 (0.8309) grad_norm 12.1787 (8.5873/1.9093) mem 68106MB [2022-12-20 02:17:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][140/1519] eta 0:23:13 lr 0.000017 time 0.9257 (1.0106) model_time 0.9254 (1.0071) loss 1.1294 (0.8388) grad_norm 6.7542 (8.5811/1.8697) mem 68106MB [2022-12-20 02:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][150/1519] eta 0:23:03 lr 0.000017 time 0.9919 (1.0106) model_time 0.9917 (1.0073) loss 0.9881 (0.8397) grad_norm 9.5408 (8.6355/1.8688) mem 68106MB [2022-12-20 02:17:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][160/1519] eta 0:22:53 lr 0.000017 time 0.9267 (1.0108) model_time 0.9265 (1.0077) loss 1.0217 (0.8394) grad_norm 9.4922 (8.6153/1.8299) mem 68106MB [2022-12-20 02:18:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][170/1519] eta 0:22:43 lr 0.000017 time 0.9169 (1.0104) model_time 0.9168 (1.0075) loss 1.1310 (0.8429) grad_norm 10.1782 (8.5948/1.7958) mem 68106MB [2022-12-20 02:18:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][180/1519] eta 0:22:32 lr 0.000017 time 0.9200 (1.0103) model_time 0.9198 (1.0075) loss 0.9180 (0.8425) grad_norm 5.9499 (8.5365/1.7777) mem 68106MB [2022-12-20 02:18:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][190/1519] eta 0:22:22 lr 0.000017 time 0.8994 (1.0099) model_time 0.8992 (1.0073) loss 0.8318 (0.8399) grad_norm 8.2261 (8.5163/1.7390) mem 68106MB [2022-12-20 02:18:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][200/1519] eta 0:22:11 lr 0.000017 time 0.9338 (1.0094) model_time 0.9337 (1.0068) loss 1.0216 (0.8386) grad_norm 8.4476 (8.5080/1.7167) mem 68106MB [2022-12-20 02:18:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][210/1519] eta 0:22:00 lr 0.000017 time 0.9307 (1.0089) model_time 0.9305 (1.0064) loss 0.7106 (0.8406) grad_norm 9.3825 (8.5243/1.6998) mem 68106MB [2022-12-20 02:18:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][220/1519] eta 0:21:50 lr 0.000017 time 0.9180 (1.0085) model_time 0.9178 (1.0061) loss 0.9008 (0.8413) grad_norm 7.7350 (8.5058/1.6720) mem 68106MB [2022-12-20 02:19:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][230/1519] eta 0:21:39 lr 0.000017 time 0.9314 (1.0085) model_time 0.9313 (1.0063) loss 0.9621 (0.8422) grad_norm 9.1359 (8.4905/1.6459) mem 68106MB [2022-12-20 02:19:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][240/1519] eta 0:21:29 lr 0.000017 time 0.9246 (1.0084) model_time 0.9245 (1.0062) loss 0.8076 (0.8453) grad_norm 7.4417 (8.4607/1.6333) mem 68106MB [2022-12-20 02:19:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][250/1519] eta 0:21:19 lr 0.000017 time 0.9255 (1.0083) model_time 0.9254 (1.0062) loss 0.8541 (0.8455) grad_norm 8.3405 (8.4464/1.6149) mem 68106MB [2022-12-20 02:19:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][260/1519] eta 0:21:09 lr 0.000017 time 0.9284 (1.0080) model_time 0.9282 (1.0059) loss 0.8154 (0.8486) grad_norm 8.7376 (8.5179/1.6467) mem 68106MB [2022-12-20 02:19:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][270/1519] eta 0:20:58 lr 0.000017 time 0.9301 (1.0077) model_time 0.9299 (1.0057) loss 0.6681 (0.8439) grad_norm 6.8299 (8.5218/1.6649) mem 68106MB [2022-12-20 02:19:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][280/1519] eta 0:20:48 lr 0.000017 time 0.9312 (1.0076) model_time 0.9311 (1.0057) loss 0.8890 (0.8453) grad_norm 6.3408 (8.4997/1.6668) mem 68106MB [2022-12-20 02:20:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][290/1519] eta 0:20:38 lr 0.000017 time 0.9308 (1.0074) model_time 0.9306 (1.0055) loss 0.7108 (0.8430) grad_norm 10.6281 (8.5509/1.7338) mem 68106MB [2022-12-20 02:20:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][300/1519] eta 0:20:28 lr 0.000017 time 0.9212 (1.0076) model_time 0.9210 (1.0058) loss 0.8345 (0.8435) grad_norm 10.2096 (8.5944/1.7895) mem 68106MB [2022-12-20 02:20:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][310/1519] eta 0:20:18 lr 0.000017 time 0.9279 (1.0075) model_time 0.9278 (1.0057) loss 0.6984 (0.8433) grad_norm 9.4606 (8.5800/1.7757) mem 68106MB [2022-12-20 02:20:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][320/1519] eta 0:20:07 lr 0.000017 time 0.9214 (1.0072) model_time 0.9212 (1.0055) loss 0.7681 (0.8424) grad_norm 8.0874 (8.5550/1.7541) mem 68106MB [2022-12-20 02:20:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][330/1519] eta 0:19:57 lr 0.000017 time 0.9950 (1.0072) model_time 0.9948 (1.0055) loss 0.8275 (0.8427) grad_norm 6.3958 (8.5491/1.7494) mem 68106MB [2022-12-20 02:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][340/1519] eta 0:19:47 lr 0.000017 time 0.9272 (1.0074) model_time 0.9271 (1.0057) loss 0.9659 (0.8429) grad_norm 8.3183 (8.5674/1.7339) mem 68106MB [2022-12-20 02:21:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][350/1519] eta 0:19:37 lr 0.000017 time 0.9163 (1.0075) model_time 0.9162 (1.0059) loss 0.8086 (0.8430) grad_norm 9.1053 (8.5953/1.7333) mem 68106MB [2022-12-20 02:21:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][360/1519] eta 0:19:27 lr 0.000017 time 0.9193 (1.0076) model_time 0.9191 (1.0060) loss 0.6870 (0.8413) grad_norm 10.3798 (8.5942/1.7204) mem 68106MB [2022-12-20 02:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][370/1519] eta 0:19:17 lr 0.000017 time 0.8907 (1.0076) model_time 0.8906 (1.0060) loss 1.0744 (0.8417) grad_norm 7.8927 (8.6012/1.7141) mem 68106MB [2022-12-20 02:21:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][380/1519] eta 0:19:07 lr 0.000017 time 0.9097 (1.0075) model_time 0.9095 (1.0060) loss 0.8437 (0.8427) grad_norm 7.9883 (8.5814/1.7095) mem 68106MB [2022-12-20 02:21:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][390/1519] eta 0:18:57 lr 0.000017 time 0.9280 (1.0073) model_time 0.9279 (1.0058) loss 0.9567 (0.8427) grad_norm 7.3637 (8.5450/1.7104) mem 68106MB [2022-12-20 02:21:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][400/1519] eta 0:18:47 lr 0.000017 time 0.9285 (1.0076) model_time 0.9283 (1.0061) loss 0.8548 (0.8429) grad_norm 6.6297 (8.5315/1.7102) mem 68106MB [2022-12-20 02:22:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][410/1519] eta 0:18:37 lr 0.000017 time 0.9276 (1.0074) model_time 0.9275 (1.0060) loss 0.8711 (0.8415) grad_norm 8.0090 (8.5233/1.7021) mem 68106MB [2022-12-20 02:22:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][420/1519] eta 0:18:27 lr 0.000017 time 0.9533 (1.0074) model_time 0.9532 (1.0060) loss 0.8271 (0.8402) grad_norm 7.3149 (8.5148/1.6940) mem 68106MB [2022-12-20 02:22:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][430/1519] eta 0:18:17 lr 0.000017 time 0.9324 (1.0079) model_time 0.9322 (1.0065) loss 0.9026 (0.8415) grad_norm 9.8085 (8.5173/1.7210) mem 68106MB [2022-12-20 02:22:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][440/1519] eta 0:18:07 lr 0.000017 time 0.9289 (1.0078) model_time 0.9288 (1.0064) loss 0.9115 (0.8418) grad_norm 11.3837 (8.5100/1.7242) mem 68106MB [2022-12-20 02:22:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][450/1519] eta 0:17:57 lr 0.000017 time 0.9232 (1.0076) model_time 0.9230 (1.0062) loss 0.6870 (0.8428) grad_norm 9.2193 (8.5038/1.7233) mem 68106MB [2022-12-20 02:22:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][460/1519] eta 0:17:46 lr 0.000017 time 0.9488 (1.0075) model_time 0.9487 (1.0062) loss 0.8738 (0.8424) grad_norm 10.1338 (8.5384/1.7288) mem 68106MB [2022-12-20 02:23:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][470/1519] eta 0:17:36 lr 0.000017 time 1.0421 (1.0075) model_time 1.0420 (1.0062) loss 1.0485 (0.8435) grad_norm 6.5344 (8.5039/1.7303) mem 68106MB [2022-12-20 02:23:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][480/1519] eta 0:17:26 lr 0.000017 time 0.9203 (1.0075) model_time 0.9201 (1.0062) loss 0.6981 (0.8419) grad_norm 7.3676 (8.4980/1.7195) mem 68106MB [2022-12-20 02:23:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][490/1519] eta 0:17:16 lr 0.000017 time 0.9186 (1.0074) model_time 0.9185 (1.0061) loss 1.1455 (0.8431) grad_norm 7.8482 (8.4923/1.7177) mem 68106MB [2022-12-20 02:23:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][500/1519] eta 0:17:06 lr 0.000017 time 0.9243 (1.0072) model_time 0.9241 (1.0060) loss 0.9039 (0.8424) grad_norm 8.2628 (8.5113/1.7102) mem 68106MB [2022-12-20 02:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][510/1519] eta 0:16:56 lr 0.000017 time 0.9334 (1.0072) model_time 0.9333 (1.0060) loss 1.1116 (0.8431) grad_norm 5.5012 (8.4942/1.7144) mem 68106MB [2022-12-20 02:23:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][520/1519] eta 0:16:46 lr 0.000017 time 0.9387 (1.0072) model_time 0.9385 (1.0060) loss 0.7754 (0.8426) grad_norm 9.2180 (8.5032/1.7107) mem 68106MB [2022-12-20 02:24:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][530/1519] eta 0:16:36 lr 0.000017 time 0.9224 (1.0071) model_time 0.9223 (1.0060) loss 0.8072 (0.8425) grad_norm 6.9939 (8.4901/1.7035) mem 68106MB [2022-12-20 02:24:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][540/1519] eta 0:16:26 lr 0.000017 time 0.9303 (1.0073) model_time 0.9302 (1.0061) loss 0.6915 (0.8424) grad_norm 6.3868 (8.4889/1.7040) mem 68106MB [2022-12-20 02:24:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][550/1519] eta 0:16:15 lr 0.000017 time 0.9317 (1.0071) model_time 0.9316 (1.0060) loss 0.7236 (0.8419) grad_norm 11.2990 (8.4892/1.7011) mem 68106MB [2022-12-20 02:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][560/1519] eta 0:16:05 lr 0.000017 time 0.9238 (1.0070) model_time 0.9237 (1.0058) loss 0.7275 (0.8419) grad_norm 6.0504 (8.4837/1.7002) mem 68106MB [2022-12-20 02:24:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][570/1519] eta 0:15:55 lr 0.000017 time 0.9314 (1.0069) model_time 0.9312 (1.0058) loss 0.7561 (0.8421) grad_norm 8.1019 (8.4594/1.7008) mem 68106MB [2022-12-20 02:24:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][580/1519] eta 0:15:45 lr 0.000017 time 0.9184 (1.0067) model_time 0.9182 (1.0056) loss 0.6662 (0.8421) grad_norm 7.5700 (8.4554/1.6871) mem 68106MB [2022-12-20 02:25:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][590/1519] eta 0:15:35 lr 0.000017 time 0.9225 (1.0066) model_time 0.9224 (1.0055) loss 0.8566 (0.8415) grad_norm 8.7315 (8.4756/1.7258) mem 68106MB [2022-12-20 02:25:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][600/1519] eta 0:15:24 lr 0.000017 time 0.9239 (1.0064) model_time 0.9237 (1.0054) loss 0.9339 (0.8420) grad_norm 6.8706 (8.4537/1.7356) mem 68106MB [2022-12-20 02:25:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][610/1519] eta 0:15:14 lr 0.000017 time 0.9199 (1.0065) model_time 0.9197 (1.0054) loss 0.8507 (0.8423) grad_norm 7.2143 (8.4535/1.7323) mem 68106MB [2022-12-20 02:25:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][620/1519] eta 0:15:04 lr 0.000017 time 0.9233 (1.0063) model_time 0.9232 (1.0053) loss 0.7274 (0.8426) grad_norm 5.9535 (8.4364/1.7397) mem 68106MB [2022-12-20 02:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][630/1519] eta 0:14:54 lr 0.000017 time 0.9212 (1.0062) model_time 0.9211 (1.0051) loss 0.9469 (0.8432) grad_norm 6.4471 (8.4230/1.7323) mem 68106MB [2022-12-20 02:25:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][640/1519] eta 0:14:44 lr 0.000017 time 0.9226 (1.0061) model_time 0.9225 (1.0050) loss 1.0981 (0.8438) grad_norm 8.0302 (8.4179/1.7457) mem 68106MB [2022-12-20 02:26:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][650/1519] eta 0:14:34 lr 0.000017 time 1.0102 (1.0066) model_time 1.0101 (1.0056) loss 0.8217 (0.8441) grad_norm 6.9861 (8.4080/1.7484) mem 68106MB [2022-12-20 02:26:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][660/1519] eta 0:14:24 lr 0.000017 time 0.9255 (1.0066) model_time 0.9254 (1.0056) loss 0.8680 (0.8445) grad_norm 10.2063 (8.3987/1.7518) mem 68106MB [2022-12-20 02:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][670/1519] eta 0:14:14 lr 0.000017 time 0.9206 (1.0066) model_time 0.9205 (1.0056) loss 0.7837 (0.8443) grad_norm 5.9880 (8.4139/1.7507) mem 68106MB [2022-12-20 02:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][680/1519] eta 0:14:04 lr 0.000017 time 0.9260 (1.0065) model_time 0.9259 (1.0055) loss 0.9385 (0.8448) grad_norm 8.4259 (8.4315/1.7488) mem 68106MB [2022-12-20 02:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][690/1519] eta 0:13:54 lr 0.000017 time 0.9249 (1.0064) model_time 0.9247 (1.0054) loss 0.9840 (0.8447) grad_norm 6.7186 (8.4307/1.7536) mem 68106MB [2022-12-20 02:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][700/1519] eta 0:13:44 lr 0.000017 time 0.9282 (1.0063) model_time 0.9280 (1.0054) loss 0.6987 (0.8439) grad_norm 8.5278 (8.4008/1.7063) mem 68106MB [2022-12-20 02:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][710/1519] eta 0:13:34 lr 0.000017 time 0.9334 (1.0063) model_time 0.9333 (1.0053) loss 1.0767 (0.8442) grad_norm 9.1958 (8.4204/1.7235) mem 68106MB [2022-12-20 02:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][720/1519] eta 0:13:24 lr 0.000017 time 0.9222 (1.0063) model_time 0.9221 (1.0054) loss 0.7344 (0.8447) grad_norm 9.4530 (8.4124/1.6470) mem 68106MB [2022-12-20 02:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][730/1519] eta 0:13:13 lr 0.000017 time 0.9191 (1.0063) model_time 0.9189 (1.0054) loss 0.7243 (0.8437) grad_norm 8.8513 (8.4024/1.6323) mem 68106MB [2022-12-20 02:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][740/1519] eta 0:13:03 lr 0.000017 time 0.9255 (1.0062) model_time 0.9253 (1.0053) loss 0.7922 (0.8437) grad_norm 9.8322 (8.4014/1.6286) mem 68106MB [2022-12-20 02:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][750/1519] eta 0:12:53 lr 0.000017 time 0.9230 (1.0062) model_time 0.9228 (1.0053) loss 0.7707 (0.8437) grad_norm 7.7022 (8.4005/1.6260) mem 68106MB [2022-12-20 02:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][760/1519] eta 0:12:43 lr 0.000017 time 0.9334 (1.0061) model_time 0.9332 (1.0052) loss 0.8854 (0.8438) grad_norm 7.9715 (8.4010/1.6294) mem 68106MB [2022-12-20 02:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][770/1519] eta 0:12:33 lr 0.000017 time 0.9470 (1.0061) model_time 0.9469 (1.0051) loss 0.7197 (0.8449) grad_norm 6.0718 (8.4036/1.6439) mem 68106MB [2022-12-20 02:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][780/1519] eta 0:12:23 lr 0.000017 time 0.9215 (1.0060) model_time 0.9214 (1.0051) loss 0.6761 (0.8442) grad_norm 7.0630 (8.3864/1.6549) mem 68106MB [2022-12-20 02:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][790/1519] eta 0:12:13 lr 0.000017 time 0.9244 (1.0059) model_time 0.9243 (1.0050) loss 0.6900 (0.8440) grad_norm 8.3924 (8.3854/1.6568) mem 68106MB [2022-12-20 02:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][800/1519] eta 0:12:03 lr 0.000017 time 0.9271 (1.0058) model_time 0.9270 (1.0050) loss 0.7474 (0.8435) grad_norm 11.0002 (8.3855/1.6674) mem 68106MB [2022-12-20 02:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][810/1519] eta 0:11:53 lr 0.000017 time 0.9241 (1.0058) model_time 0.9240 (1.0049) loss 0.9032 (0.8434) grad_norm 9.3782 (8.3828/1.6646) mem 68106MB [2022-12-20 02:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][820/1519] eta 0:11:43 lr 0.000017 time 0.9203 (1.0057) model_time 0.9202 (1.0049) loss 0.8051 (0.8432) grad_norm 9.1733 (8.3925/1.6670) mem 68106MB [2022-12-20 02:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][830/1519] eta 0:11:32 lr 0.000017 time 0.9409 (1.0056) model_time 0.9408 (1.0048) loss 1.0387 (0.8432) grad_norm 7.3605 (8.3791/1.6751) mem 68106MB [2022-12-20 02:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][840/1519] eta 0:11:22 lr 0.000017 time 0.9200 (1.0056) model_time 0.9198 (1.0048) loss 1.1753 (0.8437) grad_norm 9.5565 (8.4024/1.6803) mem 68106MB [2022-12-20 02:29:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][850/1519] eta 0:11:12 lr 0.000017 time 0.9796 (1.0056) model_time 0.9794 (1.0048) loss 0.7400 (0.8439) grad_norm 8.1483 (8.4115/1.6790) mem 68106MB [2022-12-20 02:29:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][860/1519] eta 0:11:02 lr 0.000017 time 0.9242 (1.0056) model_time 0.9241 (1.0047) loss 0.9552 (0.8443) grad_norm 8.2007 (8.3721/1.6554) mem 68106MB [2022-12-20 02:29:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][870/1519] eta 0:10:52 lr 0.000017 time 0.9253 (1.0055) model_time 0.9251 (1.0046) loss 0.7101 (0.8440) grad_norm 8.2771 (8.3625/1.6367) mem 68106MB [2022-12-20 02:29:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][880/1519] eta 0:10:42 lr 0.000017 time 0.9303 (1.0054) model_time 0.9302 (1.0046) loss 0.8847 (0.8438) grad_norm 7.2485 (8.3848/1.6518) mem 68106MB [2022-12-20 02:30:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][890/1519] eta 0:10:32 lr 0.000017 time 0.9323 (1.0054) model_time 0.9322 (1.0045) loss 0.9543 (0.8434) grad_norm 8.8525 (8.3476/1.6062) mem 68106MB [2022-12-20 02:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][900/1519] eta 0:10:22 lr 0.000017 time 0.9271 (1.0053) model_time 0.9270 (1.0044) loss 1.0244 (0.8430) grad_norm 9.6427 (8.3117/1.5665) mem 68106MB [2022-12-20 02:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][910/1519] eta 0:10:12 lr 0.000017 time 0.9188 (1.0052) model_time 0.9187 (1.0044) loss 0.6760 (0.8428) grad_norm 12.0390 (8.3358/1.5770) mem 68106MB [2022-12-20 02:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][920/1519] eta 0:10:02 lr 0.000017 time 0.9156 (1.0053) model_time 0.9139 (1.0045) loss 0.9809 (0.8431) grad_norm 7.1495 (8.3418/1.5813) mem 68106MB [2022-12-20 02:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][930/1519] eta 0:09:52 lr 0.000017 time 0.9275 (1.0052) model_time 0.9272 (1.0044) loss 1.0545 (0.8435) grad_norm 7.0532 (8.3381/1.5814) mem 68106MB [2022-12-20 02:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][940/1519] eta 0:09:41 lr 0.000017 time 0.9237 (1.0052) model_time 0.9235 (1.0043) loss 0.7222 (0.8434) grad_norm 7.6027 (8.3356/1.6324) mem 68106MB [2022-12-20 02:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][950/1519] eta 0:09:31 lr 0.000017 time 0.9323 (1.0051) model_time 0.9321 (1.0043) loss 0.7017 (0.8432) grad_norm 6.9125 (8.3259/1.6593) mem 68106MB [2022-12-20 02:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][960/1519] eta 0:09:21 lr 0.000017 time 0.9180 (1.0052) model_time 0.9178 (1.0044) loss 0.7388 (0.8429) grad_norm 6.9070 (8.3105/1.6557) mem 68106MB [2022-12-20 02:31:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][970/1519] eta 0:09:11 lr 0.000017 time 0.9906 (1.0052) model_time 0.9905 (1.0044) loss 0.6714 (0.8419) grad_norm 9.7231 (8.3121/1.6483) mem 68106MB [2022-12-20 02:31:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][980/1519] eta 0:09:01 lr 0.000017 time 0.9246 (1.0052) model_time 0.9245 (1.0044) loss 1.2218 (0.8425) grad_norm 8.1033 (8.3112/1.6413) mem 68106MB [2022-12-20 02:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][990/1519] eta 0:08:51 lr 0.000017 time 0.9236 (1.0051) model_time 0.9235 (1.0043) loss 0.7101 (0.8420) grad_norm 9.5086 (8.3370/1.6346) mem 68106MB [2022-12-20 02:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1000/1519] eta 0:08:41 lr 0.000017 time 0.9286 (1.0051) model_time 0.9284 (1.0043) loss 0.7296 (0.8413) grad_norm 7.5459 (8.3343/1.6254) mem 68106MB [2022-12-20 02:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1010/1519] eta 0:08:31 lr 0.000017 time 0.9345 (1.0051) model_time 0.9344 (1.0043) loss 0.8547 (0.8412) grad_norm 8.5347 (8.3318/1.6316) mem 68106MB [2022-12-20 02:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1020/1519] eta 0:08:21 lr 0.000017 time 0.9216 (1.0051) model_time 0.9215 (1.0044) loss 0.8176 (0.8416) grad_norm 7.1205 (8.3352/1.6534) mem 68106MB [2022-12-20 02:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1030/1519] eta 0:08:11 lr 0.000017 time 0.9841 (1.0052) model_time 0.9840 (1.0045) loss 0.7686 (0.8413) grad_norm 7.0109 (8.3251/1.6222) mem 68106MB [2022-12-20 02:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1040/1519] eta 0:08:01 lr 0.000017 time 0.9459 (1.0052) model_time 0.9458 (1.0044) loss 0.7293 (0.8406) grad_norm 7.1980 (8.3190/1.6086) mem 68106MB [2022-12-20 02:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1050/1519] eta 0:07:51 lr 0.000017 time 0.9256 (1.0052) model_time 0.9254 (1.0044) loss 0.9943 (0.8412) grad_norm 8.5878 (8.3150/1.6284) mem 68106MB [2022-12-20 02:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1060/1519] eta 0:07:41 lr 0.000017 time 0.9247 (1.0052) model_time 0.9245 (1.0044) loss 0.7744 (0.8414) grad_norm 7.1943 (8.2613/1.6186) mem 68106MB [2022-12-20 02:33:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1070/1519] eta 0:07:31 lr 0.000017 time 0.9379 (1.0051) model_time 0.9377 (1.0044) loss 1.1954 (0.8424) grad_norm 7.1207 (8.2599/1.6176) mem 68106MB [2022-12-20 02:33:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1080/1519] eta 0:07:21 lr 0.000017 time 0.9267 (1.0051) model_time 0.9266 (1.0043) loss 0.7179 (0.8416) grad_norm 8.2976 (8.2519/1.6220) mem 68106MB [2022-12-20 02:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1090/1519] eta 0:07:11 lr 0.000017 time 0.9253 (1.0050) model_time 0.9251 (1.0043) loss 0.8612 (0.8422) grad_norm 8.8431 (8.2843/1.6627) mem 68106MB [2022-12-20 02:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1100/1519] eta 0:07:01 lr 0.000017 time 0.9276 (1.0051) model_time 0.9274 (1.0043) loss 1.1334 (0.8419) grad_norm 8.4467 (8.2654/1.6541) mem 68106MB [2022-12-20 02:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1110/1519] eta 0:06:51 lr 0.000017 time 0.9303 (1.0051) model_time 0.9301 (1.0043) loss 0.7514 (0.8428) grad_norm 11.3797 (8.3013/1.6547) mem 68106MB [2022-12-20 02:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1120/1519] eta 0:06:41 lr 0.000017 time 0.9323 (1.0051) model_time 0.9322 (1.0043) loss 0.6749 (0.8428) grad_norm 8.7607 (8.2898/1.6522) mem 68106MB [2022-12-20 02:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1130/1519] eta 0:06:30 lr 0.000017 time 0.9298 (1.0050) model_time 0.9297 (1.0043) loss 0.7442 (0.8423) grad_norm 8.3788 (8.2994/1.6593) mem 68106MB [2022-12-20 02:34:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1140/1519] eta 0:06:20 lr 0.000017 time 0.9312 (1.0051) model_time 0.9310 (1.0043) loss 1.1641 (0.8428) grad_norm 5.8972 (8.2706/1.6680) mem 68106MB [2022-12-20 02:34:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1150/1519] eta 0:06:10 lr 0.000017 time 0.9881 (1.0051) model_time 0.9879 (1.0044) loss 0.7960 (0.8426) grad_norm 6.1660 (8.2438/1.6668) mem 68106MB [2022-12-20 02:34:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1160/1519] eta 0:06:00 lr 0.000017 time 0.9082 (1.0051) model_time 0.9081 (1.0044) loss 0.6931 (0.8424) grad_norm 6.2523 (8.2524/1.6813) mem 68106MB [2022-12-20 02:34:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1170/1519] eta 0:05:50 lr 0.000017 time 0.9344 (1.0051) model_time 0.9342 (1.0044) loss 1.0474 (0.8429) grad_norm 7.1804 (8.2698/1.6833) mem 68106MB [2022-12-20 02:34:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1180/1519] eta 0:05:40 lr 0.000017 time 0.9201 (1.0052) model_time 0.9200 (1.0044) loss 0.9624 (0.8426) grad_norm 7.0956 (8.2912/1.7347) mem 68106MB [2022-12-20 02:35:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1190/1519] eta 0:05:30 lr 0.000017 time 0.9245 (1.0052) model_time 0.9243 (1.0045) loss 1.0043 (0.8426) grad_norm 9.6412 (8.2628/1.6852) mem 68106MB [2022-12-20 02:35:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1200/1519] eta 0:05:20 lr 0.000017 time 0.9197 (1.0052) model_time 0.9195 (1.0045) loss 0.7251 (0.8421) grad_norm 8.3561 (8.2974/1.6717) mem 68106MB [2022-12-20 02:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1210/1519] eta 0:05:10 lr 0.000017 time 0.9250 (1.0053) model_time 0.9249 (1.0046) loss 0.7711 (0.8421) grad_norm 9.6518 (8.3355/1.7338) mem 68106MB [2022-12-20 02:35:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1220/1519] eta 0:05:00 lr 0.000017 time 0.9219 (1.0052) model_time 0.9217 (1.0045) loss 0.7495 (0.8423) grad_norm 5.7214 (8.3320/1.7319) mem 68106MB [2022-12-20 02:35:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1230/1519] eta 0:04:50 lr 0.000017 time 0.9287 (1.0052) model_time 0.9286 (1.0045) loss 0.8799 (0.8421) grad_norm 9.1683 (8.3270/1.7316) mem 68106MB [2022-12-20 02:35:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1240/1519] eta 0:04:40 lr 0.000017 time 0.9232 (1.0053) model_time 0.9231 (1.0046) loss 0.7838 (0.8422) grad_norm 9.2173 (8.3372/1.7243) mem 68106MB [2022-12-20 02:36:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1250/1519] eta 0:04:30 lr 0.000017 time 0.9292 (1.0052) model_time 0.9291 (1.0046) loss 0.8841 (0.8420) grad_norm 11.8092 (8.3483/1.7423) mem 68106MB [2022-12-20 02:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1260/1519] eta 0:04:20 lr 0.000017 time 0.9033 (1.0052) model_time 0.9032 (1.0045) loss 0.7414 (0.8415) grad_norm 7.2608 (8.3416/1.7336) mem 68106MB [2022-12-20 02:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1270/1519] eta 0:04:10 lr 0.000017 time 0.9220 (1.0052) model_time 0.9219 (1.0045) loss 1.0523 (0.8422) grad_norm 7.4795 (8.3546/1.7318) mem 68106MB [2022-12-20 02:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1280/1519] eta 0:04:00 lr 0.000017 time 0.9276 (1.0052) model_time 0.9275 (1.0045) loss 0.9161 (0.8424) grad_norm 6.3024 (8.3412/1.7526) mem 68106MB [2022-12-20 02:36:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1290/1519] eta 0:03:50 lr 0.000017 time 0.9714 (1.0052) model_time 0.9713 (1.0045) loss 0.7284 (0.8429) grad_norm 7.4682 (8.3516/1.7468) mem 68106MB [2022-12-20 02:36:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1300/1519] eta 0:03:40 lr 0.000017 time 0.9230 (1.0051) model_time 0.9229 (1.0045) loss 0.8642 (0.8427) grad_norm 8.8977 (8.3473/1.7378) mem 68106MB [2022-12-20 02:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1310/1519] eta 0:03:30 lr 0.000017 time 0.9230 (1.0051) model_time 0.9229 (1.0044) loss 0.6859 (0.8428) grad_norm 14.1613 (8.3292/1.7570) mem 68106MB [2022-12-20 02:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1320/1519] eta 0:03:20 lr 0.000017 time 0.9371 (1.0051) model_time 0.9369 (1.0044) loss 0.6844 (0.8428) grad_norm 6.2293 (8.3330/1.7790) mem 68106MB [2022-12-20 02:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1330/1519] eta 0:03:09 lr 0.000017 time 0.9238 (1.0050) model_time 0.9236 (1.0044) loss 1.0017 (0.8427) grad_norm 14.5030 (8.3936/1.9234) mem 68106MB [2022-12-20 02:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1340/1519] eta 0:02:59 lr 0.000017 time 0.9288 (1.0050) model_time 0.9287 (1.0043) loss 0.7704 (0.8428) grad_norm 11.6543 (8.4035/1.9493) mem 68106MB [2022-12-20 02:37:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1350/1519] eta 0:02:49 lr 0.000017 time 0.9218 (1.0052) model_time 0.9217 (1.0045) loss 0.6860 (0.8429) grad_norm 8.7047 (8.3995/1.9383) mem 68106MB [2022-12-20 02:37:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1360/1519] eta 0:02:39 lr 0.000017 time 0.9195 (1.0052) model_time 0.9194 (1.0045) loss 0.8094 (0.8425) grad_norm 5.8947 (8.3976/1.9380) mem 68106MB [2022-12-20 02:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1370/1519] eta 0:02:29 lr 0.000017 time 0.9190 (1.0052) model_time 0.9189 (1.0045) loss 0.8471 (0.8426) grad_norm 7.6696 (8.3885/1.9262) mem 68106MB [2022-12-20 02:38:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1380/1519] eta 0:02:19 lr 0.000017 time 0.9329 (1.0051) model_time 0.9327 (1.0045) loss 0.7177 (0.8419) grad_norm 8.4370 (8.4333/1.9263) mem 68106MB [2022-12-20 02:38:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1390/1519] eta 0:02:09 lr 0.000017 time 0.9282 (1.0051) model_time 0.9280 (1.0044) loss 0.6855 (0.8421) grad_norm 8.4207 (8.4410/1.9300) mem 68106MB [2022-12-20 02:38:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1400/1519] eta 0:01:59 lr 0.000017 time 0.9287 (1.0051) model_time 0.9285 (1.0044) loss 0.7827 (0.8425) grad_norm 6.7733 (8.4273/1.9248) mem 68106MB [2022-12-20 02:38:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1410/1519] eta 0:01:49 lr 0.000017 time 0.9044 (1.0050) model_time 0.9043 (1.0044) loss 0.8921 (0.8421) grad_norm 13.5368 (8.4427/1.9537) mem 68106MB [2022-12-20 02:38:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1420/1519] eta 0:01:39 lr 0.000017 time 0.9230 (1.0050) model_time 0.9229 (1.0044) loss 0.7272 (0.8423) grad_norm 9.2606 (8.4560/1.9808) mem 68106MB [2022-12-20 02:39:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1430/1519] eta 0:01:29 lr 0.000017 time 0.9237 (1.0050) model_time 0.9236 (1.0044) loss 1.0893 (0.8427) grad_norm 12.1830 (8.4916/1.9827) mem 68106MB [2022-12-20 02:39:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1440/1519] eta 0:01:19 lr 0.000017 time 0.9346 (1.0050) model_time 0.9345 (1.0043) loss 1.0076 (0.8425) grad_norm 8.6944 (8.4932/1.9755) mem 68106MB [2022-12-20 02:39:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1450/1519] eta 0:01:09 lr 0.000017 time 0.9310 (1.0050) model_time 0.9308 (1.0044) loss 0.9070 (0.8426) grad_norm 10.6357 (8.4973/1.9800) mem 68106MB [2022-12-20 02:39:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1460/1519] eta 0:00:59 lr 0.000017 time 0.9225 (1.0052) model_time 0.9224 (1.0045) loss 0.7977 (0.8422) grad_norm 6.2074 (8.4912/1.9887) mem 68106MB [2022-12-20 02:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1470/1519] eta 0:00:49 lr 0.000017 time 0.9757 (1.0052) model_time 0.9755 (1.0046) loss 0.7249 (0.8424) grad_norm 10.3334 (8.5340/2.0705) mem 68106MB [2022-12-20 02:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1480/1519] eta 0:00:39 lr 0.000017 time 0.9296 (1.0052) model_time 0.9295 (1.0046) loss 0.7956 (0.8425) grad_norm 7.1605 (8.5238/2.0525) mem 68106MB [2022-12-20 02:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1490/1519] eta 0:00:29 lr 0.000017 time 0.9221 (1.0052) model_time 0.9220 (1.0046) loss 0.9416 (0.8423) grad_norm 7.4117 (8.5248/2.0505) mem 68106MB [2022-12-20 02:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1500/1519] eta 0:00:19 lr 0.000017 time 0.9290 (1.0052) model_time 0.9288 (1.0046) loss 0.6748 (0.8428) grad_norm 7.8845 (8.5519/2.0524) mem 68106MB [2022-12-20 02:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [50/100][1510/1519] eta 0:00:09 lr 0.000017 time 0.9228 (1.0051) model_time 0.9227 (1.0045) loss 0.9105 (0.8426) grad_norm 9.7623 (8.5365/2.0457) mem 68106MB [2022-12-20 02:40:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 50 training takes 0:25:26 [2022-12-20 02:40:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_50.pth saving...... [2022-12-20 02:41:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_50.pth saved !!! [2022-12-20 02:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.668 (0.668) Loss 0.5086 (0.5086) Acc@1 91.319 (91.319) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 02:41:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.330) Loss 0.5070 (0.4868) Acc@1 93.056 (92.456) Acc@5 98.264 (98.643) Mem 68106MB [2022-12-20 02:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.311 (0.315) Loss 0.4773 (0.4843) Acc@1 91.319 (92.328) Acc@5 98.611 (98.528) Mem 68106MB [2022-12-20 02:41:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.310) Loss 0.5961 (0.4925) Acc@1 90.278 (92.025) Acc@5 96.875 (98.342) Mem 68106MB [2022-12-20 02:41:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.307 (0.308) Loss 0.4580 (0.4850) Acc@1 93.056 (92.141) Acc@5 98.958 (98.450) Mem 68106MB [2022-12-20 02:41:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 0.4729 (0.4817) Acc@1 90.625 (92.205) Acc@5 99.653 (98.509) Mem 68106MB [2022-12-20 02:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.4886 (0.4816) Acc@1 90.278 (92.202) Acc@5 97.917 (98.475) Mem 68106MB [2022-12-20 02:41:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5270 (0.4822) Acc@1 90.972 (92.146) Acc@5 98.264 (98.484) Mem 68106MB [2022-12-20 02:41:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.303 (0.304) Loss 0.4136 (0.4801) Acc@1 93.056 (92.138) Acc@5 98.264 (98.517) Mem 68106MB [2022-12-20 02:41:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:50] * Acc@1 92.121 Acc@5 98.514 [2022-12-20 02:41:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.1% [2022-12-20 02:41:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.17% [2022-12-20 02:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][0/1519] eta 0:45:37 lr 0.000017 time 1.8024 (1.8024) model_time 1.0027 (1.0027) loss 0.9621 (0.9621) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 02:41:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][10/1519] eta 0:26:55 lr 0.000017 time 0.9263 (1.0705) model_time 0.9262 (0.9975) loss 0.7023 (0.8258) grad_norm 8.7562 (7.1196/1.2751) mem 68106MB [2022-12-20 02:41:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][20/1519] eta 0:25:56 lr 0.000017 time 0.9257 (1.0381) model_time 0.9255 (0.9997) loss 0.7650 (0.8225) grad_norm 8.8361 (7.4422/1.2613) mem 68106MB [2022-12-20 02:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][30/1519] eta 0:25:47 lr 0.000017 time 0.9834 (1.0394) model_time 0.9832 (1.0133) loss 0.6962 (0.8260) grad_norm 7.4402 (7.4974/1.0815) mem 68106MB [2022-12-20 02:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][40/1519] eta 0:25:22 lr 0.000017 time 0.9291 (1.0295) model_time 0.9289 (1.0096) loss 0.7551 (0.8380) grad_norm 10.3214 (7.9866/1.4611) mem 68106MB [2022-12-20 02:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][50/1519] eta 0:25:03 lr 0.000017 time 0.9272 (1.0232) model_time 0.9271 (1.0072) loss 1.1240 (0.8483) grad_norm 8.5910 (8.2293/1.5257) mem 68106MB [2022-12-20 02:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][60/1519] eta 0:24:47 lr 0.000016 time 0.9279 (1.0195) model_time 0.9278 (1.0060) loss 1.0338 (0.8442) grad_norm 8.2548 (8.2432/1.4266) mem 68106MB [2022-12-20 02:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][70/1519] eta 0:24:35 lr 0.000016 time 0.9288 (1.0182) model_time 0.9287 (1.0066) loss 0.6857 (0.8429) grad_norm 8.1208 (8.0727/1.4017) mem 68106MB [2022-12-20 02:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][80/1519] eta 0:24:23 lr 0.000016 time 0.9243 (1.0171) model_time 0.9242 (1.0069) loss 0.9254 (0.8431) grad_norm 6.4796 (8.0661/1.4038) mem 68106MB [2022-12-20 02:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][90/1519] eta 0:24:10 lr 0.000016 time 0.9246 (1.0150) model_time 0.9245 (1.0059) loss 0.7132 (0.8368) grad_norm 6.8401 (8.1614/1.5826) mem 68106MB [2022-12-20 02:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][100/1519] eta 0:23:57 lr 0.000016 time 0.9267 (1.0134) model_time 0.9266 (1.0051) loss 0.9059 (0.8379) grad_norm 10.5093 (8.2308/1.6070) mem 68106MB [2022-12-20 02:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][110/1519] eta 0:23:46 lr 0.000016 time 0.9175 (1.0122) model_time 0.9173 (1.0047) loss 1.2878 (0.8472) grad_norm 7.9392 (8.2202/1.5477) mem 68106MB [2022-12-20 02:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][120/1519] eta 0:23:35 lr 0.000016 time 0.9236 (1.0115) model_time 0.9235 (1.0046) loss 0.8381 (0.8479) grad_norm 7.2066 (8.2386/1.5767) mem 68106MB [2022-12-20 02:43:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][130/1519] eta 0:23:23 lr 0.000016 time 0.9194 (1.0105) model_time 0.9192 (1.0040) loss 0.9119 (0.8575) grad_norm 7.7381 (8.3308/1.6241) mem 68106MB [2022-12-20 02:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][140/1519] eta 0:23:13 lr 0.000016 time 0.9276 (1.0107) model_time 0.9271 (1.0047) loss 0.9198 (0.8596) grad_norm 7.6070 (8.3010/1.5800) mem 68106MB [2022-12-20 02:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][150/1519] eta 0:23:03 lr 0.000016 time 0.9208 (1.0107) model_time 0.9207 (1.0051) loss 0.6954 (0.8601) grad_norm 8.4142 (8.3785/1.5622) mem 68106MB [2022-12-20 02:44:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][160/1519] eta 0:22:53 lr 0.000016 time 0.9885 (1.0105) model_time 0.9883 (1.0053) loss 0.7202 (0.8600) grad_norm 6.7960 (8.3391/1.5355) mem 68106MB [2022-12-20 02:44:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][170/1519] eta 0:22:42 lr 0.000016 time 0.9286 (1.0102) model_time 0.9285 (1.0052) loss 0.7051 (0.8587) grad_norm 11.6723 (8.3850/1.5476) mem 68106MB [2022-12-20 02:44:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][180/1519] eta 0:22:31 lr 0.000016 time 0.9333 (1.0096) model_time 0.9331 (1.0049) loss 0.6741 (0.8547) grad_norm 7.9169 (8.3325/1.5255) mem 68106MB [2022-12-20 02:44:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][190/1519] eta 0:22:21 lr 0.000016 time 0.9161 (1.0093) model_time 0.9160 (1.0047) loss 0.9158 (0.8538) grad_norm 10.8689 (8.3368/1.5128) mem 68106MB [2022-12-20 02:44:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][200/1519] eta 0:22:10 lr 0.000016 time 0.9344 (1.0089) model_time 0.9343 (1.0046) loss 0.7358 (0.8555) grad_norm 8.4216 (8.3338/1.4960) mem 68106MB [2022-12-20 02:45:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][210/1519] eta 0:22:00 lr 0.000016 time 0.9870 (1.0090) model_time 0.9869 (1.0049) loss 0.6932 (0.8534) grad_norm 7.2714 (8.3058/1.4762) mem 68106MB [2022-12-20 02:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][220/1519] eta 0:21:49 lr 0.000016 time 0.9168 (1.0084) model_time 0.9167 (1.0045) loss 0.6849 (0.8485) grad_norm 6.5983 (8.3403/1.5192) mem 68106MB [2022-12-20 02:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][230/1519] eta 0:21:39 lr 0.000016 time 0.9285 (1.0080) model_time 0.9284 (1.0042) loss 0.7381 (0.8481) grad_norm 7.2721 (8.3676/1.5191) mem 68106MB [2022-12-20 02:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][240/1519] eta 0:21:28 lr 0.000016 time 0.9200 (1.0077) model_time 0.9198 (1.0040) loss 0.8679 (0.8478) grad_norm 8.7645 (8.3839/1.5378) mem 68106MB [2022-12-20 02:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][250/1519] eta 0:21:18 lr 0.000016 time 0.9460 (1.0077) model_time 0.9459 (1.0042) loss 0.9264 (0.8468) grad_norm 8.0342 (8.3834/1.5428) mem 68106MB [2022-12-20 02:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][260/1519] eta 0:21:08 lr 0.000016 time 0.9303 (1.0074) model_time 0.9302 (1.0040) loss 0.6847 (0.8463) grad_norm 9.8596 (8.3901/1.5490) mem 68106MB [2022-12-20 02:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][270/1519] eta 0:20:58 lr 0.000016 time 0.9282 (1.0075) model_time 0.9281 (1.0042) loss 0.8678 (0.8465) grad_norm 6.4498 (8.3585/1.5424) mem 68106MB [2022-12-20 02:46:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][280/1519] eta 0:20:47 lr 0.000016 time 0.9187 (1.0072) model_time 0.9185 (1.0040) loss 0.8491 (0.8457) grad_norm 6.9740 (8.3815/1.5458) mem 68106MB [2022-12-20 02:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][290/1519] eta 0:20:37 lr 0.000016 time 0.9186 (1.0069) model_time 0.9184 (1.0038) loss 0.8480 (0.8455) grad_norm 8.4640 (8.3610/1.5267) mem 68106MB [2022-12-20 02:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][300/1519] eta 0:20:27 lr 0.000016 time 0.9365 (1.0067) model_time 0.9364 (1.0037) loss 0.9209 (0.8431) grad_norm 10.1774 (8.3671/1.5170) mem 68106MB [2022-12-20 02:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][310/1519] eta 0:20:17 lr 0.000016 time 1.0243 (1.0069) model_time 1.0242 (1.0040) loss 1.1568 (0.8439) grad_norm 9.6333 (8.3647/1.5219) mem 68106MB [2022-12-20 02:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][320/1519] eta 0:20:08 lr 0.000016 time 0.9170 (1.0076) model_time 0.9169 (1.0047) loss 0.7007 (0.8433) grad_norm 10.2839 (8.3439/1.5353) mem 68106MB [2022-12-20 02:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][330/1519] eta 0:19:57 lr 0.000016 time 0.9324 (1.0075) model_time 0.9323 (1.0047) loss 0.7283 (0.8426) grad_norm 7.8659 (8.3643/1.5670) mem 68106MB [2022-12-20 02:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][340/1519] eta 0:19:48 lr 0.000016 time 0.9922 (1.0078) model_time 0.9921 (1.0051) loss 0.7901 (0.8424) grad_norm 6.1282 (8.3561/1.5570) mem 68106MB [2022-12-20 02:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][350/1519] eta 0:19:37 lr 0.000016 time 0.9214 (1.0076) model_time 0.9213 (1.0050) loss 0.6791 (0.8413) grad_norm 7.2371 (8.4242/1.6189) mem 68106MB [2022-12-20 02:47:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][360/1519] eta 0:19:27 lr 0.000016 time 0.9219 (1.0073) model_time 0.9218 (1.0048) loss 0.9424 (0.8397) grad_norm 6.0311 (8.4611/1.7895) mem 68106MB [2022-12-20 02:47:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][370/1519] eta 0:19:17 lr 0.000016 time 0.9217 (1.0071) model_time 0.9215 (1.0046) loss 0.9129 (0.8399) grad_norm 6.4076 (8.4217/1.7826) mem 68106MB [2022-12-20 02:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][380/1519] eta 0:19:07 lr 0.000016 time 0.9208 (1.0071) model_time 0.9207 (1.0046) loss 0.6815 (0.8403) grad_norm 8.6003 (8.4008/1.7664) mem 68106MB [2022-12-20 02:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][390/1519] eta 0:18:57 lr 0.000016 time 0.9030 (1.0072) model_time 0.9029 (1.0049) loss 0.7587 (0.8397) grad_norm 7.8558 (8.4071/1.7477) mem 68106MB [2022-12-20 02:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][400/1519] eta 0:18:46 lr 0.000016 time 0.9254 (1.0070) model_time 0.9253 (1.0047) loss 0.6958 (0.8375) grad_norm 8.9874 (8.4166/1.7306) mem 68106MB [2022-12-20 02:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][410/1519] eta 0:18:36 lr 0.000016 time 0.9270 (1.0069) model_time 0.9268 (1.0046) loss 0.8326 (0.8373) grad_norm 8.2011 (8.4324/1.7329) mem 68106MB [2022-12-20 02:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][420/1519] eta 0:18:26 lr 0.000016 time 0.8837 (1.0068) model_time 0.8836 (1.0046) loss 0.7451 (0.8391) grad_norm 8.9288 (8.4613/1.7537) mem 68106MB [2022-12-20 02:48:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][430/1519] eta 0:18:16 lr 0.000016 time 0.9228 (1.0067) model_time 0.9227 (1.0045) loss 1.0848 (0.8405) grad_norm 10.3125 (8.4783/1.7566) mem 68106MB [2022-12-20 02:48:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][440/1519] eta 0:18:06 lr 0.000016 time 0.9277 (1.0068) model_time 0.9275 (1.0046) loss 0.7551 (0.8397) grad_norm 11.2407 (8.5107/1.7779) mem 68106MB [2022-12-20 02:49:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][450/1519] eta 0:17:56 lr 0.000016 time 0.9218 (1.0067) model_time 0.9216 (1.0046) loss 0.7050 (0.8395) grad_norm 8.6299 (8.5121/1.7609) mem 68106MB [2022-12-20 02:49:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][460/1519] eta 0:17:46 lr 0.000016 time 0.9223 (1.0067) model_time 0.9221 (1.0047) loss 0.9908 (0.8427) grad_norm 11.1345 (8.5120/1.7552) mem 68106MB [2022-12-20 02:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][470/1519] eta 0:17:36 lr 0.000016 time 0.9453 (1.0069) model_time 0.9451 (1.0048) loss 0.7427 (0.8413) grad_norm 14.1129 (8.5746/1.8203) mem 68106MB [2022-12-20 02:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][480/1519] eta 0:17:26 lr 0.000016 time 0.9326 (1.0068) model_time 0.9325 (1.0048) loss 0.9612 (0.8419) grad_norm 7.5421 (8.5678/1.8068) mem 68106MB [2022-12-20 02:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][490/1519] eta 0:17:16 lr 0.000016 time 1.0305 (1.0070) model_time 1.0304 (1.0050) loss 0.7559 (0.8426) grad_norm 7.9434 (8.5633/1.8045) mem 68106MB [2022-12-20 02:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][500/1519] eta 0:17:05 lr 0.000016 time 0.9224 (1.0069) model_time 0.9222 (1.0049) loss 0.6964 (0.8424) grad_norm 10.0909 (8.5854/1.8325) mem 68106MB [2022-12-20 02:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][510/1519] eta 0:16:55 lr 0.000016 time 0.9346 (1.0068) model_time 0.9345 (1.0049) loss 0.9737 (0.8426) grad_norm 7.5451 (8.5710/1.8215) mem 68106MB [2022-12-20 02:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][520/1519] eta 0:16:46 lr 0.000016 time 0.9289 (1.0071) model_time 0.9288 (1.0052) loss 0.7177 (0.8418) grad_norm 9.0715 (8.5738/1.8208) mem 68106MB [2022-12-20 02:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][530/1519] eta 0:16:35 lr 0.000016 time 0.9202 (1.0070) model_time 0.9201 (1.0052) loss 0.9144 (0.8419) grad_norm 8.8635 (8.5453/1.8219) mem 68106MB [2022-12-20 02:50:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][540/1519] eta 0:16:25 lr 0.000016 time 0.9285 (1.0068) model_time 0.9284 (1.0050) loss 0.7380 (0.8427) grad_norm 6.6528 (8.5392/1.8096) mem 68106MB [2022-12-20 02:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][550/1519] eta 0:16:15 lr 0.000016 time 0.9228 (1.0067) model_time 0.9226 (1.0049) loss 0.9769 (0.8428) grad_norm 7.7367 (8.5239/1.7986) mem 68106MB [2022-12-20 02:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][560/1519] eta 0:16:05 lr 0.000016 time 0.9192 (1.0066) model_time 0.9191 (1.0049) loss 0.8300 (0.8428) grad_norm 6.5934 (8.5345/1.8001) mem 68106MB [2022-12-20 02:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][570/1519] eta 0:15:55 lr 0.000016 time 0.9259 (1.0065) model_time 0.9257 (1.0047) loss 1.0783 (0.8429) grad_norm 9.4610 (8.5418/1.7863) mem 68106MB [2022-12-20 02:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][580/1519] eta 0:15:45 lr 0.000016 time 0.9267 (1.0065) model_time 0.9266 (1.0048) loss 0.8859 (0.8422) grad_norm 16.2924 (8.5795/1.8545) mem 68106MB [2022-12-20 02:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][590/1519] eta 0:15:34 lr 0.000016 time 0.9225 (1.0064) model_time 0.9224 (1.0047) loss 0.8174 (0.8427) grad_norm 7.1334 (8.5765/1.8531) mem 68106MB [2022-12-20 02:51:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][600/1519] eta 0:15:24 lr 0.000016 time 0.9949 (1.0064) model_time 0.9947 (1.0048) loss 1.2781 (0.8434) grad_norm 10.6957 (8.5615/1.8539) mem 68106MB [2022-12-20 02:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][610/1519] eta 0:15:14 lr 0.000016 time 0.9256 (1.0063) model_time 0.9255 (1.0047) loss 1.0747 (0.8439) grad_norm 13.6120 (8.6216/1.8907) mem 68106MB [2022-12-20 02:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][620/1519] eta 0:15:04 lr 0.000016 time 0.9197 (1.0062) model_time 0.9180 (1.0046) loss 0.9697 (0.8442) grad_norm 7.1901 (8.6370/1.8919) mem 68106MB [2022-12-20 02:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][630/1519] eta 0:14:54 lr 0.000016 time 0.9084 (1.0063) model_time 0.9083 (1.0047) loss 0.9740 (0.8441) grad_norm 7.1721 (8.6460/1.8961) mem 68106MB [2022-12-20 02:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][640/1519] eta 0:14:44 lr 0.000016 time 0.9240 (1.0062) model_time 0.9238 (1.0046) loss 1.1071 (0.8459) grad_norm 10.5095 (8.6510/1.9146) mem 68106MB [2022-12-20 02:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][650/1519] eta 0:14:34 lr 0.000016 time 1.1803 (1.0066) model_time 1.1802 (1.0050) loss 0.6852 (0.8465) grad_norm 6.3110 (8.6314/1.9125) mem 68106MB [2022-12-20 02:52:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][660/1519] eta 0:14:24 lr 0.000016 time 0.9443 (1.0067) model_time 0.9442 (1.0052) loss 0.7332 (0.8463) grad_norm 9.4024 (8.6234/1.9224) mem 68106MB [2022-12-20 02:52:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][670/1519] eta 0:14:14 lr 0.000016 time 0.9218 (1.0066) model_time 0.9217 (1.0051) loss 0.6935 (0.8461) grad_norm 9.7462 (8.6383/1.9187) mem 68106MB [2022-12-20 02:52:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][680/1519] eta 0:14:04 lr 0.000016 time 0.9293 (1.0065) model_time 0.9291 (1.0049) loss 1.1303 (0.8457) grad_norm 7.1772 (8.6486/1.9122) mem 68106MB [2022-12-20 02:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][690/1519] eta 0:13:54 lr 0.000016 time 0.9168 (1.0064) model_time 0.9167 (1.0049) loss 1.0505 (0.8449) grad_norm 7.2770 (8.6282/1.8937) mem 68106MB [2022-12-20 02:53:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][700/1519] eta 0:13:44 lr 0.000016 time 0.9211 (1.0068) model_time 0.9210 (1.0054) loss 0.8253 (0.8450) grad_norm 7.3118 (8.6109/1.8871) mem 68106MB [2022-12-20 02:53:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][710/1519] eta 0:13:34 lr 0.000016 time 0.9234 (1.0068) model_time 0.9232 (1.0053) loss 0.7240 (0.8446) grad_norm 5.9967 (8.6141/1.8917) mem 68106MB [2022-12-20 02:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][720/1519] eta 0:13:24 lr 0.000016 time 0.9246 (1.0067) model_time 0.9245 (1.0053) loss 0.7515 (0.8447) grad_norm 6.7526 (8.6141/1.8843) mem 68106MB [2022-12-20 02:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][730/1519] eta 0:13:14 lr 0.000016 time 0.9333 (1.0066) model_time 0.9331 (1.0052) loss 0.8291 (0.8438) grad_norm 8.6160 (8.5909/1.8692) mem 68106MB [2022-12-20 02:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][740/1519] eta 0:13:04 lr 0.000016 time 0.9277 (1.0066) model_time 0.9275 (1.0052) loss 0.7972 (0.8435) grad_norm 6.7559 (8.5949/1.8732) mem 68106MB [2022-12-20 02:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][750/1519] eta 0:12:54 lr 0.000016 time 0.9298 (1.0065) model_time 0.9296 (1.0052) loss 0.7744 (0.8440) grad_norm 8.7399 (8.5794/1.8743) mem 68106MB [2022-12-20 02:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][760/1519] eta 0:12:43 lr 0.000016 time 0.9191 (1.0065) model_time 0.9190 (1.0051) loss 1.1111 (0.8439) grad_norm 7.4104 (8.5953/1.8749) mem 68106MB [2022-12-20 02:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][770/1519] eta 0:12:33 lr 0.000016 time 0.9289 (1.0064) model_time 0.9287 (1.0051) loss 0.8019 (0.8442) grad_norm 12.0350 (8.6100/1.9016) mem 68106MB [2022-12-20 02:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][780/1519] eta 0:12:23 lr 0.000016 time 0.9216 (1.0064) model_time 0.9215 (1.0050) loss 0.9763 (0.8449) grad_norm 8.1307 (8.6212/1.8975) mem 68106MB [2022-12-20 02:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][790/1519] eta 0:12:13 lr 0.000016 time 0.9770 (1.0064) model_time 0.9769 (1.0050) loss 0.8019 (0.8455) grad_norm 9.2510 (8.6346/1.8959) mem 68106MB [2022-12-20 02:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][800/1519] eta 0:12:03 lr 0.000016 time 0.9233 (1.0063) model_time 0.9232 (1.0050) loss 0.8327 (0.8462) grad_norm 9.0218 (8.6370/1.8949) mem 68106MB [2022-12-20 02:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][810/1519] eta 0:11:53 lr 0.000016 time 0.9260 (1.0064) model_time 0.9259 (1.0051) loss 0.7481 (0.8460) grad_norm 9.0076 (8.6690/1.8988) mem 68106MB [2022-12-20 02:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][820/1519] eta 0:11:43 lr 0.000016 time 0.9382 (1.0063) model_time 0.9380 (1.0050) loss 0.8133 (0.8454) grad_norm 8.0469 (8.6578/1.8800) mem 68106MB [2022-12-20 02:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][830/1519] eta 0:11:33 lr 0.000016 time 1.0043 (1.0063) model_time 1.0042 (1.0050) loss 0.7046 (0.8451) grad_norm 12.2612 (8.6691/1.8890) mem 68106MB [2022-12-20 02:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][840/1519] eta 0:11:23 lr 0.000016 time 0.9249 (1.0064) model_time 0.9248 (1.0051) loss 0.8955 (0.8445) grad_norm 8.5287 (8.6646/1.8801) mem 68106MB [2022-12-20 02:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][850/1519] eta 0:11:13 lr 0.000016 time 0.9192 (1.0063) model_time 0.9191 (1.0051) loss 0.8639 (0.8440) grad_norm 8.8253 (8.6609/1.8843) mem 68106MB [2022-12-20 02:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][860/1519] eta 0:11:03 lr 0.000016 time 0.9462 (1.0063) model_time 0.9460 (1.0050) loss 0.7407 (0.8443) grad_norm 7.7590 (8.6583/1.8754) mem 68106MB [2022-12-20 02:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][870/1519] eta 0:10:53 lr 0.000016 time 0.9204 (1.0062) model_time 0.9202 (1.0049) loss 0.8750 (0.8445) grad_norm 7.4518 (8.6873/1.8978) mem 68106MB [2022-12-20 02:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][880/1519] eta 0:10:42 lr 0.000016 time 0.9231 (1.0062) model_time 0.9230 (1.0050) loss 0.6875 (0.8443) grad_norm 5.5576 (8.6707/1.9012) mem 68106MB [2022-12-20 02:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][890/1519] eta 0:10:32 lr 0.000016 time 0.9813 (1.0062) model_time 0.9811 (1.0050) loss 0.8893 (0.8437) grad_norm 10.9181 (8.6827/1.9032) mem 68106MB [2022-12-20 02:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][900/1519] eta 0:10:22 lr 0.000016 time 0.9251 (1.0062) model_time 0.9250 (1.0049) loss 0.9321 (0.8434) grad_norm 10.6106 (8.7031/1.9047) mem 68106MB [2022-12-20 02:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][910/1519] eta 0:10:12 lr 0.000016 time 0.9317 (1.0061) model_time 0.9316 (1.0049) loss 0.8383 (0.8431) grad_norm 12.3118 (8.7056/1.9092) mem 68106MB [2022-12-20 02:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][920/1519] eta 0:10:02 lr 0.000016 time 0.9250 (1.0061) model_time 0.9249 (1.0049) loss 0.6993 (0.8428) grad_norm 8.4272 (8.7293/1.8923) mem 68106MB [2022-12-20 02:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][930/1519] eta 0:09:52 lr 0.000016 time 0.9420 (1.0060) model_time 0.9419 (1.0048) loss 0.9681 (0.8425) grad_norm 13.2746 (8.7215/1.8966) mem 68106MB [2022-12-20 02:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][940/1519] eta 0:09:42 lr 0.000016 time 0.9247 (1.0060) model_time 0.9246 (1.0048) loss 1.0312 (0.8433) grad_norm 7.0713 (8.7506/1.9066) mem 68106MB [2022-12-20 02:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][950/1519] eta 0:09:32 lr 0.000016 time 0.9706 (1.0060) model_time 0.9705 (1.0048) loss 1.0371 (0.8433) grad_norm 9.6731 (8.7163/1.8747) mem 68106MB [2022-12-20 02:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][960/1519] eta 0:09:22 lr 0.000016 time 0.9034 (1.0060) model_time 0.9032 (1.0048) loss 0.9483 (0.8439) grad_norm 4.8790 (8.6888/1.7871) mem 68106MB [2022-12-20 02:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][970/1519] eta 0:09:12 lr 0.000016 time 0.9219 (1.0059) model_time 0.9217 (1.0048) loss 0.8072 (0.8439) grad_norm 7.8209 (8.7011/1.7831) mem 68106MB [2022-12-20 02:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][980/1519] eta 0:09:02 lr 0.000016 time 0.9205 (1.0059) model_time 0.9204 (1.0047) loss 0.9448 (0.8437) grad_norm 10.4885 (8.7333/1.8318) mem 68106MB [2022-12-20 02:58:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][990/1519] eta 0:08:52 lr 0.000016 time 0.9276 (1.0058) model_time 0.9275 (1.0046) loss 0.9612 (0.8428) grad_norm 7.2262 (8.7185/1.8373) mem 68106MB [2022-12-20 02:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1000/1519] eta 0:08:41 lr 0.000016 time 0.9218 (1.0058) model_time 0.9216 (1.0046) loss 1.1451 (0.8431) grad_norm 7.0566 (8.6884/1.8495) mem 68106MB [2022-12-20 02:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1010/1519] eta 0:08:31 lr 0.000016 time 0.9241 (1.0058) model_time 0.9239 (1.0047) loss 0.9067 (0.8434) grad_norm 6.7799 (8.6729/1.8451) mem 68106MB [2022-12-20 02:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1020/1519] eta 0:08:21 lr 0.000016 time 0.9136 (1.0058) model_time 0.9134 (1.0047) loss 1.3195 (0.8438) grad_norm 6.0543 (8.6432/1.8392) mem 68106MB [2022-12-20 02:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1030/1519] eta 0:08:11 lr 0.000016 time 0.9237 (1.0057) model_time 0.9235 (1.0046) loss 0.7782 (0.8439) grad_norm 9.6830 (8.6542/1.8522) mem 68106MB [2022-12-20 02:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1040/1519] eta 0:08:01 lr 0.000016 time 0.9330 (1.0057) model_time 0.9328 (1.0046) loss 0.7240 (0.8435) grad_norm 11.3174 (8.6334/1.8391) mem 68106MB [2022-12-20 02:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1050/1519] eta 0:07:51 lr 0.000016 time 0.9288 (1.0057) model_time 0.9287 (1.0046) loss 0.6982 (0.8433) grad_norm 5.0525 (8.6087/1.8542) mem 68106MB [2022-12-20 02:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1060/1519] eta 0:07:41 lr 0.000016 time 0.9260 (1.0057) model_time 0.9259 (1.0046) loss 0.9778 (0.8437) grad_norm 8.1148 (8.6013/1.8500) mem 68106MB [2022-12-20 02:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1070/1519] eta 0:07:31 lr 0.000016 time 0.9887 (1.0057) model_time 0.9885 (1.0046) loss 0.9677 (0.8436) grad_norm 9.0181 (8.5498/1.7938) mem 68106MB [2022-12-20 02:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1080/1519] eta 0:07:21 lr 0.000016 time 0.9870 (1.0057) model_time 0.9869 (1.0046) loss 0.9906 (0.8434) grad_norm 7.9563 (8.5446/1.8080) mem 68106MB [2022-12-20 02:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1090/1519] eta 0:07:11 lr 0.000016 time 0.9253 (1.0057) model_time 0.9252 (1.0046) loss 0.7674 (0.8432) grad_norm 9.2006 (8.5690/1.8092) mem 68106MB [2022-12-20 02:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1100/1519] eta 0:07:01 lr 0.000016 time 0.9231 (1.0056) model_time 0.9228 (1.0045) loss 0.7905 (0.8425) grad_norm 6.0572 (8.5430/1.7789) mem 68106MB [2022-12-20 03:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1110/1519] eta 0:06:51 lr 0.000016 time 0.9570 (1.0056) model_time 0.9569 (1.0046) loss 0.9004 (0.8431) grad_norm 8.7317 (8.5523/1.7803) mem 68106MB [2022-12-20 03:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1120/1519] eta 0:06:41 lr 0.000016 time 0.9227 (1.0056) model_time 0.9225 (1.0046) loss 0.8154 (0.8428) grad_norm 9.1902 (8.5687/1.7834) mem 68106MB [2022-12-20 03:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1130/1519] eta 0:06:31 lr 0.000016 time 0.9869 (1.0056) model_time 0.9868 (1.0046) loss 0.7658 (0.8426) grad_norm 9.5391 (8.6001/1.7789) mem 68106MB [2022-12-20 03:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1140/1519] eta 0:06:21 lr 0.000016 time 0.9335 (1.0056) model_time 0.9334 (1.0046) loss 1.1341 (0.8433) grad_norm 8.9284 (8.6090/1.7773) mem 68106MB [2022-12-20 03:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1150/1519] eta 0:06:11 lr 0.000016 time 0.9261 (1.0058) model_time 0.9260 (1.0047) loss 0.6699 (0.8428) grad_norm 8.6579 (8.6132/1.7768) mem 68106MB [2022-12-20 03:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1160/1519] eta 0:06:01 lr 0.000016 time 0.9274 (1.0058) model_time 0.9273 (1.0047) loss 0.6694 (0.8419) grad_norm 8.9437 (8.6156/1.8000) mem 68106MB [2022-12-20 03:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1170/1519] eta 0:05:50 lr 0.000016 time 0.9250 (1.0057) model_time 0.9249 (1.0047) loss 0.8334 (0.8416) grad_norm 8.6558 (8.5977/1.8026) mem 68106MB [2022-12-20 03:01:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1180/1519] eta 0:05:40 lr 0.000016 time 0.9207 (1.0057) model_time 0.9206 (1.0047) loss 0.6694 (0.8416) grad_norm 8.3482 (8.5610/1.7236) mem 68106MB [2022-12-20 03:01:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1190/1519] eta 0:05:30 lr 0.000016 time 0.9196 (1.0057) model_time 0.9195 (1.0047) loss 0.7083 (0.8419) grad_norm 7.4827 (8.5533/1.7187) mem 68106MB [2022-12-20 03:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1200/1519] eta 0:05:20 lr 0.000016 time 0.9281 (1.0057) model_time 0.9280 (1.0047) loss 0.6844 (0.8412) grad_norm 11.4763 (8.5835/1.7160) mem 68106MB [2022-12-20 03:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1210/1519] eta 0:05:10 lr 0.000016 time 0.9290 (1.0057) model_time 0.9289 (1.0047) loss 1.0205 (0.8419) grad_norm 9.5456 (8.5447/1.6620) mem 68106MB [2022-12-20 03:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1220/1519] eta 0:05:00 lr 0.000016 time 0.9294 (1.0056) model_time 0.9293 (1.0046) loss 0.9901 (0.8416) grad_norm 7.5602 (8.5334/1.6601) mem 68106MB [2022-12-20 03:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1230/1519] eta 0:04:50 lr 0.000016 time 0.9263 (1.0056) model_time 0.9260 (1.0046) loss 0.7469 (0.8416) grad_norm 8.6004 (8.5335/1.6571) mem 68106MB [2022-12-20 03:02:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1240/1519] eta 0:04:40 lr 0.000016 time 0.9333 (1.0056) model_time 0.9332 (1.0046) loss 0.7766 (0.8418) grad_norm 12.2952 (8.5410/1.6451) mem 68106MB [2022-12-20 03:02:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1250/1519] eta 0:04:30 lr 0.000016 time 1.0100 (1.0056) model_time 1.0098 (1.0046) loss 0.7148 (0.8415) grad_norm 7.0741 (8.5536/1.6510) mem 68106MB [2022-12-20 03:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1260/1519] eta 0:04:20 lr 0.000016 time 0.9971 (1.0057) model_time 0.9970 (1.0048) loss 0.8114 (0.8410) grad_norm 9.4163 (8.5680/1.6429) mem 68106MB [2022-12-20 03:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1270/1519] eta 0:04:10 lr 0.000016 time 0.9206 (1.0057) model_time 0.9205 (1.0048) loss 0.8230 (0.8410) grad_norm 8.8441 (8.5739/1.6387) mem 68106MB [2022-12-20 03:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1280/1519] eta 0:04:00 lr 0.000016 time 0.9256 (1.0058) model_time 0.9253 (1.0048) loss 0.6753 (0.8408) grad_norm 10.1697 (8.6010/1.6609) mem 68106MB [2022-12-20 03:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1290/1519] eta 0:03:50 lr 0.000016 time 0.9259 (1.0058) model_time 0.9257 (1.0048) loss 0.7311 (0.8409) grad_norm 11.6603 (8.6286/1.6950) mem 68106MB [2022-12-20 03:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1300/1519] eta 0:03:40 lr 0.000016 time 0.9251 (1.0057) model_time 0.9249 (1.0048) loss 0.8660 (0.8411) grad_norm 6.2341 (8.6334/1.7043) mem 68106MB [2022-12-20 03:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1310/1519] eta 0:03:30 lr 0.000016 time 0.9284 (1.0058) model_time 0.9283 (1.0048) loss 0.9218 (0.8408) grad_norm 9.7544 (8.6476/1.7118) mem 68106MB [2022-12-20 03:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1320/1519] eta 0:03:20 lr 0.000016 time 1.0731 (1.0058) model_time 1.0730 (1.0049) loss 0.9259 (0.8416) grad_norm 8.4208 (8.6523/1.7115) mem 68106MB [2022-12-20 03:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1330/1519] eta 0:03:10 lr 0.000016 time 0.9265 (1.0060) model_time 0.9264 (1.0050) loss 0.7672 (0.8416) grad_norm 6.0158 (8.6519/1.7247) mem 68106MB [2022-12-20 03:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1340/1519] eta 0:03:00 lr 0.000016 time 0.9235 (1.0059) model_time 0.9234 (1.0050) loss 0.8800 (0.8411) grad_norm 7.4006 (8.6588/1.7269) mem 68106MB [2022-12-20 03:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1350/1519] eta 0:02:49 lr 0.000016 time 0.9253 (1.0059) model_time 0.9251 (1.0049) loss 0.9368 (0.8414) grad_norm 5.9524 (8.6516/1.7321) mem 68106MB [2022-12-20 03:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1360/1519] eta 0:02:39 lr 0.000016 time 0.9890 (1.0059) model_time 0.9889 (1.0049) loss 0.7024 (0.8409) grad_norm 6.3134 (8.6455/1.7302) mem 68106MB [2022-12-20 03:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1370/1519] eta 0:02:29 lr 0.000016 time 0.9396 (1.0058) model_time 0.9395 (1.0049) loss 0.7648 (0.8408) grad_norm 9.4825 (8.6407/1.7035) mem 68106MB [2022-12-20 03:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1380/1519] eta 0:02:19 lr 0.000016 time 0.9235 (1.0058) model_time 0.9234 (1.0049) loss 0.7492 (0.8409) grad_norm 8.6220 (8.6633/1.7144) mem 68106MB [2022-12-20 03:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1390/1519] eta 0:02:09 lr 0.000016 time 0.9254 (1.0058) model_time 0.9253 (1.0049) loss 0.7281 (0.8409) grad_norm 7.6135 (8.6616/1.7448) mem 68106MB [2022-12-20 03:04:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1400/1519] eta 0:01:59 lr 0.000016 time 0.9209 (1.0058) model_time 0.9208 (1.0049) loss 0.6781 (0.8403) grad_norm 8.9330 (8.6498/1.7490) mem 68106MB [2022-12-20 03:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1410/1519] eta 0:01:49 lr 0.000016 time 0.9259 (1.0058) model_time 0.9257 (1.0049) loss 0.9009 (0.8401) grad_norm 8.0968 (8.6358/1.7425) mem 68106MB [2022-12-20 03:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1420/1519] eta 0:01:39 lr 0.000016 time 0.9320 (1.0057) model_time 0.9318 (1.0048) loss 0.9441 (0.8399) grad_norm 6.9371 (8.6174/1.7522) mem 68106MB [2022-12-20 03:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1430/1519] eta 0:01:29 lr 0.000016 time 0.9205 (1.0058) model_time 0.9204 (1.0049) loss 1.2487 (0.8399) grad_norm 8.2610 (8.5737/1.7484) mem 68106MB [2022-12-20 03:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1440/1519] eta 0:01:19 lr 0.000016 time 0.9258 (1.0058) model_time 0.9256 (1.0049) loss 0.6727 (0.8400) grad_norm 8.0271 (8.5620/1.7521) mem 68106MB [2022-12-20 03:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1450/1519] eta 0:01:09 lr 0.000016 time 0.9215 (1.0058) model_time 0.9214 (1.0049) loss 0.8461 (0.8397) grad_norm 10.9545 (8.5676/1.7512) mem 68106MB [2022-12-20 03:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1460/1519] eta 0:00:59 lr 0.000016 time 0.9143 (1.0062) model_time 0.9142 (1.0053) loss 0.7425 (0.8399) grad_norm 6.2087 (8.5338/1.7711) mem 68106MB [2022-12-20 03:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1470/1519] eta 0:00:49 lr 0.000016 time 0.9234 (1.0061) model_time 0.9231 (1.0052) loss 0.7507 (0.8399) grad_norm 10.0953 (8.5082/1.7459) mem 68106MB [2022-12-20 03:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1480/1519] eta 0:00:39 lr 0.000016 time 0.9285 (1.0061) model_time 0.9284 (1.0052) loss 0.7165 (0.8401) grad_norm 11.1730 (8.5107/1.7458) mem 68106MB [2022-12-20 03:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1490/1519] eta 0:00:29 lr 0.000016 time 0.9047 (1.0061) model_time 0.9046 (1.0052) loss 0.6700 (0.8399) grad_norm 6.6498 (8.5012/1.7440) mem 68106MB [2022-12-20 03:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1500/1519] eta 0:00:19 lr 0.000016 time 1.0444 (1.0061) model_time 1.0443 (1.0052) loss 0.9907 (0.8398) grad_norm 9.5623 (8.4927/1.7375) mem 68106MB [2022-12-20 03:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [51/100][1510/1519] eta 0:00:09 lr 0.000016 time 0.9303 (1.0061) model_time 0.9302 (1.0052) loss 0.6924 (0.8400) grad_norm 9.3317 (8.5249/1.7764) mem 68106MB [2022-12-20 03:06:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 51 training takes 0:25:28 [2022-12-20 03:06:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_51.pth saving...... [2022-12-20 03:07:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_51.pth saved !!! [2022-12-20 03:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.681 (0.681) Loss 0.5227 (0.5227) Acc@1 90.972 (90.972) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 03:07:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.331) Loss 0.5109 (0.4970) Acc@1 92.708 (92.203) Acc@5 97.917 (98.674) Mem 68106MB [2022-12-20 03:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.302 (0.315) Loss 0.4723 (0.4950) Acc@1 92.361 (92.278) Acc@5 99.306 (98.528) Mem 68106MB [2022-12-20 03:07:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.5929 (0.5009) Acc@1 89.931 (91.935) Acc@5 97.569 (98.398) Mem 68106MB [2022-12-20 03:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.307) Loss 0.4576 (0.4910) Acc@1 92.708 (92.090) Acc@5 98.958 (98.493) Mem 68106MB [2022-12-20 03:07:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.304 (0.305) Loss 0.4665 (0.4879) Acc@1 91.667 (92.170) Acc@5 99.653 (98.523) Mem 68106MB [2022-12-20 03:07:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.306 (0.304) Loss 0.5042 (0.4877) Acc@1 90.972 (92.133) Acc@5 97.917 (98.497) Mem 68106MB [2022-12-20 03:07:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5361 (0.4889) Acc@1 93.403 (92.102) Acc@5 98.611 (98.508) Mem 68106MB [2022-12-20 03:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4269 (0.4872) Acc@1 93.056 (92.147) Acc@5 98.611 (98.543) Mem 68106MB [2022-12-20 03:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:51] * Acc@1 92.113 Acc@5 98.543 [2022-12-20 03:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.1% [2022-12-20 03:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.17% [2022-12-20 03:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][0/1519] eta 0:45:22 lr 0.000016 time 1.7922 (1.7922) model_time 1.0500 (1.0500) loss 0.7416 (0.7416) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 03:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][10/1519] eta 0:26:57 lr 0.000016 time 0.9326 (1.0721) model_time 0.9321 (1.0041) loss 0.7290 (0.9163) grad_norm 11.9138 (8.3143/1.9174) mem 68106MB [2022-12-20 03:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][20/1519] eta 0:25:57 lr 0.000016 time 0.9334 (1.0388) model_time 0.9332 (1.0030) loss 0.7757 (0.8873) grad_norm 8.4755 (8.1438/1.4436) mem 68106MB [2022-12-20 03:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][30/1519] eta 0:25:30 lr 0.000016 time 0.9816 (1.0275) model_time 0.9814 (1.0032) loss 1.0341 (0.8789) grad_norm 6.2516 (8.3634/1.5315) mem 68106MB [2022-12-20 03:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][40/1519] eta 0:25:11 lr 0.000016 time 0.9348 (1.0219) model_time 0.9347 (1.0034) loss 0.7294 (0.8756) grad_norm 7.5924 (8.3493/1.3462) mem 68106MB [2022-12-20 03:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][50/1519] eta 0:24:55 lr 0.000016 time 0.9269 (1.0184) model_time 0.9267 (1.0034) loss 0.7180 (0.8584) grad_norm 11.6616 (8.4726/1.7180) mem 68106MB [2022-12-20 03:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][60/1519] eta 0:24:43 lr 0.000016 time 0.9870 (1.0169) model_time 0.9869 (1.0043) loss 0.6886 (0.8522) grad_norm 8.7804 (8.4844/1.6302) mem 68106MB [2022-12-20 03:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][70/1519] eta 0:24:30 lr 0.000016 time 0.9411 (1.0150) model_time 0.9410 (1.0042) loss 0.9103 (0.8388) grad_norm 7.2487 (8.5002/1.6151) mem 68106MB [2022-12-20 03:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][80/1519] eta 0:24:19 lr 0.000016 time 0.9413 (1.0143) model_time 0.9411 (1.0047) loss 0.7350 (0.8363) grad_norm 9.7592 (8.6348/1.6819) mem 68106MB [2022-12-20 03:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][90/1519] eta 0:24:07 lr 0.000016 time 0.9317 (1.0126) model_time 0.9315 (1.0041) loss 0.6937 (0.8396) grad_norm 8.9324 (8.5899/1.6431) mem 68106MB [2022-12-20 03:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][100/1519] eta 0:23:56 lr 0.000016 time 0.9320 (1.0124) model_time 0.9318 (1.0046) loss 0.9695 (0.8391) grad_norm 7.9586 (8.6216/1.6041) mem 68106MB [2022-12-20 03:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][110/1519] eta 0:23:45 lr 0.000016 time 0.9358 (1.0114) model_time 0.9356 (1.0043) loss 0.7757 (0.8343) grad_norm 9.4729 (8.6367/1.5972) mem 68106MB [2022-12-20 03:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][120/1519] eta 0:23:37 lr 0.000016 time 0.9417 (1.0135) model_time 0.9416 (1.0070) loss 0.7591 (0.8356) grad_norm 7.8458 (8.6793/1.6352) mem 68106MB [2022-12-20 03:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][130/1519] eta 0:23:26 lr 0.000016 time 0.9283 (1.0127) model_time 0.9282 (1.0066) loss 0.7129 (0.8339) grad_norm 7.0255 (8.6425/1.5917) mem 68106MB [2022-12-20 03:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][140/1519] eta 0:23:15 lr 0.000016 time 0.9311 (1.0119) model_time 0.9308 (1.0063) loss 0.8280 (0.8336) grad_norm 8.1905 (8.6281/1.5439) mem 68106MB [2022-12-20 03:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][150/1519] eta 0:23:05 lr 0.000016 time 1.0265 (1.0121) model_time 1.0264 (1.0068) loss 0.7200 (0.8308) grad_norm 8.7703 (8.5990/1.5101) mem 68106MB [2022-12-20 03:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][160/1519] eta 0:22:54 lr 0.000016 time 0.9321 (1.0113) model_time 0.9319 (1.0063) loss 1.1297 (0.8302) grad_norm 8.3090 (8.5494/1.4942) mem 68106MB [2022-12-20 03:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][170/1519] eta 0:22:43 lr 0.000016 time 0.9315 (1.0108) model_time 0.9314 (1.0061) loss 1.0714 (0.8363) grad_norm 9.1545 (8.5552/1.4702) mem 68106MB [2022-12-20 03:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][180/1519] eta 0:22:32 lr 0.000016 time 0.9329 (1.0102) model_time 0.9327 (1.0057) loss 1.0952 (0.8356) grad_norm 9.8546 (8.5458/1.4635) mem 68106MB [2022-12-20 03:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][190/1519] eta 0:22:21 lr 0.000016 time 0.9319 (1.0097) model_time 0.9317 (1.0054) loss 1.0088 (0.8407) grad_norm 7.5884 (8.5122/1.4429) mem 68106MB [2022-12-20 03:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][200/1519] eta 0:22:11 lr 0.000016 time 0.9284 (1.0091) model_time 0.9283 (1.0051) loss 1.0887 (0.8453) grad_norm 6.5291 (8.5402/1.4950) mem 68106MB [2022-12-20 03:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][210/1519] eta 0:22:00 lr 0.000016 time 0.9868 (1.0090) model_time 0.9866 (1.0051) loss 0.7083 (0.8442) grad_norm 7.0743 (8.5144/1.4833) mem 68106MB [2022-12-20 03:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][220/1519] eta 0:21:50 lr 0.000016 time 0.9373 (1.0086) model_time 0.9372 (1.0048) loss 0.7043 (0.8441) grad_norm 14.5621 (8.5172/1.5878) mem 68106MB [2022-12-20 03:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][230/1519] eta 0:21:40 lr 0.000016 time 0.9321 (1.0086) model_time 0.9320 (1.0050) loss 0.7459 (0.8430) grad_norm 6.6925 (8.5085/1.5762) mem 68106MB [2022-12-20 03:11:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][240/1519] eta 0:21:29 lr 0.000016 time 0.9924 (1.0086) model_time 0.9923 (1.0051) loss 1.0397 (0.8418) grad_norm 8.1734 (8.4887/1.5571) mem 68106MB [2022-12-20 03:12:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][250/1519] eta 0:21:19 lr 0.000016 time 0.9237 (1.0086) model_time 0.9236 (1.0052) loss 1.0522 (0.8453) grad_norm 8.0719 (8.5725/1.6437) mem 68106MB [2022-12-20 03:12:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][260/1519] eta 0:21:09 lr 0.000016 time 0.9368 (1.0087) model_time 0.9367 (1.0054) loss 0.7358 (0.8450) grad_norm 8.2104 (8.5945/1.6571) mem 68106MB [2022-12-20 03:12:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][270/1519] eta 0:20:59 lr 0.000016 time 0.9330 (1.0083) model_time 0.9328 (1.0051) loss 0.7125 (0.8474) grad_norm 5.9349 (8.6398/1.7708) mem 68106MB [2022-12-20 03:12:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][280/1519] eta 0:20:48 lr 0.000016 time 0.9313 (1.0079) model_time 0.9311 (1.0049) loss 0.8975 (0.8449) grad_norm 8.9905 (8.6217/1.7612) mem 68106MB [2022-12-20 03:12:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][290/1519] eta 0:20:38 lr 0.000016 time 0.9320 (1.0077) model_time 0.9319 (1.0048) loss 0.7123 (0.8432) grad_norm 18.0090 (8.7364/1.9901) mem 68106MB [2022-12-20 03:12:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][300/1519] eta 0:20:28 lr 0.000016 time 0.9954 (1.0080) model_time 0.9952 (1.0051) loss 0.7111 (0.8413) grad_norm 7.3488 (8.7224/1.9702) mem 68106MB [2022-12-20 03:13:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][310/1519] eta 0:20:18 lr 0.000016 time 0.9339 (1.0082) model_time 0.9337 (1.0054) loss 0.6929 (0.8401) grad_norm 8.7129 (8.7163/1.9771) mem 68106MB [2022-12-20 03:13:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][320/1519] eta 0:20:08 lr 0.000016 time 0.9309 (1.0079) model_time 0.9307 (1.0052) loss 0.9677 (0.8413) grad_norm 10.9062 (8.7021/1.9632) mem 68106MB [2022-12-20 03:13:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][330/1519] eta 0:19:58 lr 0.000016 time 0.9347 (1.0077) model_time 0.9346 (1.0051) loss 1.0812 (0.8416) grad_norm 8.9731 (8.6737/1.9666) mem 68106MB [2022-12-20 03:13:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][340/1519] eta 0:19:47 lr 0.000016 time 0.9262 (1.0074) model_time 0.9260 (1.0049) loss 0.7238 (0.8408) grad_norm 8.7333 (8.6802/1.9404) mem 68106MB [2022-12-20 03:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][350/1519] eta 0:19:37 lr 0.000016 time 0.9308 (1.0074) model_time 0.9306 (1.0049) loss 0.6855 (0.8409) grad_norm 7.1078 (8.6849/1.9258) mem 68106MB [2022-12-20 03:13:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][360/1519] eta 0:19:27 lr 0.000016 time 0.9487 (1.0072) model_time 0.9486 (1.0047) loss 0.8327 (0.8401) grad_norm 8.6991 (8.6782/1.9086) mem 68106MB [2022-12-20 03:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][370/1519] eta 0:19:17 lr 0.000016 time 0.9332 (1.0070) model_time 0.9330 (1.0046) loss 0.7689 (0.8380) grad_norm 7.6950 (8.6819/1.8903) mem 68106MB [2022-12-20 03:14:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][380/1519] eta 0:19:06 lr 0.000016 time 0.9315 (1.0069) model_time 0.9313 (1.0046) loss 1.1560 (0.8380) grad_norm 7.7644 (8.6677/1.8695) mem 68106MB [2022-12-20 03:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][390/1519] eta 0:18:56 lr 0.000016 time 0.9340 (1.0069) model_time 0.9339 (1.0046) loss 0.8067 (0.8362) grad_norm 10.2262 (8.6714/1.8666) mem 68106MB [2022-12-20 03:14:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][400/1519] eta 0:18:46 lr 0.000016 time 0.9412 (1.0068) model_time 0.9410 (1.0045) loss 0.7048 (0.8365) grad_norm 13.5772 (8.7131/1.8846) mem 68106MB [2022-12-20 03:14:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][410/1519] eta 0:18:36 lr 0.000016 time 0.9347 (1.0070) model_time 0.9345 (1.0048) loss 0.9976 (0.8356) grad_norm 7.8431 (8.6961/1.8784) mem 68106MB [2022-12-20 03:14:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][420/1519] eta 0:18:26 lr 0.000016 time 0.9610 (1.0070) model_time 0.9608 (1.0048) loss 0.9109 (0.8358) grad_norm 11.5755 (8.6827/1.8846) mem 68106MB [2022-12-20 03:15:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][430/1519] eta 0:18:16 lr 0.000016 time 0.8968 (1.0070) model_time 0.8967 (1.0049) loss 0.8046 (0.8352) grad_norm 10.0290 (8.7016/1.9053) mem 68106MB [2022-12-20 03:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][440/1519] eta 0:18:06 lr 0.000016 time 0.9325 (1.0069) model_time 0.9323 (1.0048) loss 0.6983 (0.8351) grad_norm 8.8802 (8.7261/1.9021) mem 68106MB [2022-12-20 03:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][450/1519] eta 0:17:56 lr 0.000016 time 0.9365 (1.0068) model_time 0.9364 (1.0048) loss 0.9900 (0.8342) grad_norm 7.3132 (8.7170/1.8859) mem 68106MB [2022-12-20 03:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][460/1519] eta 0:17:46 lr 0.000016 time 0.9354 (1.0070) model_time 0.9352 (1.0050) loss 0.7102 (0.8343) grad_norm 8.7414 (8.7088/1.8788) mem 68106MB [2022-12-20 03:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][470/1519] eta 0:17:36 lr 0.000016 time 0.9300 (1.0070) model_time 0.9298 (1.0050) loss 0.7772 (0.8333) grad_norm 7.6177 (8.7238/1.8727) mem 68106MB [2022-12-20 03:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][480/1519] eta 0:17:26 lr 0.000016 time 0.9915 (1.0070) model_time 0.9913 (1.0051) loss 0.7442 (0.8317) grad_norm 7.2510 (8.7089/1.8676) mem 68106MB [2022-12-20 03:16:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][490/1519] eta 0:17:16 lr 0.000016 time 0.9325 (1.0069) model_time 0.9323 (1.0050) loss 1.0426 (0.8343) grad_norm 11.6131 (8.7115/1.8665) mem 68106MB [2022-12-20 03:16:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][500/1519] eta 0:17:05 lr 0.000016 time 0.9276 (1.0067) model_time 0.9274 (1.0049) loss 0.8924 (0.8352) grad_norm 9.0308 (8.6958/1.8597) mem 68106MB [2022-12-20 03:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][510/1519] eta 0:16:55 lr 0.000016 time 0.9353 (1.0066) model_time 0.9350 (1.0048) loss 1.0267 (0.8358) grad_norm 8.2086 (8.7226/1.8671) mem 68106MB [2022-12-20 03:16:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][520/1519] eta 0:16:45 lr 0.000016 time 0.9352 (1.0066) model_time 0.9350 (1.0048) loss 0.6909 (0.8346) grad_norm 10.0566 (8.7404/1.8709) mem 68106MB [2022-12-20 03:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][530/1519] eta 0:16:35 lr 0.000016 time 0.9317 (1.0067) model_time 0.9316 (1.0049) loss 0.8147 (0.8327) grad_norm 13.1994 (8.7579/1.8984) mem 68106MB [2022-12-20 03:16:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][540/1519] eta 0:16:25 lr 0.000016 time 0.9327 (1.0065) model_time 0.9326 (1.0048) loss 0.8597 (0.8329) grad_norm 8.4944 (8.7555/1.9251) mem 68106MB [2022-12-20 03:17:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][550/1519] eta 0:16:15 lr 0.000016 time 0.9243 (1.0065) model_time 0.9242 (1.0048) loss 0.9094 (0.8332) grad_norm 10.4159 (8.7454/1.9250) mem 68106MB [2022-12-20 03:17:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][560/1519] eta 0:16:05 lr 0.000016 time 0.9302 (1.0065) model_time 0.9300 (1.0048) loss 0.7226 (0.8326) grad_norm 9.0131 (8.7507/1.9142) mem 68106MB [2022-12-20 03:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][570/1519] eta 0:15:55 lr 0.000016 time 0.8941 (1.0067) model_time 0.8940 (1.0050) loss 0.7005 (0.8320) grad_norm 8.2969 (8.7445/1.9021) mem 68106MB [2022-12-20 03:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][580/1519] eta 0:15:45 lr 0.000016 time 0.9252 (1.0065) model_time 0.9251 (1.0049) loss 1.0286 (0.8314) grad_norm 10.5527 (8.7416/1.9085) mem 68106MB [2022-12-20 03:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][590/1519] eta 0:15:34 lr 0.000016 time 0.9264 (1.0064) model_time 0.9263 (1.0047) loss 0.9636 (0.8319) grad_norm 8.5319 (8.7515/1.8960) mem 68106MB [2022-12-20 03:17:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][600/1519] eta 0:15:24 lr 0.000016 time 0.9192 (1.0062) model_time 0.9191 (1.0046) loss 0.7953 (0.8322) grad_norm 7.6517 (8.7305/1.8901) mem 68106MB [2022-12-20 03:18:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][610/1519] eta 0:15:14 lr 0.000016 time 0.9338 (1.0063) model_time 0.9337 (1.0047) loss 1.1672 (0.8323) grad_norm 6.8226 (8.7104/1.8891) mem 68106MB [2022-12-20 03:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][620/1519] eta 0:15:04 lr 0.000016 time 0.9284 (1.0063) model_time 0.9282 (1.0047) loss 1.0301 (0.8322) grad_norm 7.3488 (8.7091/1.8938) mem 68106MB [2022-12-20 03:18:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][630/1519] eta 0:14:54 lr 0.000016 time 0.9217 (1.0062) model_time 0.9216 (1.0046) loss 0.8386 (0.8319) grad_norm 9.1339 (8.7312/1.9057) mem 68106MB [2022-12-20 03:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][640/1519] eta 0:14:44 lr 0.000016 time 0.9290 (1.0065) model_time 0.9289 (1.0049) loss 0.9051 (0.8319) grad_norm 8.1400 (8.7225/1.9178) mem 68106MB [2022-12-20 03:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][650/1519] eta 0:14:34 lr 0.000016 time 0.9298 (1.0065) model_time 0.9296 (1.0050) loss 0.6933 (0.8316) grad_norm 6.7852 (8.7111/1.8964) mem 68106MB [2022-12-20 03:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][660/1519] eta 0:14:24 lr 0.000016 time 0.9307 (1.0064) model_time 0.9305 (1.0049) loss 0.6830 (0.8315) grad_norm 8.0048 (8.7188/1.9090) mem 68106MB [2022-12-20 03:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][670/1519] eta 0:14:14 lr 0.000016 time 0.9191 (1.0064) model_time 0.9190 (1.0049) loss 0.6955 (0.8318) grad_norm 8.1127 (8.7296/1.9343) mem 68106MB [2022-12-20 03:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][680/1519] eta 0:14:04 lr 0.000016 time 0.9329 (1.0064) model_time 0.9328 (1.0049) loss 1.0288 (0.8315) grad_norm 6.5818 (8.6993/1.9328) mem 68106MB [2022-12-20 03:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][690/1519] eta 0:13:54 lr 0.000016 time 0.9691 (1.0063) model_time 0.9689 (1.0049) loss 0.7321 (0.8307) grad_norm 7.1462 (8.7076/1.9470) mem 68106MB [2022-12-20 03:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][700/1519] eta 0:13:44 lr 0.000016 time 0.9238 (1.0063) model_time 0.9236 (1.0048) loss 0.9259 (0.8301) grad_norm 9.4479 (8.7179/1.9611) mem 68106MB [2022-12-20 03:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][710/1519] eta 0:13:34 lr 0.000016 time 0.9295 (1.0062) model_time 0.9293 (1.0048) loss 0.8780 (0.8308) grad_norm 6.5595 (8.7274/1.9709) mem 68106MB [2022-12-20 03:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][720/1519] eta 0:13:23 lr 0.000016 time 0.9344 (1.0062) model_time 0.9342 (1.0047) loss 0.8204 (0.8309) grad_norm 6.7812 (8.7136/1.9578) mem 68106MB [2022-12-20 03:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][730/1519] eta 0:13:13 lr 0.000016 time 0.9280 (1.0062) model_time 0.9278 (1.0049) loss 0.8141 (0.8308) grad_norm 6.4374 (8.7015/1.9664) mem 68106MB [2022-12-20 03:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][740/1519] eta 0:13:03 lr 0.000016 time 0.9275 (1.0062) model_time 0.9274 (1.0049) loss 0.6977 (0.8297) grad_norm 7.4781 (8.6988/1.9723) mem 68106MB [2022-12-20 03:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][750/1519] eta 0:12:53 lr 0.000016 time 0.9288 (1.0063) model_time 0.9287 (1.0049) loss 0.7893 (0.8300) grad_norm 7.2201 (8.7361/2.0338) mem 68106MB [2022-12-20 03:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][760/1519] eta 0:12:43 lr 0.000016 time 0.9635 (1.0062) model_time 0.9634 (1.0049) loss 0.7363 (0.8309) grad_norm 9.0204 (8.7454/2.0325) mem 68106MB [2022-12-20 03:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][770/1519] eta 0:12:33 lr 0.000016 time 0.9612 (1.0062) model_time 0.9610 (1.0049) loss 0.7262 (0.8312) grad_norm 8.8080 (8.7451/2.0297) mem 68106MB [2022-12-20 03:20:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][780/1519] eta 0:12:23 lr 0.000016 time 0.9257 (1.0061) model_time 0.9249 (1.0048) loss 0.8687 (0.8310) grad_norm 10.2845 (8.7483/2.0267) mem 68106MB [2022-12-20 03:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][790/1519] eta 0:12:13 lr 0.000016 time 0.9193 (1.0060) model_time 0.9192 (1.0047) loss 0.8940 (0.8317) grad_norm 6.7859 (8.7572/2.0258) mem 68106MB [2022-12-20 03:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][800/1519] eta 0:12:03 lr 0.000016 time 0.9286 (1.0060) model_time 0.9285 (1.0047) loss 0.7379 (0.8307) grad_norm 7.1186 (8.7370/2.0161) mem 68106MB [2022-12-20 03:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][810/1519] eta 0:11:53 lr 0.000016 time 0.9284 (1.0059) model_time 0.9281 (1.0046) loss 0.7733 (0.8306) grad_norm 10.4108 (8.7379/2.0168) mem 68106MB [2022-12-20 03:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][820/1519] eta 0:11:43 lr 0.000016 time 0.9287 (1.0058) model_time 0.9285 (1.0046) loss 0.6876 (0.8302) grad_norm 9.3111 (8.7319/1.9817) mem 68106MB [2022-12-20 03:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][830/1519] eta 0:11:32 lr 0.000016 time 0.9240 (1.0057) model_time 0.9238 (1.0045) loss 0.7942 (0.8309) grad_norm 7.4832 (8.7361/1.9819) mem 68106MB [2022-12-20 03:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][840/1519] eta 0:11:22 lr 0.000016 time 0.9081 (1.0058) model_time 0.9080 (1.0045) loss 0.9084 (0.8319) grad_norm 9.5794 (8.7286/1.9907) mem 68106MB [2022-12-20 03:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][850/1519] eta 0:11:12 lr 0.000016 time 0.9288 (1.0058) model_time 0.9286 (1.0045) loss 0.8817 (0.8325) grad_norm 8.5778 (8.7019/1.9609) mem 68106MB [2022-12-20 03:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][860/1519] eta 0:11:02 lr 0.000016 time 0.9637 (1.0058) model_time 0.9635 (1.0046) loss 1.0310 (0.8331) grad_norm 9.1019 (8.6851/1.9499) mem 68106MB [2022-12-20 03:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][870/1519] eta 0:10:52 lr 0.000016 time 0.9750 (1.0060) model_time 0.9748 (1.0048) loss 0.7616 (0.8336) grad_norm 8.0085 (8.6493/1.9019) mem 68106MB [2022-12-20 03:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][880/1519] eta 0:10:42 lr 0.000016 time 0.9059 (1.0061) model_time 0.9058 (1.0048) loss 0.7814 (0.8330) grad_norm 10.1943 (8.6681/1.9017) mem 68106MB [2022-12-20 03:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][890/1519] eta 0:10:32 lr 0.000016 time 0.9353 (1.0060) model_time 0.9351 (1.0048) loss 0.9967 (0.8331) grad_norm 9.9145 (8.6091/1.7820) mem 68106MB [2022-12-20 03:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][900/1519] eta 0:10:22 lr 0.000016 time 0.9282 (1.0060) model_time 0.9280 (1.0048) loss 1.1392 (0.8345) grad_norm 8.7539 (8.6155/1.7817) mem 68106MB [2022-12-20 03:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][910/1519] eta 0:10:12 lr 0.000016 time 0.9234 (1.0059) model_time 0.9232 (1.0047) loss 0.6709 (0.8332) grad_norm 10.3444 (8.6310/1.8036) mem 68106MB [2022-12-20 03:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][920/1519] eta 0:10:02 lr 0.000016 time 0.9241 (1.0059) model_time 0.9239 (1.0048) loss 0.7294 (0.8328) grad_norm 9.9410 (8.6413/1.8041) mem 68106MB [2022-12-20 03:23:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][930/1519] eta 0:09:52 lr 0.000016 time 0.9286 (1.0060) model_time 0.9284 (1.0049) loss 1.0029 (0.8328) grad_norm 6.6036 (8.6346/1.8027) mem 68106MB [2022-12-20 03:23:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][940/1519] eta 0:09:42 lr 0.000016 time 0.9272 (1.0060) model_time 0.9270 (1.0048) loss 1.0346 (0.8335) grad_norm 10.2706 (8.6338/1.8041) mem 68106MB [2022-12-20 03:23:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][950/1519] eta 0:09:32 lr 0.000016 time 0.9086 (1.0060) model_time 0.9084 (1.0049) loss 0.6759 (0.8331) grad_norm 10.4274 (8.6480/1.8042) mem 68106MB [2022-12-20 03:23:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][960/1519] eta 0:09:22 lr 0.000016 time 0.9269 (1.0061) model_time 0.9268 (1.0049) loss 0.6740 (0.8331) grad_norm 6.7079 (8.6393/1.8037) mem 68106MB [2022-12-20 03:24:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][970/1519] eta 0:09:12 lr 0.000016 time 0.9279 (1.0061) model_time 0.9277 (1.0049) loss 0.7731 (0.8327) grad_norm 7.5532 (8.6418/1.8082) mem 68106MB [2022-12-20 03:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][980/1519] eta 0:09:02 lr 0.000016 time 0.9333 (1.0060) model_time 0.9331 (1.0049) loss 0.9197 (0.8326) grad_norm 6.8366 (8.6344/1.8189) mem 68106MB [2022-12-20 03:24:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][990/1519] eta 0:08:52 lr 0.000016 time 0.9258 (1.0060) model_time 0.9255 (1.0048) loss 0.9769 (0.8333) grad_norm 10.5876 (8.6397/1.8248) mem 68106MB [2022-12-20 03:24:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1000/1519] eta 0:08:42 lr 0.000016 time 0.9356 (1.0059) model_time 0.9355 (1.0048) loss 1.0488 (0.8346) grad_norm 9.7334 (8.5942/1.8065) mem 68106MB [2022-12-20 03:24:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1010/1519] eta 0:08:32 lr 0.000016 time 0.9226 (1.0059) model_time 0.9225 (1.0048) loss 0.7048 (0.8341) grad_norm 8.7057 (8.6005/1.8082) mem 68106MB [2022-12-20 03:24:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1020/1519] eta 0:08:21 lr 0.000016 time 0.9307 (1.0059) model_time 0.9306 (1.0048) loss 1.0140 (0.8338) grad_norm 8.1096 (8.6252/1.8153) mem 68106MB [2022-12-20 03:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1030/1519] eta 0:08:11 lr 0.000016 time 0.9285 (1.0058) model_time 0.9283 (1.0047) loss 0.7367 (0.8335) grad_norm 10.4978 (8.6136/1.7992) mem 68106MB [2022-12-20 03:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1040/1519] eta 0:08:01 lr 0.000016 time 0.9259 (1.0059) model_time 0.9257 (1.0048) loss 1.0976 (0.8334) grad_norm 7.2588 (8.5904/1.7894) mem 68106MB [2022-12-20 03:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1050/1519] eta 0:07:51 lr 0.000016 time 0.9222 (1.0058) model_time 0.9220 (1.0047) loss 0.9678 (0.8336) grad_norm 6.3958 (8.5802/1.7979) mem 68106MB [2022-12-20 03:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1060/1519] eta 0:07:41 lr 0.000016 time 0.9298 (1.0060) model_time 0.9296 (1.0049) loss 0.9072 (0.8335) grad_norm 8.4677 (8.5985/1.8102) mem 68106MB [2022-12-20 03:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1070/1519] eta 0:07:31 lr 0.000016 time 0.9224 (1.0059) model_time 0.9223 (1.0048) loss 0.7565 (0.8335) grad_norm 10.0783 (8.5969/1.8205) mem 68106MB [2022-12-20 03:25:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1080/1519] eta 0:07:21 lr 0.000016 time 0.9303 (1.0059) model_time 0.9301 (1.0048) loss 0.7853 (0.8331) grad_norm 11.8529 (8.6098/1.8222) mem 68106MB [2022-12-20 03:26:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1090/1519] eta 0:07:11 lr 0.000016 time 0.9220 (1.0058) model_time 0.9219 (1.0048) loss 0.7280 (0.8331) grad_norm 9.6548 (8.6126/1.8289) mem 68106MB [2022-12-20 03:26:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1100/1519] eta 0:07:01 lr 0.000016 time 0.9607 (1.0059) model_time 0.9606 (1.0049) loss 0.7399 (0.8328) grad_norm 8.8566 (8.6402/1.8351) mem 68106MB [2022-12-20 03:26:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1110/1519] eta 0:06:51 lr 0.000016 time 0.9275 (1.0059) model_time 0.9274 (1.0048) loss 0.6704 (0.8329) grad_norm 10.2104 (8.6046/1.8216) mem 68106MB [2022-12-20 03:26:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1120/1519] eta 0:06:41 lr 0.000016 time 0.9227 (1.0058) model_time 0.9225 (1.0048) loss 0.8489 (0.8336) grad_norm 9.2128 (8.5746/1.8162) mem 68106MB [2022-12-20 03:26:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1130/1519] eta 0:06:31 lr 0.000016 time 0.9295 (1.0058) model_time 0.9293 (1.0048) loss 0.7782 (0.8331) grad_norm 13.4501 (8.5595/1.7989) mem 68106MB [2022-12-20 03:26:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1140/1519] eta 0:06:21 lr 0.000016 time 0.9327 (1.0058) model_time 0.9326 (1.0047) loss 0.7027 (0.8329) grad_norm 10.2606 (8.5521/1.7649) mem 68106MB [2022-12-20 03:27:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1150/1519] eta 0:06:11 lr 0.000016 time 0.9312 (1.0057) model_time 0.9311 (1.0047) loss 1.0928 (0.8326) grad_norm 8.0999 (8.5592/1.7505) mem 68106MB [2022-12-20 03:27:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1160/1519] eta 0:06:01 lr 0.000016 time 0.9231 (1.0057) model_time 0.9229 (1.0047) loss 1.0451 (0.8326) grad_norm 9.7045 (8.5507/1.7590) mem 68106MB [2022-12-20 03:27:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1170/1519] eta 0:05:50 lr 0.000016 time 0.9220 (1.0057) model_time 0.9219 (1.0047) loss 0.8835 (0.8335) grad_norm 5.2333 (8.5475/1.7896) mem 68106MB [2022-12-20 03:27:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1180/1519] eta 0:05:40 lr 0.000016 time 0.9202 (1.0058) model_time 0.9201 (1.0048) loss 0.7507 (0.8329) grad_norm 14.1657 (8.5679/1.8061) mem 68106MB [2022-12-20 03:27:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1190/1519] eta 0:05:30 lr 0.000016 time 0.9177 (1.0058) model_time 0.9176 (1.0048) loss 0.7290 (0.8326) grad_norm 8.9593 (8.5680/1.8086) mem 68106MB [2022-12-20 03:27:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1200/1519] eta 0:05:20 lr 0.000016 time 0.9273 (1.0058) model_time 0.9270 (1.0048) loss 0.8000 (0.8326) grad_norm 6.5849 (8.5836/1.8073) mem 68106MB [2022-12-20 03:28:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1210/1519] eta 0:05:10 lr 0.000016 time 0.9377 (1.0057) model_time 0.9376 (1.0048) loss 1.1342 (0.8331) grad_norm 7.0767 (8.6262/1.8548) mem 68106MB [2022-12-20 03:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1220/1519] eta 0:05:00 lr 0.000016 time 0.9271 (1.0057) model_time 0.9270 (1.0048) loss 0.9304 (0.8334) grad_norm 10.0171 (8.6317/1.8580) mem 68106MB [2022-12-20 03:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1230/1519] eta 0:04:50 lr 0.000016 time 0.9208 (1.0057) model_time 0.9206 (1.0047) loss 0.7859 (0.8339) grad_norm 11.8309 (8.6245/1.8612) mem 68106MB [2022-12-20 03:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1240/1519] eta 0:04:40 lr 0.000016 time 0.9698 (1.0058) model_time 0.9697 (1.0048) loss 1.0234 (0.8347) grad_norm 9.5761 (8.6596/1.8755) mem 68106MB [2022-12-20 03:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1250/1519] eta 0:04:30 lr 0.000016 time 0.9289 (1.0057) model_time 0.9287 (1.0048) loss 0.6927 (0.8340) grad_norm 5.6885 (8.6480/1.8794) mem 68106MB [2022-12-20 03:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1260/1519] eta 0:04:20 lr 0.000016 time 0.9254 (1.0057) model_time 0.9253 (1.0048) loss 0.8720 (0.8343) grad_norm 7.7840 (8.6366/1.8638) mem 68106MB [2022-12-20 03:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1270/1519] eta 0:04:10 lr 0.000016 time 0.9270 (1.0058) model_time 0.9268 (1.0048) loss 0.8725 (0.8341) grad_norm 11.9440 (8.6397/1.8469) mem 68106MB [2022-12-20 03:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1280/1519] eta 0:04:00 lr 0.000016 time 0.9808 (1.0058) model_time 0.9807 (1.0048) loss 0.6966 (0.8339) grad_norm 8.0415 (8.6606/1.8373) mem 68106MB [2022-12-20 03:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1290/1519] eta 0:03:50 lr 0.000016 time 0.9254 (1.0058) model_time 0.9253 (1.0048) loss 0.8317 (0.8341) grad_norm 8.2378 (8.6470/1.8230) mem 68106MB [2022-12-20 03:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1300/1519] eta 0:03:40 lr 0.000016 time 0.9303 (1.0057) model_time 0.9300 (1.0048) loss 1.0459 (0.8346) grad_norm 6.9145 (8.6380/1.8076) mem 68106MB [2022-12-20 03:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1310/1519] eta 0:03:30 lr 0.000016 time 0.9205 (1.0057) model_time 0.9204 (1.0047) loss 0.8345 (0.8351) grad_norm 10.1895 (8.6310/1.7904) mem 68106MB [2022-12-20 03:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1320/1519] eta 0:03:20 lr 0.000016 time 0.9551 (1.0056) model_time 0.9549 (1.0047) loss 0.7447 (0.8346) grad_norm 8.4220 (8.6352/1.7889) mem 68106MB [2022-12-20 03:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1330/1519] eta 0:03:10 lr 0.000016 time 0.9264 (1.0057) model_time 0.9262 (1.0047) loss 0.6732 (0.8342) grad_norm 9.7973 (8.6453/1.7826) mem 68106MB [2022-12-20 03:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1340/1519] eta 0:03:00 lr 0.000016 time 0.9250 (1.0056) model_time 0.9248 (1.0047) loss 0.9225 (0.8338) grad_norm 8.8211 (8.6496/1.7753) mem 68106MB [2022-12-20 03:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1350/1519] eta 0:02:49 lr 0.000016 time 0.9263 (1.0056) model_time 0.9261 (1.0047) loss 1.0916 (0.8340) grad_norm 7.9824 (8.6115/1.7197) mem 68106MB [2022-12-20 03:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1360/1519] eta 0:02:39 lr 0.000016 time 0.9212 (1.0058) model_time 0.9211 (1.0049) loss 1.0556 (0.8341) grad_norm 8.0089 (8.6163/1.7166) mem 68106MB [2022-12-20 03:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1370/1519] eta 0:02:29 lr 0.000016 time 0.9597 (1.0059) model_time 0.9595 (1.0050) loss 0.6728 (0.8341) grad_norm 8.5277 (8.6048/1.7202) mem 68106MB [2022-12-20 03:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1380/1519] eta 0:02:19 lr 0.000016 time 0.9193 (1.0059) model_time 0.9191 (1.0050) loss 0.8983 (0.8342) grad_norm 10.8225 (8.6118/1.7206) mem 68106MB [2022-12-20 03:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1390/1519] eta 0:02:09 lr 0.000016 time 0.9280 (1.0058) model_time 0.9279 (1.0049) loss 0.7769 (0.8343) grad_norm 11.4281 (8.6347/1.7386) mem 68106MB [2022-12-20 03:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1400/1519] eta 0:01:59 lr 0.000016 time 0.9300 (1.0058) model_time 0.9298 (1.0049) loss 1.1452 (0.8345) grad_norm 8.5782 (8.6742/1.7885) mem 68106MB [2022-12-20 03:31:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1410/1519] eta 0:01:49 lr 0.000016 time 0.9216 (1.0058) model_time 0.9214 (1.0049) loss 0.8567 (0.8347) grad_norm 10.9950 (8.6744/1.7969) mem 68106MB [2022-12-20 03:31:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1420/1519] eta 0:01:39 lr 0.000016 time 0.9755 (1.0059) model_time 0.9754 (1.0050) loss 0.7521 (0.8344) grad_norm 8.7991 (8.6687/1.7977) mem 68106MB [2022-12-20 03:31:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1430/1519] eta 0:01:29 lr 0.000016 time 0.9215 (1.0058) model_time 0.9214 (1.0049) loss 0.7936 (0.8347) grad_norm 14.2960 (8.6836/1.8244) mem 68106MB [2022-12-20 03:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1440/1519] eta 0:01:19 lr 0.000015 time 0.9227 (1.0057) model_time 0.9226 (1.0049) loss 1.2364 (0.8349) grad_norm 6.7484 (8.6913/1.8206) mem 68106MB [2022-12-20 03:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1450/1519] eta 0:01:09 lr 0.000015 time 0.9283 (1.0057) model_time 0.9281 (1.0048) loss 0.6708 (0.8348) grad_norm 7.7579 (8.6640/1.8302) mem 68106MB [2022-12-20 03:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1460/1519] eta 0:00:59 lr 0.000015 time 0.9234 (1.0057) model_time 0.9232 (1.0048) loss 1.0890 (0.8349) grad_norm 6.4737 (8.7094/2.0179) mem 68106MB [2022-12-20 03:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1470/1519] eta 0:00:49 lr 0.000015 time 0.9258 (1.0057) model_time 0.9257 (1.0048) loss 0.7042 (0.8350) grad_norm 8.3638 (8.7316/2.0211) mem 68106MB [2022-12-20 03:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1480/1519] eta 0:00:39 lr 0.000015 time 0.9250 (1.0056) model_time 0.9248 (1.0048) loss 1.2058 (0.8352) grad_norm 8.7639 (8.7161/2.0183) mem 68106MB [2022-12-20 03:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1490/1519] eta 0:00:29 lr 0.000015 time 0.9246 (1.0056) model_time 0.9245 (1.0047) loss 0.6785 (0.8349) grad_norm 7.7646 (8.7219/2.0193) mem 68106MB [2022-12-20 03:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1500/1519] eta 0:00:19 lr 0.000015 time 0.9215 (1.0056) model_time 0.9213 (1.0047) loss 0.8577 (0.8350) grad_norm 10.5068 (8.7046/2.0276) mem 68106MB [2022-12-20 03:33:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [52/100][1510/1519] eta 0:00:09 lr 0.000015 time 0.9260 (1.0057) model_time 0.9259 (1.0048) loss 0.8875 (0.8355) grad_norm 8.1225 (8.6990/1.9982) mem 68106MB [2022-12-20 03:33:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 52 training takes 0:25:27 [2022-12-20 03:33:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_52.pth saving...... [2022-12-20 03:33:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_52.pth saved !!! [2022-12-20 03:33:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.617 (0.617) Loss 0.5203 (0.5203) Acc@1 90.278 (90.278) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 03:33:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.326) Loss 0.5157 (0.4934) Acc@1 91.667 (92.172) Acc@5 98.611 (98.643) Mem 68106MB [2022-12-20 03:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.312) Loss 0.4690 (0.4911) Acc@1 91.667 (92.245) Acc@5 98.958 (98.495) Mem 68106MB [2022-12-20 03:33:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.301 (0.307) Loss 0.6087 (0.4979) Acc@1 89.583 (92.081) Acc@5 97.569 (98.376) Mem 68106MB [2022-12-20 03:33:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.305) Loss 0.4504 (0.4894) Acc@1 93.750 (92.192) Acc@5 98.958 (98.484) Mem 68106MB [2022-12-20 03:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.304) Loss 0.4742 (0.4863) Acc@1 92.014 (92.218) Acc@5 99.653 (98.529) Mem 68106MB [2022-12-20 03:33:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.303) Loss 0.4889 (0.4861) Acc@1 90.625 (92.185) Acc@5 97.917 (98.514) Mem 68106MB [2022-12-20 03:34:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5360 (0.4878) Acc@1 92.014 (92.092) Acc@5 97.917 (98.508) Mem 68106MB [2022-12-20 03:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.302) Loss 0.4287 (0.4867) Acc@1 92.361 (92.095) Acc@5 98.958 (98.551) Mem 68106MB [2022-12-20 03:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:52] * Acc@1 92.068 Acc@5 98.535 [2022-12-20 03:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.1% [2022-12-20 03:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.17% [2022-12-20 03:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][0/1519] eta 0:45:04 lr 0.000015 time 1.7806 (1.7806) model_time 1.0389 (1.0389) loss 0.7467 (0.7467) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 03:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][10/1519] eta 0:27:06 lr 0.000015 time 0.9885 (1.0776) model_time 0.9884 (1.0098) loss 0.7910 (0.8133) grad_norm 5.0426 (8.3621/2.2191) mem 68106MB [2022-12-20 03:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][20/1519] eta 0:26:18 lr 0.000015 time 0.9257 (1.0529) model_time 0.9256 (1.0173) loss 0.7817 (0.8275) grad_norm 6.8477 (8.2126/1.7426) mem 68106MB [2022-12-20 03:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][30/1519] eta 0:25:50 lr 0.000015 time 0.9333 (1.0412) model_time 0.9331 (1.0169) loss 1.0820 (0.8379) grad_norm 9.1733 (8.0674/1.6464) mem 68106MB [2022-12-20 03:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][40/1519] eta 0:25:29 lr 0.000015 time 0.9241 (1.0345) model_time 0.9240 (1.0160) loss 0.6880 (0.8306) grad_norm 8.1356 (8.4803/2.3741) mem 68106MB [2022-12-20 03:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][50/1519] eta 0:25:13 lr 0.000015 time 0.9483 (1.0303) model_time 0.9482 (1.0155) loss 0.6737 (0.8624) grad_norm 9.0072 (8.6440/2.2583) mem 68106MB [2022-12-20 03:35:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][60/1519] eta 0:24:56 lr 0.000015 time 0.9222 (1.0254) model_time 0.9220 (1.0129) loss 0.6683 (0.8473) grad_norm 12.4932 (8.6111/2.2772) mem 68106MB [2022-12-20 03:35:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][70/1519] eta 0:24:41 lr 0.000015 time 0.9262 (1.0223) model_time 0.9260 (1.0115) loss 0.9217 (0.8377) grad_norm 5.5355 (8.5123/2.2328) mem 68106MB [2022-12-20 03:35:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][80/1519] eta 0:24:27 lr 0.000015 time 0.9362 (1.0199) model_time 0.9360 (1.0104) loss 0.9086 (0.8388) grad_norm 10.4179 (8.6258/2.2352) mem 68106MB [2022-12-20 03:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][90/1519] eta 0:24:14 lr 0.000015 time 0.9320 (1.0178) model_time 0.9319 (1.0093) loss 0.7137 (0.8325) grad_norm 11.3496 (8.6404/2.1828) mem 68106MB [2022-12-20 03:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][100/1519] eta 0:24:01 lr 0.000015 time 0.9267 (1.0162) model_time 0.9265 (1.0085) loss 0.9257 (0.8307) grad_norm 6.9463 (8.6290/2.1115) mem 68106MB [2022-12-20 03:35:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][110/1519] eta 0:23:49 lr 0.000015 time 0.9199 (1.0146) model_time 0.9197 (1.0076) loss 0.7091 (0.8240) grad_norm 11.1768 (8.5489/2.1110) mem 68106MB [2022-12-20 03:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][120/1519] eta 0:23:36 lr 0.000015 time 0.9239 (1.0123) model_time 0.9237 (1.0058) loss 0.8733 (0.8288) grad_norm 6.5297 (8.4918/2.0680) mem 68106MB [2022-12-20 03:36:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][130/1519] eta 0:23:26 lr 0.000015 time 0.9249 (1.0126) model_time 0.9247 (1.0066) loss 0.8243 (0.8328) grad_norm 9.6869 (8.4930/2.0090) mem 68106MB [2022-12-20 03:36:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][140/1519] eta 0:23:15 lr 0.000015 time 0.9209 (1.0121) model_time 0.9207 (1.0065) loss 0.9116 (0.8324) grad_norm 12.6102 (8.5139/2.0290) mem 68106MB [2022-12-20 03:36:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][150/1519] eta 0:23:04 lr 0.000015 time 0.9278 (1.0116) model_time 0.9277 (1.0063) loss 0.8806 (0.8302) grad_norm 8.9841 (8.5231/1.9862) mem 68106MB [2022-12-20 03:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][160/1519] eta 0:22:54 lr 0.000015 time 1.0000 (1.0117) model_time 0.9999 (1.0068) loss 0.9296 (0.8287) grad_norm 7.6378 (8.4898/1.9839) mem 68106MB [2022-12-20 03:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][170/1519] eta 0:22:44 lr 0.000015 time 0.9208 (1.0117) model_time 0.9206 (1.0071) loss 1.0117 (0.8343) grad_norm 8.2229 (8.4944/1.9291) mem 68106MB [2022-12-20 03:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][180/1519] eta 0:22:33 lr 0.000015 time 0.9347 (1.0110) model_time 0.9345 (1.0066) loss 0.7550 (0.8356) grad_norm 9.6102 (8.4575/1.9074) mem 68106MB [2022-12-20 03:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][190/1519] eta 0:22:22 lr 0.000015 time 0.9247 (1.0105) model_time 0.9246 (1.0063) loss 0.9460 (0.8356) grad_norm 6.1197 (8.4280/1.8849) mem 68106MB [2022-12-20 03:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][200/1519] eta 0:22:11 lr 0.000015 time 0.9250 (1.0098) model_time 0.9248 (1.0058) loss 0.9234 (0.8372) grad_norm 11.2334 (8.4703/1.8767) mem 68106MB [2022-12-20 03:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][210/1519] eta 0:22:02 lr 0.000015 time 0.8875 (1.0102) model_time 0.8874 (1.0064) loss 0.8135 (0.8366) grad_norm 6.2135 (8.4534/1.9026) mem 68106MB [2022-12-20 03:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][220/1519] eta 0:21:51 lr 0.000015 time 0.9334 (1.0099) model_time 0.9332 (1.0062) loss 1.0103 (0.8372) grad_norm 6.6658 (8.4103/1.9058) mem 68106MB [2022-12-20 03:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][230/1519] eta 0:21:41 lr 0.000015 time 0.9348 (1.0096) model_time 0.9347 (1.0061) loss 0.7894 (0.8372) grad_norm 9.9393 (8.5410/2.3541) mem 68106MB [2022-12-20 03:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][240/1519] eta 0:21:30 lr 0.000015 time 0.9211 (1.0091) model_time 0.9210 (1.0057) loss 0.8922 (0.8370) grad_norm 9.1841 (8.5600/2.3569) mem 68106MB [2022-12-20 03:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][250/1519] eta 0:21:19 lr 0.000015 time 0.9177 (1.0086) model_time 0.9176 (1.0053) loss 0.7358 (0.8362) grad_norm 6.6221 (8.5567/2.3284) mem 68106MB [2022-12-20 03:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][260/1519] eta 0:21:09 lr 0.000015 time 0.9229 (1.0082) model_time 0.9228 (1.0050) loss 0.8490 (0.8366) grad_norm 6.9526 (8.5256/2.2933) mem 68106MB [2022-12-20 03:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][270/1519] eta 0:20:58 lr 0.000015 time 0.9217 (1.0077) model_time 0.9216 (1.0047) loss 1.0063 (0.8358) grad_norm 8.1422 (8.5030/2.2554) mem 68106MB [2022-12-20 03:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][280/1519] eta 0:20:48 lr 0.000015 time 0.9298 (1.0075) model_time 0.9297 (1.0045) loss 0.8364 (0.8362) grad_norm 8.2424 (8.5181/2.2224) mem 68106MB [2022-12-20 03:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][290/1519] eta 0:20:38 lr 0.000015 time 0.9232 (1.0074) model_time 0.9231 (1.0045) loss 0.8069 (0.8370) grad_norm 11.1269 (8.5300/2.2049) mem 68106MB [2022-12-20 03:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][300/1519] eta 0:20:27 lr 0.000015 time 0.9211 (1.0072) model_time 0.9210 (1.0044) loss 1.1215 (0.8367) grad_norm 8.4524 (8.5378/2.1792) mem 68106MB [2022-12-20 03:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][310/1519] eta 0:20:17 lr 0.000015 time 0.9227 (1.0068) model_time 0.9226 (1.0041) loss 0.6743 (0.8372) grad_norm 8.3384 (8.5082/2.1592) mem 68106MB [2022-12-20 03:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][320/1519] eta 0:20:06 lr 0.000015 time 0.9233 (1.0066) model_time 0.9231 (1.0040) loss 0.7792 (0.8363) grad_norm 8.9537 (8.5218/2.1587) mem 68106MB [2022-12-20 03:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][330/1519] eta 0:19:56 lr 0.000015 time 0.9269 (1.0067) model_time 0.9267 (1.0041) loss 0.7550 (0.8346) grad_norm 8.0030 (8.5332/2.1499) mem 68106MB [2022-12-20 03:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][340/1519] eta 0:19:46 lr 0.000015 time 0.9954 (1.0067) model_time 0.9952 (1.0042) loss 0.9155 (0.8339) grad_norm 9.1617 (8.5748/2.2206) mem 68106MB [2022-12-20 03:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][350/1519] eta 0:19:36 lr 0.000015 time 0.9211 (1.0068) model_time 0.9209 (1.0043) loss 0.7967 (0.8345) grad_norm 7.0170 (8.5159/2.2190) mem 68106MB [2022-12-20 03:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][360/1519] eta 0:19:26 lr 0.000015 time 0.9214 (1.0066) model_time 0.9212 (1.0043) loss 0.9231 (0.8353) grad_norm 9.4860 (8.5234/2.2053) mem 68106MB [2022-12-20 03:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][370/1519] eta 0:19:16 lr 0.000015 time 0.9176 (1.0066) model_time 0.9175 (1.0042) loss 0.7084 (0.8356) grad_norm 11.5667 (8.5629/2.2281) mem 68106MB [2022-12-20 03:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][380/1519] eta 0:19:06 lr 0.000015 time 0.9295 (1.0065) model_time 0.9293 (1.0042) loss 0.9492 (0.8339) grad_norm 7.2441 (8.5318/2.2161) mem 68106MB [2022-12-20 03:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][390/1519] eta 0:18:56 lr 0.000015 time 0.9390 (1.0067) model_time 0.9389 (1.0045) loss 1.1322 (0.8339) grad_norm 10.3897 (8.5460/2.2005) mem 68106MB [2022-12-20 03:40:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][400/1519] eta 0:18:46 lr 0.000015 time 0.9337 (1.0067) model_time 0.9336 (1.0045) loss 0.6887 (0.8331) grad_norm 7.6151 (8.5228/2.1856) mem 68106MB [2022-12-20 03:40:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][410/1519] eta 0:18:36 lr 0.000015 time 0.9205 (1.0065) model_time 0.9203 (1.0044) loss 0.7383 (0.8328) grad_norm 6.4295 (8.5018/2.1793) mem 68106MB [2022-12-20 03:41:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][420/1519] eta 0:18:25 lr 0.000015 time 0.9361 (1.0064) model_time 0.9359 (1.0043) loss 0.7068 (0.8322) grad_norm 5.4161 (8.4744/2.1672) mem 68106MB [2022-12-20 03:41:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][430/1519] eta 0:18:15 lr 0.000015 time 0.9258 (1.0062) model_time 0.9257 (1.0041) loss 0.8749 (0.8336) grad_norm 10.2800 (8.5342/2.1869) mem 68106MB [2022-12-20 03:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][440/1519] eta 0:18:05 lr 0.000015 time 0.9255 (1.0062) model_time 0.9253 (1.0042) loss 0.7525 (0.8348) grad_norm 7.5415 (8.5346/2.1718) mem 68106MB [2022-12-20 03:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][450/1519] eta 0:17:55 lr 0.000015 time 0.9271 (1.0064) model_time 0.9269 (1.0045) loss 0.6592 (0.8338) grad_norm 8.6410 (8.5398/2.1617) mem 68106MB [2022-12-20 03:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][460/1519] eta 0:17:45 lr 0.000015 time 0.9267 (1.0064) model_time 0.9265 (1.0044) loss 0.7736 (0.8321) grad_norm 10.4626 (8.5685/2.1735) mem 68106MB [2022-12-20 03:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][470/1519] eta 0:17:35 lr 0.000015 time 0.9268 (1.0066) model_time 0.9267 (1.0046) loss 0.7082 (0.8333) grad_norm 8.5799 (8.5439/2.1594) mem 68106MB [2022-12-20 03:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][480/1519] eta 0:17:25 lr 0.000015 time 0.9855 (1.0065) model_time 0.9854 (1.0046) loss 0.7392 (0.8332) grad_norm 7.2750 (8.5266/2.1425) mem 68106MB [2022-12-20 03:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][490/1519] eta 0:17:15 lr 0.000015 time 0.9225 (1.0065) model_time 0.9224 (1.0046) loss 1.0419 (0.8337) grad_norm 7.9147 (8.5074/2.1301) mem 68106MB [2022-12-20 03:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][500/1519] eta 0:17:05 lr 0.000015 time 0.9198 (1.0063) model_time 0.9197 (1.0045) loss 0.7138 (0.8335) grad_norm 8.4781 (8.4844/2.1179) mem 68106MB [2022-12-20 03:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][510/1519] eta 0:16:55 lr 0.000015 time 0.9225 (1.0064) model_time 0.9224 (1.0046) loss 0.9103 (0.8347) grad_norm 7.2659 (8.4868/2.1051) mem 68106MB [2022-12-20 03:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][520/1519] eta 0:16:45 lr 0.000015 time 0.9995 (1.0065) model_time 0.9993 (1.0048) loss 0.9256 (0.8361) grad_norm 7.5673 (8.4967/2.0946) mem 68106MB [2022-12-20 03:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][530/1519] eta 0:16:35 lr 0.000015 time 0.9375 (1.0066) model_time 0.9373 (1.0048) loss 0.8919 (0.8362) grad_norm 10.6328 (8.4965/2.0832) mem 68106MB [2022-12-20 03:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][540/1519] eta 0:16:25 lr 0.000015 time 0.9205 (1.0065) model_time 0.9203 (1.0048) loss 0.7675 (0.8358) grad_norm 8.5209 (8.5092/2.0785) mem 68106MB [2022-12-20 03:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][550/1519] eta 0:16:15 lr 0.000015 time 0.9279 (1.0065) model_time 0.9278 (1.0048) loss 0.7124 (0.8354) grad_norm 9.8949 (8.4958/2.0690) mem 68106MB [2022-12-20 03:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][560/1519] eta 0:16:05 lr 0.000015 time 0.9266 (1.0064) model_time 0.9265 (1.0048) loss 0.8259 (0.8344) grad_norm 11.9553 (8.5211/2.0650) mem 68106MB [2022-12-20 03:43:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][570/1519] eta 0:15:55 lr 0.000015 time 0.9309 (1.0063) model_time 0.9307 (1.0047) loss 0.6995 (0.8343) grad_norm 10.6998 (8.5434/2.0697) mem 68106MB [2022-12-20 03:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][580/1519] eta 0:15:44 lr 0.000015 time 0.9334 (1.0062) model_time 0.9333 (1.0046) loss 0.8234 (0.8349) grad_norm 9.5401 (8.5199/2.0684) mem 68106MB [2022-12-20 03:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][590/1519] eta 0:15:34 lr 0.000015 time 0.9221 (1.0061) model_time 0.9220 (1.0045) loss 0.7364 (0.8333) grad_norm 7.4106 (8.5238/2.0657) mem 68106MB [2022-12-20 03:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][600/1519] eta 0:15:24 lr 0.000015 time 0.9270 (1.0064) model_time 0.9269 (1.0048) loss 1.1915 (0.8352) grad_norm 7.6556 (8.5133/2.0566) mem 68106MB [2022-12-20 03:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][610/1519] eta 0:15:14 lr 0.000015 time 0.8932 (1.0063) model_time 0.8930 (1.0047) loss 0.6931 (0.8351) grad_norm 8.0907 (8.5191/2.0734) mem 68106MB [2022-12-20 03:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][620/1519] eta 0:15:04 lr 0.000015 time 0.9315 (1.0065) model_time 0.9313 (1.0049) loss 0.9014 (0.8363) grad_norm 7.0511 (8.5027/2.0743) mem 68106MB [2022-12-20 03:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][630/1519] eta 0:14:54 lr 0.000015 time 0.9212 (1.0064) model_time 0.9211 (1.0048) loss 0.7432 (0.8359) grad_norm 8.3956 (8.5083/2.0705) mem 68106MB [2022-12-20 03:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][640/1519] eta 0:14:44 lr 0.000015 time 0.9433 (1.0064) model_time 0.9432 (1.0049) loss 0.9216 (0.8363) grad_norm 5.4061 (8.4722/2.0220) mem 68106MB [2022-12-20 03:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][650/1519] eta 0:14:34 lr 0.000015 time 0.9343 (1.0066) model_time 0.9342 (1.0051) loss 0.7604 (0.8362) grad_norm 8.3994 (8.4478/2.0303) mem 68106MB [2022-12-20 03:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][660/1519] eta 0:14:24 lr 0.000015 time 1.0126 (1.0067) model_time 1.0125 (1.0052) loss 1.0647 (0.8375) grad_norm 7.2740 (8.4706/2.0298) mem 68106MB [2022-12-20 03:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][670/1519] eta 0:14:14 lr 0.000015 time 0.9281 (1.0068) model_time 0.9279 (1.0053) loss 0.8018 (0.8364) grad_norm 10.2270 (8.4644/2.0160) mem 68106MB [2022-12-20 03:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][680/1519] eta 0:14:04 lr 0.000015 time 0.9408 (1.0067) model_time 0.9406 (1.0053) loss 0.7645 (0.8363) grad_norm 7.5106 (8.4646/1.9925) mem 68106MB [2022-12-20 03:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][690/1519] eta 0:13:54 lr 0.000015 time 0.9493 (1.0067) model_time 0.9491 (1.0053) loss 0.7010 (0.8375) grad_norm 7.1045 (8.4646/1.9881) mem 68106MB [2022-12-20 03:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][700/1519] eta 0:13:44 lr 0.000015 time 0.9753 (1.0069) model_time 0.9752 (1.0055) loss 0.9678 (0.8373) grad_norm 7.0260 (8.4374/1.9945) mem 68106MB [2022-12-20 03:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][710/1519] eta 0:13:34 lr 0.000015 time 0.9322 (1.0068) model_time 0.9321 (1.0055) loss 0.8356 (0.8376) grad_norm 7.6572 (8.4743/1.9882) mem 68106MB [2022-12-20 03:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][720/1519] eta 0:13:24 lr 0.000015 time 0.9211 (1.0067) model_time 0.9210 (1.0053) loss 0.9935 (0.8373) grad_norm 8.8386 (8.4585/1.9831) mem 68106MB [2022-12-20 03:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][730/1519] eta 0:13:14 lr 0.000015 time 0.9242 (1.0066) model_time 0.9241 (1.0053) loss 0.7560 (0.8370) grad_norm 9.7830 (8.4644/1.9808) mem 68106MB [2022-12-20 03:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][740/1519] eta 0:13:04 lr 0.000015 time 0.9324 (1.0065) model_time 0.9322 (1.0052) loss 0.6715 (0.8365) grad_norm 7.8623 (8.4542/1.9648) mem 68106MB [2022-12-20 03:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][750/1519] eta 0:12:53 lr 0.000015 time 0.9285 (1.0065) model_time 0.9284 (1.0051) loss 0.7020 (0.8363) grad_norm 9.7941 (8.4474/1.9635) mem 68106MB [2022-12-20 03:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][760/1519] eta 0:12:43 lr 0.000015 time 0.9199 (1.0064) model_time 0.9198 (1.0051) loss 0.9162 (0.8362) grad_norm 6.4221 (8.4600/1.9542) mem 68106MB [2022-12-20 03:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][770/1519] eta 0:12:33 lr 0.000015 time 0.9290 (1.0063) model_time 0.9288 (1.0050) loss 0.6670 (0.8360) grad_norm 12.1660 (8.4841/1.9834) mem 68106MB [2022-12-20 03:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][780/1519] eta 0:12:23 lr 0.000015 time 0.9247 (1.0062) model_time 0.9246 (1.0050) loss 0.7046 (0.8367) grad_norm 9.2850 (8.5166/1.9988) mem 68106MB [2022-12-20 03:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][790/1519] eta 0:12:13 lr 0.000015 time 0.9238 (1.0061) model_time 0.9237 (1.0049) loss 0.9494 (0.8372) grad_norm 11.2554 (8.5188/2.0041) mem 68106MB [2022-12-20 03:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][800/1519] eta 0:12:03 lr 0.000015 time 0.9308 (1.0061) model_time 0.9290 (1.0048) loss 0.6905 (0.8370) grad_norm 10.1444 (8.4950/2.0022) mem 68106MB [2022-12-20 03:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][810/1519] eta 0:11:53 lr 0.000015 time 0.9272 (1.0060) model_time 0.9271 (1.0048) loss 0.9870 (0.8379) grad_norm 7.3445 (8.4967/1.9836) mem 68106MB [2022-12-20 03:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][820/1519] eta 0:11:43 lr 0.000015 time 0.9262 (1.0064) model_time 0.9261 (1.0052) loss 0.6561 (0.8373) grad_norm 6.9985 (8.4983/1.9727) mem 68106MB [2022-12-20 03:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][830/1519] eta 0:11:33 lr 0.000015 time 1.0886 (1.0065) model_time 1.0884 (1.0053) loss 0.9798 (0.8381) grad_norm 6.9123 (8.4307/1.7724) mem 68106MB [2022-12-20 03:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][840/1519] eta 0:11:23 lr 0.000015 time 0.9942 (1.0065) model_time 0.9941 (1.0053) loss 0.9624 (0.8386) grad_norm 10.4306 (8.4725/1.9123) mem 68106MB [2022-12-20 03:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][850/1519] eta 0:11:13 lr 0.000015 time 0.9205 (1.0064) model_time 0.9204 (1.0052) loss 0.7407 (0.8394) grad_norm 6.4144 (8.4580/1.9226) mem 68106MB [2022-12-20 03:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][860/1519] eta 0:11:03 lr 0.000015 time 0.9182 (1.0066) model_time 0.9180 (1.0053) loss 0.6932 (0.8383) grad_norm 7.0454 (8.4556/1.9273) mem 68106MB [2022-12-20 03:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][870/1519] eta 0:10:53 lr 0.000015 time 0.9342 (1.0065) model_time 0.9341 (1.0053) loss 0.7926 (0.8379) grad_norm 7.5427 (8.4503/1.9296) mem 68106MB [2022-12-20 03:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][880/1519] eta 0:10:43 lr 0.000015 time 0.9082 (1.0064) model_time 0.9081 (1.0053) loss 1.1984 (0.8381) grad_norm 9.4049 (8.4446/1.9384) mem 68106MB [2022-12-20 03:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][890/1519] eta 0:10:32 lr 0.000015 time 0.9352 (1.0063) model_time 0.9350 (1.0052) loss 0.9341 (0.8379) grad_norm 6.3593 (8.4445/1.9391) mem 68106MB [2022-12-20 03:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][900/1519] eta 0:10:22 lr 0.000015 time 0.9278 (1.0062) model_time 0.9277 (1.0051) loss 0.7180 (0.8383) grad_norm 9.3608 (8.4218/1.9414) mem 68106MB [2022-12-20 03:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][910/1519] eta 0:10:12 lr 0.000015 time 0.9682 (1.0063) model_time 0.9681 (1.0051) loss 0.6796 (0.8377) grad_norm 8.3268 (8.4501/1.9435) mem 68106MB [2022-12-20 03:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][920/1519] eta 0:10:02 lr 0.000015 time 0.9193 (1.0062) model_time 0.9191 (1.0050) loss 0.9327 (0.8379) grad_norm 6.5721 (8.4366/1.9287) mem 68106MB [2022-12-20 03:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][930/1519] eta 0:09:52 lr 0.000015 time 0.9197 (1.0061) model_time 0.9196 (1.0050) loss 0.9474 (0.8383) grad_norm 8.6864 (8.4440/1.9233) mem 68106MB [2022-12-20 03:49:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][940/1519] eta 0:09:42 lr 0.000015 time 0.9135 (1.0063) model_time 0.9134 (1.0051) loss 0.8816 (0.8383) grad_norm 6.4220 (8.4360/1.8843) mem 68106MB [2022-12-20 03:50:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][950/1519] eta 0:09:32 lr 0.000015 time 0.9222 (1.0062) model_time 0.9220 (1.0051) loss 1.0326 (0.8381) grad_norm 5.9501 (8.4724/1.8851) mem 68106MB [2022-12-20 03:50:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][960/1519] eta 0:09:22 lr 0.000015 time 0.9419 (1.0062) model_time 0.9417 (1.0051) loss 1.0545 (0.8383) grad_norm 11.0397 (8.4830/1.8819) mem 68106MB [2022-12-20 03:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][970/1519] eta 0:09:12 lr 0.000015 time 0.9269 (1.0062) model_time 0.9267 (1.0051) loss 0.8345 (0.8381) grad_norm 12.1604 (8.5085/1.9398) mem 68106MB [2022-12-20 03:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][980/1519] eta 0:09:02 lr 0.000015 time 0.9257 (1.0064) model_time 0.9255 (1.0053) loss 0.9169 (0.8377) grad_norm 11.1774 (8.5528/1.9435) mem 68106MB [2022-12-20 03:50:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][990/1519] eta 0:08:52 lr 0.000015 time 0.9289 (1.0063) model_time 0.9287 (1.0052) loss 0.7023 (0.8371) grad_norm 8.4366 (8.5572/1.9462) mem 68106MB [2022-12-20 03:50:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1000/1519] eta 0:08:42 lr 0.000015 time 0.9266 (1.0063) model_time 0.9264 (1.0052) loss 0.8490 (0.8363) grad_norm 7.3737 (8.5691/1.9388) mem 68106MB [2022-12-20 03:51:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1010/1519] eta 0:08:32 lr 0.000015 time 1.0728 (1.0064) model_time 1.0726 (1.0053) loss 0.9287 (0.8366) grad_norm 8.3682 (8.6043/1.9544) mem 68106MB [2022-12-20 03:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1020/1519] eta 0:08:22 lr 0.000015 time 0.9987 (1.0065) model_time 0.9986 (1.0054) loss 0.8028 (0.8368) grad_norm 11.4945 (8.6201/1.9707) mem 68106MB [2022-12-20 03:51:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1030/1519] eta 0:08:12 lr 0.000015 time 0.9272 (1.0065) model_time 0.9271 (1.0054) loss 0.7184 (0.8368) grad_norm 9.0669 (8.5963/1.9441) mem 68106MB [2022-12-20 03:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1040/1519] eta 0:08:02 lr 0.000015 time 0.9219 (1.0064) model_time 0.9217 (1.0053) loss 0.9007 (0.8366) grad_norm 9.8339 (8.6001/1.9555) mem 68106MB [2022-12-20 03:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1050/1519] eta 0:07:51 lr 0.000015 time 0.9215 (1.0063) model_time 0.9214 (1.0053) loss 0.6941 (0.8368) grad_norm 7.9864 (8.6171/1.9641) mem 68106MB [2022-12-20 03:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1060/1519] eta 0:07:41 lr 0.000015 time 0.9207 (1.0062) model_time 0.9206 (1.0052) loss 0.7788 (0.8367) grad_norm 6.4401 (8.5930/1.9484) mem 68106MB [2022-12-20 03:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1070/1519] eta 0:07:31 lr 0.000015 time 0.9218 (1.0062) model_time 0.9216 (1.0051) loss 0.8960 (0.8368) grad_norm 7.2175 (8.5945/1.9464) mem 68106MB [2022-12-20 03:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1080/1519] eta 0:07:21 lr 0.000015 time 0.9273 (1.0061) model_time 0.9271 (1.0051) loss 0.9250 (0.8377) grad_norm 11.8820 (8.6197/1.9533) mem 68106MB [2022-12-20 03:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1090/1519] eta 0:07:11 lr 0.000015 time 0.9809 (1.0061) model_time 0.9808 (1.0051) loss 0.8220 (0.8384) grad_norm 7.0028 (8.6505/1.9627) mem 68106MB [2022-12-20 03:52:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1100/1519] eta 0:07:01 lr 0.000015 time 0.9247 (1.0060) model_time 0.9246 (1.0050) loss 0.8193 (0.8386) grad_norm 9.2943 (8.6703/1.9592) mem 68106MB [2022-12-20 03:52:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1110/1519] eta 0:06:51 lr 0.000015 time 0.9206 (1.0060) model_time 0.9205 (1.0050) loss 0.8147 (0.8380) grad_norm 6.9057 (8.6537/1.9605) mem 68106MB [2022-12-20 03:52:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1120/1519] eta 0:06:41 lr 0.000015 time 0.9254 (1.0060) model_time 0.9252 (1.0050) loss 0.8283 (0.8379) grad_norm 6.7156 (8.6296/1.9683) mem 68106MB [2022-12-20 03:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1130/1519] eta 0:06:31 lr 0.000015 time 0.9187 (1.0059) model_time 0.9186 (1.0049) loss 0.8032 (0.8372) grad_norm 6.3864 (8.6309/1.9745) mem 68106MB [2022-12-20 03:53:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1140/1519] eta 0:06:21 lr 0.000015 time 0.9252 (1.0059) model_time 0.9251 (1.0049) loss 0.7798 (0.8374) grad_norm 7.6508 (8.5969/1.9669) mem 68106MB [2022-12-20 03:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1150/1519] eta 0:06:11 lr 0.000015 time 0.9284 (1.0059) model_time 0.9283 (1.0049) loss 0.8111 (0.8371) grad_norm 8.0434 (8.5933/1.9684) mem 68106MB [2022-12-20 03:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1160/1519] eta 0:06:01 lr 0.000015 time 0.9209 (1.0060) model_time 0.9208 (1.0050) loss 0.9381 (0.8372) grad_norm 9.9722 (8.5777/1.9712) mem 68106MB [2022-12-20 03:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1170/1519] eta 0:05:51 lr 0.000015 time 0.9227 (1.0060) model_time 0.9226 (1.0050) loss 1.2102 (0.8377) grad_norm 6.2865 (8.5411/1.9522) mem 68106MB [2022-12-20 03:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1180/1519] eta 0:05:41 lr 0.000015 time 0.9272 (1.0060) model_time 0.9271 (1.0050) loss 0.7167 (0.8376) grad_norm 8.1630 (8.5576/1.9330) mem 68106MB [2022-12-20 03:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1190/1519] eta 0:05:30 lr 0.000015 time 0.9320 (1.0060) model_time 0.9319 (1.0050) loss 0.6655 (0.8377) grad_norm 8.2817 (8.5551/1.9432) mem 68106MB [2022-12-20 03:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1200/1519] eta 0:05:20 lr 0.000015 time 0.9172 (1.0060) model_time 0.9171 (1.0050) loss 0.8860 (0.8373) grad_norm 7.9399 (8.5591/1.9360) mem 68106MB [2022-12-20 03:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1210/1519] eta 0:05:10 lr 0.000015 time 0.9279 (1.0059) model_time 0.9276 (1.0049) loss 1.0325 (0.8376) grad_norm 9.1817 (8.5574/1.9107) mem 68106MB [2022-12-20 03:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1220/1519] eta 0:05:00 lr 0.000015 time 0.9267 (1.0059) model_time 0.9265 (1.0049) loss 0.7379 (0.8374) grad_norm 13.1484 (8.6051/1.9284) mem 68106MB [2022-12-20 03:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1230/1519] eta 0:04:50 lr 0.000015 time 0.9342 (1.0059) model_time 0.9340 (1.0049) loss 0.7215 (0.8369) grad_norm 9.3301 (8.6277/1.9178) mem 68106MB [2022-12-20 03:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1240/1519] eta 0:04:40 lr 0.000015 time 0.9612 (1.0058) model_time 0.9610 (1.0049) loss 0.7899 (0.8367) grad_norm 8.5536 (8.6283/1.9229) mem 68106MB [2022-12-20 03:55:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1250/1519] eta 0:04:30 lr 0.000015 time 0.9331 (1.0058) model_time 0.9329 (1.0048) loss 0.7287 (0.8373) grad_norm 13.7209 (8.6686/1.9153) mem 68106MB [2022-12-20 03:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1260/1519] eta 0:04:20 lr 0.000015 time 0.9313 (1.0058) model_time 0.9311 (1.0049) loss 0.6796 (0.8369) grad_norm 6.6918 (8.6681/1.9387) mem 68106MB [2022-12-20 03:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1270/1519] eta 0:04:10 lr 0.000015 time 1.0289 (1.0059) model_time 1.0288 (1.0049) loss 0.9869 (0.8370) grad_norm 10.5745 (8.6945/1.9534) mem 68106MB [2022-12-20 03:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1280/1519] eta 0:04:00 lr 0.000015 time 0.9224 (1.0059) model_time 0.9222 (1.0049) loss 1.3086 (0.8384) grad_norm 7.8503 (8.6965/1.9595) mem 68106MB [2022-12-20 03:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1290/1519] eta 0:03:50 lr 0.000015 time 0.9198 (1.0059) model_time 0.9197 (1.0049) loss 0.8637 (0.8380) grad_norm 6.6423 (8.6959/1.9761) mem 68106MB [2022-12-20 03:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1300/1519] eta 0:03:40 lr 0.000015 time 0.9231 (1.0058) model_time 0.9230 (1.0049) loss 0.7518 (0.8384) grad_norm 9.7380 (8.7202/1.9695) mem 68106MB [2022-12-20 03:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1310/1519] eta 0:03:30 lr 0.000015 time 0.9274 (1.0058) model_time 0.9272 (1.0049) loss 0.9500 (0.8381) grad_norm 8.5026 (8.7498/1.9973) mem 68106MB [2022-12-20 03:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1320/1519] eta 0:03:20 lr 0.000015 time 0.9195 (1.0057) model_time 0.9194 (1.0048) loss 0.9123 (0.8380) grad_norm 10.6116 (8.7770/1.9935) mem 68106MB [2022-12-20 03:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1330/1519] eta 0:03:10 lr 0.000015 time 0.9149 (1.0061) model_time 0.9148 (1.0052) loss 0.8919 (0.8387) grad_norm 10.1426 (8.7923/1.9957) mem 68106MB [2022-12-20 03:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1340/1519] eta 0:03:00 lr 0.000015 time 0.9366 (1.0061) model_time 0.9364 (1.0052) loss 0.7936 (0.8388) grad_norm 8.7780 (8.7953/1.9932) mem 68106MB [2022-12-20 03:56:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1350/1519] eta 0:02:50 lr 0.000015 time 0.9297 (1.0061) model_time 0.9296 (1.0052) loss 0.7690 (0.8391) grad_norm 8.2029 (8.8261/2.0032) mem 68106MB [2022-12-20 03:56:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1360/1519] eta 0:02:39 lr 0.000015 time 0.9353 (1.0061) model_time 0.9352 (1.0052) loss 0.7025 (0.8388) grad_norm 7.7321 (8.8090/2.0062) mem 68106MB [2022-12-20 03:57:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1370/1519] eta 0:02:29 lr 0.000015 time 0.9206 (1.0061) model_time 0.9205 (1.0052) loss 1.3698 (0.8391) grad_norm 10.1172 (8.7999/1.9951) mem 68106MB [2022-12-20 03:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1380/1519] eta 0:02:19 lr 0.000015 time 0.9216 (1.0060) model_time 0.9214 (1.0052) loss 0.6982 (0.8390) grad_norm 6.8726 (8.7817/1.9726) mem 68106MB [2022-12-20 03:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1390/1519] eta 0:02:09 lr 0.000015 time 0.9345 (1.0060) model_time 0.9344 (1.0051) loss 0.6852 (0.8383) grad_norm 8.7270 (8.7996/1.9809) mem 68106MB [2022-12-20 03:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1400/1519] eta 0:01:59 lr 0.000015 time 0.9363 (1.0060) model_time 0.9362 (1.0051) loss 0.8735 (0.8383) grad_norm 13.4049 (8.8225/1.9850) mem 68106MB [2022-12-20 03:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1410/1519] eta 0:01:49 lr 0.000015 time 0.9337 (1.0059) model_time 0.9335 (1.0051) loss 0.7016 (0.8380) grad_norm 7.4471 (8.8270/1.9859) mem 68106MB [2022-12-20 03:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1420/1519] eta 0:01:39 lr 0.000015 time 0.9182 (1.0059) model_time 0.9181 (1.0050) loss 0.7523 (0.8384) grad_norm 6.4509 (8.8203/1.9928) mem 68106MB [2022-12-20 03:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1430/1519] eta 0:01:29 lr 0.000015 time 0.9319 (1.0058) model_time 0.9318 (1.0050) loss 0.7190 (0.8386) grad_norm 7.8575 (8.8388/1.9960) mem 68106MB [2022-12-20 03:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1440/1519] eta 0:01:19 lr 0.000015 time 0.9924 (1.0058) model_time 0.9922 (1.0050) loss 0.9525 (0.8384) grad_norm 6.8863 (8.7834/1.8632) mem 68106MB [2022-12-20 03:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1450/1519] eta 0:01:09 lr 0.000015 time 0.9296 (1.0058) model_time 0.9294 (1.0049) loss 0.8059 (0.8383) grad_norm 7.9636 (8.7838/1.8475) mem 68106MB [2022-12-20 03:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1460/1519] eta 0:00:59 lr 0.000015 time 0.9268 (1.0058) model_time 0.9266 (1.0049) loss 0.9984 (0.8386) grad_norm 8.1225 (8.8044/1.8349) mem 68106MB [2022-12-20 03:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1470/1519] eta 0:00:49 lr 0.000015 time 0.9363 (1.0058) model_time 0.9361 (1.0049) loss 0.7337 (0.8386) grad_norm 7.6462 (8.8246/1.8293) mem 68106MB [2022-12-20 03:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1480/1519] eta 0:00:39 lr 0.000015 time 1.1867 (1.0060) model_time 1.1866 (1.0051) loss 1.2252 (0.8388) grad_norm 6.5650 (8.8298/1.8323) mem 68106MB [2022-12-20 03:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1490/1519] eta 0:00:29 lr 0.000015 time 0.9288 (1.0059) model_time 0.9287 (1.0051) loss 0.9187 (0.8389) grad_norm 6.5024 (8.8192/1.8420) mem 68106MB [2022-12-20 03:59:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1500/1519] eta 0:00:19 lr 0.000015 time 0.9238 (1.0060) model_time 0.9236 (1.0051) loss 0.7001 (0.8391) grad_norm 9.4352 (8.8499/1.8293) mem 68106MB [2022-12-20 03:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [53/100][1510/1519] eta 0:00:09 lr 0.000015 time 0.9213 (1.0060) model_time 0.9212 (1.0052) loss 0.9729 (0.8389) grad_norm 8.3077 (8.8138/1.8322) mem 68106MB [2022-12-20 03:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 53 training takes 0:25:28 [2022-12-20 03:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_53.pth saving...... [2022-12-20 03:59:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_53.pth saved !!! [2022-12-20 04:00:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.649 (0.649) Loss 0.5143 (0.5143) Acc@1 90.972 (90.972) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 04:00:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.329) Loss 0.5123 (0.4954) Acc@1 92.361 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 04:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.313) Loss 0.4681 (0.4938) Acc@1 93.750 (92.460) Acc@5 98.958 (98.462) Mem 68106MB [2022-12-20 04:00:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.308) Loss 0.6179 (0.5003) Acc@1 88.889 (92.126) Acc@5 98.264 (98.398) Mem 68106MB [2022-12-20 04:00:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.306) Loss 0.4597 (0.4927) Acc@1 93.750 (92.234) Acc@5 98.958 (98.467) Mem 68106MB [2022-12-20 04:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.304) Loss 0.4918 (0.4902) Acc@1 91.319 (92.279) Acc@5 99.306 (98.529) Mem 68106MB [2022-12-20 04:00:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.303) Loss 0.5081 (0.4900) Acc@1 90.972 (92.276) Acc@5 97.917 (98.492) Mem 68106MB [2022-12-20 04:00:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5219 (0.4911) Acc@1 92.361 (92.205) Acc@5 98.264 (98.494) Mem 68106MB [2022-12-20 04:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.300 (0.302) Loss 0.4172 (0.4896) Acc@1 92.708 (92.211) Acc@5 98.611 (98.530) Mem 68106MB [2022-12-20 04:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:53] * Acc@1 92.191 Acc@5 98.531 [2022-12-20 04:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.2% [2022-12-20 04:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 04:00:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 04:00:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.19% [2022-12-20 04:00:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][0/1519] eta 0:34:18 lr 0.000015 time 1.3551 (1.3551) model_time 0.9397 (0.9397) loss 0.6746 (0.6746) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 04:01:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][10/1519] eta 0:26:00 lr 0.000015 time 0.9271 (1.0342) model_time 0.9270 (0.9962) loss 0.6993 (0.8309) grad_norm 9.2439 (9.1992/1.0890) mem 68106MB [2022-12-20 04:01:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][20/1519] eta 0:25:30 lr 0.000015 time 0.9928 (1.0212) model_time 0.9927 (1.0011) loss 0.8805 (0.8023) grad_norm 6.7009 (9.0670/1.6888) mem 68106MB [2022-12-20 04:01:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][30/1519] eta 0:25:08 lr 0.000015 time 0.9213 (1.0131) model_time 0.9212 (0.9993) loss 0.9846 (0.8294) grad_norm 8.1793 (8.5002/1.6448) mem 68106MB [2022-12-20 04:01:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][40/1519] eta 0:24:54 lr 0.000015 time 0.9265 (1.0108) model_time 0.9263 (1.0003) loss 0.6835 (0.8268) grad_norm 9.7422 (8.3294/1.5858) mem 68106MB [2022-12-20 04:01:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][50/1519] eta 0:24:44 lr 0.000015 time 0.9234 (1.0106) model_time 0.9233 (1.0021) loss 0.6794 (0.8161) grad_norm 7.8879 (8.2893/1.4830) mem 68106MB [2022-12-20 04:01:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][60/1519] eta 0:24:32 lr 0.000015 time 0.9219 (1.0093) model_time 0.9218 (1.0021) loss 0.6920 (0.8078) grad_norm 7.0166 (8.6279/1.9688) mem 68106MB [2022-12-20 04:02:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][70/1519] eta 0:24:22 lr 0.000015 time 1.0104 (1.0092) model_time 1.0103 (1.0029) loss 1.2342 (0.8161) grad_norm 7.8619 (8.5468/1.8549) mem 68106MB [2022-12-20 04:02:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][80/1519] eta 0:24:11 lr 0.000015 time 0.9274 (1.0087) model_time 0.9272 (1.0032) loss 0.9742 (0.8219) grad_norm 8.2861 (8.3733/1.8170) mem 68106MB [2022-12-20 04:02:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][90/1519] eta 0:23:59 lr 0.000015 time 0.9319 (1.0075) model_time 0.9317 (1.0026) loss 0.8438 (0.8212) grad_norm 8.3669 (8.4099/1.9237) mem 68106MB [2022-12-20 04:02:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][100/1519] eta 0:23:48 lr 0.000015 time 0.9238 (1.0067) model_time 0.9235 (1.0023) loss 0.7237 (0.8183) grad_norm 9.9776 (8.4501/1.8924) mem 68106MB [2022-12-20 04:02:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][110/1519] eta 0:23:38 lr 0.000015 time 0.9228 (1.0067) model_time 0.9227 (1.0026) loss 0.7346 (0.8176) grad_norm 7.3367 (8.4304/1.8188) mem 68106MB [2022-12-20 04:02:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][120/1519] eta 0:23:28 lr 0.000015 time 0.9267 (1.0066) model_time 0.9266 (1.0028) loss 0.7329 (0.8185) grad_norm 7.7594 (8.5157/1.9239) mem 68106MB [2022-12-20 04:03:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][130/1519] eta 0:23:18 lr 0.000015 time 0.9210 (1.0066) model_time 0.9209 (1.0031) loss 0.6908 (0.8194) grad_norm 9.2568 (8.6335/1.9632) mem 68106MB [2022-12-20 04:03:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][140/1519] eta 0:23:07 lr 0.000015 time 0.9292 (1.0064) model_time 0.9290 (1.0032) loss 0.8952 (0.8190) grad_norm 10.7983 (8.6877/1.9437) mem 68106MB [2022-12-20 04:03:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][150/1519] eta 0:22:56 lr 0.000015 time 0.9217 (1.0058) model_time 0.9216 (1.0027) loss 0.7058 (0.8193) grad_norm 7.3882 (8.6939/1.8903) mem 68106MB [2022-12-20 04:03:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][160/1519] eta 0:22:46 lr 0.000015 time 0.9247 (1.0056) model_time 0.9246 (1.0027) loss 0.6840 (0.8249) grad_norm 8.6064 (8.6449/1.8639) mem 68106MB [2022-12-20 04:03:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][170/1519] eta 0:22:36 lr 0.000015 time 0.9230 (1.0052) model_time 0.9229 (1.0025) loss 0.8955 (0.8271) grad_norm 5.4413 (8.6224/1.9041) mem 68106MB [2022-12-20 04:03:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][180/1519] eta 0:22:26 lr 0.000015 time 0.9295 (1.0053) model_time 0.9294 (1.0027) loss 0.6884 (0.8234) grad_norm 8.2051 (8.6780/2.0934) mem 68106MB [2022-12-20 04:04:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][190/1519] eta 0:22:15 lr 0.000015 time 0.9244 (1.0051) model_time 0.9243 (1.0026) loss 0.9871 (0.8246) grad_norm 9.0988 (8.6230/2.0643) mem 68106MB [2022-12-20 04:04:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][200/1519] eta 0:22:06 lr 0.000015 time 0.9218 (1.0055) model_time 0.9217 (1.0031) loss 0.7631 (0.8246) grad_norm 8.3570 (8.5698/2.0434) mem 68106MB [2022-12-20 04:04:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][210/1519] eta 0:21:55 lr 0.000015 time 0.9238 (1.0052) model_time 0.9237 (1.0029) loss 0.6663 (0.8229) grad_norm 6.8180 (8.5471/2.0178) mem 68106MB [2022-12-20 04:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][220/1519] eta 0:21:45 lr 0.000015 time 0.9263 (1.0049) model_time 0.9261 (1.0027) loss 0.7469 (0.8255) grad_norm 7.7616 (8.5900/2.0021) mem 68106MB [2022-12-20 04:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][230/1519] eta 0:21:35 lr 0.000015 time 0.9218 (1.0050) model_time 0.9217 (1.0029) loss 0.9285 (0.8255) grad_norm 10.7478 (8.5884/1.9809) mem 68106MB [2022-12-20 04:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][240/1519] eta 0:21:25 lr 0.000015 time 0.9048 (1.0050) model_time 0.9047 (1.0030) loss 0.6728 (0.8229) grad_norm 6.2707 (8.5111/1.9834) mem 68106MB [2022-12-20 04:05:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][250/1519] eta 0:21:15 lr 0.000015 time 0.9204 (1.0050) model_time 0.9203 (1.0030) loss 0.8133 (0.8231) grad_norm 8.9450 (8.5706/2.0281) mem 68106MB [2022-12-20 04:05:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][260/1519] eta 0:21:05 lr 0.000015 time 1.0219 (1.0052) model_time 1.0218 (1.0033) loss 0.7270 (0.8281) grad_norm 6.6472 (8.5596/2.0021) mem 68106MB [2022-12-20 04:05:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][270/1519] eta 0:20:55 lr 0.000015 time 0.9254 (1.0050) model_time 0.9252 (1.0032) loss 1.0527 (0.8268) grad_norm 7.8597 (8.5315/1.9822) mem 68106MB [2022-12-20 04:05:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][280/1519] eta 0:20:45 lr 0.000015 time 0.9263 (1.0049) model_time 0.9262 (1.0031) loss 0.9506 (0.8286) grad_norm 7.3235 (8.5266/1.9537) mem 68106MB [2022-12-20 04:05:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][290/1519] eta 0:20:35 lr 0.000015 time 0.9177 (1.0049) model_time 0.9176 (1.0032) loss 1.4244 (0.8306) grad_norm 9.5469 (8.5036/1.9379) mem 68106MB [2022-12-20 04:05:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][300/1519] eta 0:20:25 lr 0.000015 time 0.9243 (1.0052) model_time 0.9242 (1.0035) loss 0.9504 (0.8300) grad_norm 8.5879 (8.5196/1.9341) mem 68106MB [2022-12-20 04:06:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][310/1519] eta 0:20:15 lr 0.000015 time 0.9256 (1.0053) model_time 0.9255 (1.0036) loss 0.7640 (0.8307) grad_norm 6.7849 (8.5464/1.9354) mem 68106MB [2022-12-20 04:06:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][320/1519] eta 0:20:05 lr 0.000015 time 0.9210 (1.0051) model_time 0.9209 (1.0035) loss 0.9955 (0.8307) grad_norm 7.5098 (8.5668/1.9323) mem 68106MB [2022-12-20 04:06:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][330/1519] eta 0:19:54 lr 0.000015 time 0.9235 (1.0049) model_time 0.9233 (1.0033) loss 0.7343 (0.8287) grad_norm 6.3677 (8.5274/1.9239) mem 68106MB [2022-12-20 04:06:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][340/1519] eta 0:19:44 lr 0.000015 time 0.9241 (1.0049) model_time 0.9240 (1.0034) loss 0.8141 (0.8282) grad_norm 9.2223 (8.4936/1.9212) mem 68106MB [2022-12-20 04:06:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][350/1519] eta 0:19:34 lr 0.000015 time 0.9342 (1.0049) model_time 0.9339 (1.0034) loss 0.7679 (0.8306) grad_norm 5.9191 (8.4522/1.9202) mem 68106MB [2022-12-20 04:06:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][360/1519] eta 0:19:24 lr 0.000015 time 0.9207 (1.0047) model_time 0.9205 (1.0033) loss 0.9235 (0.8302) grad_norm 6.7523 (8.4381/1.9138) mem 68106MB [2022-12-20 04:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][370/1519] eta 0:19:14 lr 0.000015 time 1.0710 (1.0051) model_time 1.0709 (1.0037) loss 0.8180 (0.8303) grad_norm 9.5779 (8.4502/1.9344) mem 68106MB [2022-12-20 04:07:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][380/1519] eta 0:19:04 lr 0.000015 time 0.9287 (1.0051) model_time 0.9284 (1.0037) loss 0.7013 (0.8285) grad_norm 7.4104 (8.4622/1.9310) mem 68106MB [2022-12-20 04:07:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][390/1519] eta 0:18:54 lr 0.000015 time 0.9242 (1.0052) model_time 0.9240 (1.0038) loss 0.6564 (0.8283) grad_norm 9.1714 (8.4702/1.9285) mem 68106MB [2022-12-20 04:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][400/1519] eta 0:18:44 lr 0.000015 time 0.9457 (1.0051) model_time 0.9455 (1.0037) loss 0.7645 (0.8301) grad_norm 7.2721 (8.4415/1.9138) mem 68106MB [2022-12-20 04:07:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][410/1519] eta 0:18:34 lr 0.000015 time 0.9361 (1.0049) model_time 0.9360 (1.0036) loss 0.8095 (0.8312) grad_norm 6.5852 (8.4419/1.9021) mem 68106MB [2022-12-20 04:07:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][420/1519] eta 0:18:24 lr 0.000015 time 0.9368 (1.0052) model_time 0.9366 (1.0038) loss 0.7266 (0.8307) grad_norm 8.1158 (8.4082/1.8951) mem 68106MB [2022-12-20 04:08:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][430/1519] eta 0:18:14 lr 0.000015 time 0.9228 (1.0051) model_time 0.9227 (1.0038) loss 1.1991 (0.8313) grad_norm 6.4787 (8.4199/1.9251) mem 68106MB [2022-12-20 04:08:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][440/1519] eta 0:18:04 lr 0.000015 time 1.0335 (1.0053) model_time 1.0333 (1.0041) loss 0.6928 (0.8313) grad_norm 5.5999 (8.4187/1.9304) mem 68106MB [2022-12-20 04:08:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][450/1519] eta 0:17:54 lr 0.000015 time 0.9338 (1.0055) model_time 0.9337 (1.0042) loss 0.6715 (0.8312) grad_norm 8.5243 (8.4020/1.9153) mem 68106MB [2022-12-20 04:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][460/1519] eta 0:17:44 lr 0.000015 time 0.9303 (1.0055) model_time 0.9302 (1.0042) loss 1.0336 (0.8317) grad_norm 7.5369 (8.4002/1.9140) mem 68106MB [2022-12-20 04:08:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][470/1519] eta 0:17:34 lr 0.000015 time 0.9400 (1.0055) model_time 0.9398 (1.0043) loss 0.8668 (0.8320) grad_norm 7.7816 (8.3980/1.8976) mem 68106MB [2022-12-20 04:08:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][480/1519] eta 0:17:24 lr 0.000015 time 0.9299 (1.0055) model_time 0.9297 (1.0044) loss 0.6694 (0.8322) grad_norm 10.1676 (8.4010/1.8820) mem 68106MB [2022-12-20 04:09:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][490/1519] eta 0:17:14 lr 0.000015 time 0.9519 (1.0057) model_time 0.9517 (1.0045) loss 0.7743 (0.8316) grad_norm 8.2360 (8.4040/1.8645) mem 68106MB [2022-12-20 04:09:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][500/1519] eta 0:17:04 lr 0.000015 time 0.9397 (1.0055) model_time 0.9396 (1.0044) loss 0.6702 (0.8309) grad_norm 9.3080 (8.4345/1.8607) mem 68106MB [2022-12-20 04:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][510/1519] eta 0:16:54 lr 0.000015 time 0.9273 (1.0055) model_time 0.9272 (1.0043) loss 0.6963 (0.8300) grad_norm 6.9447 (8.4258/1.8484) mem 68106MB [2022-12-20 04:09:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][520/1519] eta 0:16:44 lr 0.000015 time 0.9369 (1.0053) model_time 0.9367 (1.0042) loss 0.6668 (0.8298) grad_norm 8.3025 (8.4227/1.8397) mem 68106MB [2022-12-20 04:09:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][530/1519] eta 0:16:34 lr 0.000015 time 0.9238 (1.0052) model_time 0.9236 (1.0041) loss 0.7349 (0.8290) grad_norm 10.0843 (8.4342/1.8383) mem 68106MB [2022-12-20 04:09:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][540/1519] eta 0:16:24 lr 0.000015 time 0.9261 (1.0052) model_time 0.9259 (1.0041) loss 0.7492 (0.8278) grad_norm 5.9780 (8.4326/1.8394) mem 68106MB [2022-12-20 04:10:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][550/1519] eta 0:16:14 lr 0.000015 time 0.9807 (1.0052) model_time 0.9806 (1.0041) loss 0.6788 (0.8274) grad_norm 7.6290 (8.4201/1.8322) mem 68106MB [2022-12-20 04:10:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][560/1519] eta 0:16:03 lr 0.000015 time 0.9358 (1.0052) model_time 0.9357 (1.0041) loss 0.6888 (0.8275) grad_norm 8.7146 (8.4499/1.8428) mem 68106MB [2022-12-20 04:10:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][570/1519] eta 0:15:53 lr 0.000015 time 0.9233 (1.0052) model_time 0.9231 (1.0042) loss 0.6694 (0.8286) grad_norm 8.6713 (8.5001/1.8837) mem 68106MB [2022-12-20 04:10:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][580/1519] eta 0:15:43 lr 0.000015 time 0.9315 (1.0051) model_time 0.9313 (1.0041) loss 1.0083 (0.8292) grad_norm 12.2453 (8.5156/1.8819) mem 68106MB [2022-12-20 04:10:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][590/1519] eta 0:15:33 lr 0.000015 time 0.9272 (1.0050) model_time 0.9266 (1.0040) loss 0.6928 (0.8303) grad_norm 7.7016 (8.5080/1.8683) mem 68106MB [2022-12-20 04:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][600/1519] eta 0:15:23 lr 0.000015 time 0.9227 (1.0050) model_time 0.9225 (1.0039) loss 0.7111 (0.8297) grad_norm 8.2174 (8.5048/1.8543) mem 68106MB [2022-12-20 04:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][610/1519] eta 0:15:13 lr 0.000015 time 0.9326 (1.0050) model_time 0.9324 (1.0040) loss 0.8244 (0.8292) grad_norm 8.9325 (8.4893/1.8531) mem 68106MB [2022-12-20 04:11:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][620/1519] eta 0:15:03 lr 0.000015 time 0.9035 (1.0050) model_time 0.9034 (1.0040) loss 0.6911 (0.8305) grad_norm 7.3184 (8.4725/1.8375) mem 68106MB [2022-12-20 04:11:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][630/1519] eta 0:14:53 lr 0.000015 time 0.9241 (1.0049) model_time 0.9240 (1.0039) loss 0.6963 (0.8310) grad_norm 7.5333 (8.5085/1.8622) mem 68106MB [2022-12-20 04:11:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][640/1519] eta 0:14:43 lr 0.000015 time 0.9391 (1.0049) model_time 0.9390 (1.0039) loss 0.6923 (0.8326) grad_norm 8.0121 (8.5204/1.8545) mem 68106MB [2022-12-20 04:11:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][650/1519] eta 0:14:33 lr 0.000015 time 0.9287 (1.0048) model_time 0.9284 (1.0038) loss 0.9618 (0.8316) grad_norm 7.8042 (8.5077/1.8593) mem 68106MB [2022-12-20 04:11:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][660/1519] eta 0:14:23 lr 0.000015 time 0.9193 (1.0048) model_time 0.9192 (1.0038) loss 0.7245 (0.8305) grad_norm 8.5564 (8.4709/1.8072) mem 68106MB [2022-12-20 04:12:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][670/1519] eta 0:14:13 lr 0.000015 time 0.9205 (1.0048) model_time 0.9203 (1.0038) loss 0.6939 (0.8301) grad_norm 6.0784 (8.4759/1.8117) mem 68106MB [2022-12-20 04:12:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][680/1519] eta 0:14:03 lr 0.000015 time 0.9278 (1.0048) model_time 0.9274 (1.0038) loss 0.6954 (0.8301) grad_norm 9.8474 (8.5018/1.8093) mem 68106MB [2022-12-20 04:12:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][690/1519] eta 0:13:53 lr 0.000015 time 0.9153 (1.0049) model_time 0.9151 (1.0040) loss 0.7833 (0.8306) grad_norm 6.9743 (8.4960/1.7853) mem 68106MB [2022-12-20 04:12:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][700/1519] eta 0:13:43 lr 0.000015 time 0.9317 (1.0049) model_time 0.9316 (1.0040) loss 0.8593 (0.8304) grad_norm 6.6997 (8.4737/1.7849) mem 68106MB [2022-12-20 04:12:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][710/1519] eta 0:13:32 lr 0.000015 time 0.9261 (1.0049) model_time 0.9259 (1.0040) loss 0.9146 (0.8306) grad_norm 7.0785 (8.4916/1.8110) mem 68106MB [2022-12-20 04:12:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][720/1519] eta 0:13:22 lr 0.000015 time 0.9246 (1.0048) model_time 0.9244 (1.0039) loss 0.8364 (0.8302) grad_norm 5.7452 (8.4586/1.7877) mem 68106MB [2022-12-20 04:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][730/1519] eta 0:13:13 lr 0.000015 time 0.8844 (1.0051) model_time 0.8842 (1.0042) loss 0.9877 (0.8315) grad_norm 8.7693 (8.4358/1.7636) mem 68106MB [2022-12-20 04:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][740/1519] eta 0:13:02 lr 0.000015 time 0.9242 (1.0050) model_time 0.9241 (1.0041) loss 0.6712 (0.8312) grad_norm 10.4508 (8.4342/1.7784) mem 68106MB [2022-12-20 04:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][750/1519] eta 0:12:52 lr 0.000015 time 0.9366 (1.0050) model_time 0.9364 (1.0041) loss 0.6825 (0.8315) grad_norm 9.3076 (8.4257/1.7814) mem 68106MB [2022-12-20 04:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][760/1519] eta 0:12:42 lr 0.000015 time 0.9572 (1.0051) model_time 0.9571 (1.0042) loss 1.0293 (0.8313) grad_norm 9.8099 (8.4319/1.7840) mem 68106MB [2022-12-20 04:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][770/1519] eta 0:12:32 lr 0.000015 time 0.9306 (1.0050) model_time 0.9305 (1.0041) loss 0.7084 (0.8301) grad_norm 9.1037 (8.4257/1.7669) mem 68106MB [2022-12-20 04:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][780/1519] eta 0:12:22 lr 0.000015 time 0.9263 (1.0049) model_time 0.9261 (1.0041) loss 0.7380 (0.8298) grad_norm 7.8541 (8.4016/1.6904) mem 68106MB [2022-12-20 04:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][790/1519] eta 0:12:12 lr 0.000015 time 0.9372 (1.0051) model_time 0.9370 (1.0042) loss 0.9068 (0.8297) grad_norm 13.6336 (8.4415/1.7277) mem 68106MB [2022-12-20 04:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][800/1519] eta 0:12:02 lr 0.000015 time 0.9346 (1.0051) model_time 0.9345 (1.0042) loss 1.2089 (0.8298) grad_norm 12.6622 (8.4652/1.7538) mem 68106MB [2022-12-20 04:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][810/1519] eta 0:11:52 lr 0.000015 time 0.9306 (1.0050) model_time 0.9304 (1.0042) loss 0.8022 (0.8295) grad_norm 7.7290 (8.5054/1.8061) mem 68106MB [2022-12-20 04:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][820/1519] eta 0:11:42 lr 0.000015 time 0.9271 (1.0050) model_time 0.9270 (1.0041) loss 0.7077 (0.8290) grad_norm 7.8703 (8.4984/1.8136) mem 68106MB [2022-12-20 04:14:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][830/1519] eta 0:11:32 lr 0.000015 time 0.9346 (1.0049) model_time 0.9344 (1.0040) loss 0.7139 (0.8299) grad_norm 7.3226 (8.5014/1.8149) mem 68106MB [2022-12-20 04:14:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][840/1519] eta 0:11:22 lr 0.000015 time 0.9276 (1.0048) model_time 0.9275 (1.0040) loss 0.8939 (0.8293) grad_norm 7.9056 (8.5269/1.8063) mem 68106MB [2022-12-20 04:15:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][850/1519] eta 0:11:12 lr 0.000015 time 0.9757 (1.0048) model_time 0.9755 (1.0040) loss 0.8412 (0.8303) grad_norm 8.4078 (8.4837/1.7764) mem 68106MB [2022-12-20 04:15:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][860/1519] eta 0:11:02 lr 0.000015 time 0.9307 (1.0047) model_time 0.9305 (1.0039) loss 0.7121 (0.8301) grad_norm 8.7409 (8.4952/1.8033) mem 68106MB [2022-12-20 04:15:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][870/1519] eta 0:10:52 lr 0.000015 time 0.9221 (1.0049) model_time 0.9220 (1.0040) loss 1.0530 (0.8301) grad_norm 9.6755 (8.5025/1.8043) mem 68106MB [2022-12-20 04:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][880/1519] eta 0:10:42 lr 0.000015 time 0.9122 (1.0049) model_time 0.9121 (1.0041) loss 1.1114 (0.8305) grad_norm 9.3570 (8.4894/1.8125) mem 68106MB [2022-12-20 04:15:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][890/1519] eta 0:10:32 lr 0.000015 time 0.9218 (1.0048) model_time 0.9217 (1.0040) loss 0.8014 (0.8310) grad_norm 6.7996 (8.4920/1.8101) mem 68106MB [2022-12-20 04:15:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][900/1519] eta 0:10:21 lr 0.000015 time 0.9275 (1.0047) model_time 0.9273 (1.0039) loss 0.8019 (0.8314) grad_norm 7.3148 (8.4970/1.8171) mem 68106MB [2022-12-20 04:16:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][910/1519] eta 0:10:11 lr 0.000015 time 0.9708 (1.0047) model_time 0.9706 (1.0039) loss 0.8906 (0.8325) grad_norm 7.4928 (8.4850/1.8178) mem 68106MB [2022-12-20 04:16:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][920/1519] eta 0:10:01 lr 0.000015 time 0.9302 (1.0047) model_time 0.9300 (1.0039) loss 0.8975 (0.8325) grad_norm 5.0178 (8.4446/1.8215) mem 68106MB [2022-12-20 04:16:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][930/1519] eta 0:09:51 lr 0.000015 time 0.9141 (1.0046) model_time 0.9139 (1.0038) loss 0.7265 (0.8322) grad_norm 11.6168 (8.4730/1.8225) mem 68106MB [2022-12-20 04:16:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][940/1519] eta 0:09:41 lr 0.000015 time 0.9340 (1.0047) model_time 0.9338 (1.0039) loss 0.7583 (0.8328) grad_norm 7.6513 (8.5016/1.8226) mem 68106MB [2022-12-20 04:16:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][950/1519] eta 0:09:31 lr 0.000015 time 0.9222 (1.0046) model_time 0.9221 (1.0038) loss 0.9590 (0.8330) grad_norm 11.8228 (8.5300/1.8233) mem 68106MB [2022-12-20 04:16:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][960/1519] eta 0:09:21 lr 0.000015 time 0.9240 (1.0046) model_time 0.9238 (1.0038) loss 0.8848 (0.8326) grad_norm 8.9249 (8.5519/1.8185) mem 68106MB [2022-12-20 04:17:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][970/1519] eta 0:09:11 lr 0.000015 time 0.9777 (1.0046) model_time 0.9776 (1.0038) loss 0.6717 (0.8324) grad_norm 7.1569 (8.5261/1.7989) mem 68106MB [2022-12-20 04:17:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][980/1519] eta 0:09:01 lr 0.000015 time 0.9273 (1.0046) model_time 0.9271 (1.0038) loss 0.7918 (0.8320) grad_norm 7.4925 (8.5190/1.7869) mem 68106MB [2022-12-20 04:17:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][990/1519] eta 0:08:51 lr 0.000015 time 0.9211 (1.0046) model_time 0.9209 (1.0038) loss 0.7679 (0.8328) grad_norm 6.1667 (8.5200/1.7817) mem 68106MB [2022-12-20 04:17:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1000/1519] eta 0:08:41 lr 0.000015 time 0.9321 (1.0045) model_time 0.9319 (1.0037) loss 0.6968 (0.8330) grad_norm 8.6970 (8.5457/1.7864) mem 68106MB [2022-12-20 04:17:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1010/1519] eta 0:08:31 lr 0.000015 time 0.9207 (1.0046) model_time 0.9206 (1.0038) loss 0.8162 (0.8328) grad_norm 8.7626 (8.5334/1.7854) mem 68106MB [2022-12-20 04:17:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1020/1519] eta 0:08:21 lr 0.000015 time 0.9333 (1.0045) model_time 0.9332 (1.0038) loss 0.8559 (0.8329) grad_norm 6.4582 (8.5389/1.7847) mem 68106MB [2022-12-20 04:18:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1030/1519] eta 0:08:11 lr 0.000015 time 0.9336 (1.0045) model_time 0.9335 (1.0037) loss 0.8637 (0.8327) grad_norm 11.4954 (8.5321/1.7643) mem 68106MB [2022-12-20 04:18:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1040/1519] eta 0:08:01 lr 0.000015 time 0.9414 (1.0044) model_time 0.9413 (1.0037) loss 1.1204 (0.8329) grad_norm 9.9746 (8.5164/1.7569) mem 68106MB [2022-12-20 04:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1050/1519] eta 0:07:51 lr 0.000015 time 0.9209 (1.0046) model_time 0.9208 (1.0038) loss 1.0102 (0.8334) grad_norm 6.2608 (8.5125/1.7594) mem 68106MB [2022-12-20 04:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1060/1519] eta 0:07:41 lr 0.000015 time 0.9320 (1.0046) model_time 0.9318 (1.0038) loss 0.7358 (0.8327) grad_norm 7.2024 (8.5303/1.7546) mem 68106MB [2022-12-20 04:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1070/1519] eta 0:07:31 lr 0.000015 time 0.9334 (1.0046) model_time 0.9333 (1.0039) loss 0.7430 (0.8321) grad_norm 9.7560 (8.5232/1.7594) mem 68106MB [2022-12-20 04:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1080/1519] eta 0:07:21 lr 0.000015 time 0.9301 (1.0046) model_time 0.9299 (1.0039) loss 0.7450 (0.8321) grad_norm 7.2983 (8.5318/1.7675) mem 68106MB [2022-12-20 04:19:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1090/1519] eta 0:07:10 lr 0.000015 time 0.9663 (1.0046) model_time 0.9661 (1.0039) loss 0.7209 (0.8317) grad_norm 7.4011 (8.5074/1.7794) mem 68106MB [2022-12-20 04:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1100/1519] eta 0:07:00 lr 0.000015 time 0.9756 (1.0046) model_time 0.9754 (1.0039) loss 0.6995 (0.8312) grad_norm 5.7492 (8.4836/1.7905) mem 68106MB [2022-12-20 04:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1110/1519] eta 0:06:50 lr 0.000015 time 0.9322 (1.0046) model_time 0.9320 (1.0039) loss 0.6706 (0.8308) grad_norm 5.7142 (8.4792/1.7955) mem 68106MB [2022-12-20 04:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1120/1519] eta 0:06:40 lr 0.000015 time 0.9312 (1.0046) model_time 0.9310 (1.0039) loss 0.9059 (0.8314) grad_norm 11.6727 (8.5051/1.8109) mem 68106MB [2022-12-20 04:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1130/1519] eta 0:06:30 lr 0.000015 time 0.9247 (1.0045) model_time 0.9245 (1.0038) loss 0.9779 (0.8318) grad_norm 9.0333 (8.5080/1.8176) mem 68106MB [2022-12-20 04:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1140/1519] eta 0:06:20 lr 0.000015 time 0.9354 (1.0046) model_time 0.9352 (1.0039) loss 0.8761 (0.8315) grad_norm 5.3834 (8.5010/1.8115) mem 68106MB [2022-12-20 04:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1150/1519] eta 0:06:10 lr 0.000015 time 0.9869 (1.0046) model_time 0.9867 (1.0039) loss 0.7746 (0.8316) grad_norm 8.1255 (8.5251/1.8153) mem 68106MB [2022-12-20 04:20:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1160/1519] eta 0:06:00 lr 0.000015 time 0.9250 (1.0045) model_time 0.9249 (1.0038) loss 0.9886 (0.8322) grad_norm 7.6993 (8.4873/1.7946) mem 68106MB [2022-12-20 04:20:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1170/1519] eta 0:05:50 lr 0.000015 time 0.9218 (1.0046) model_time 0.9217 (1.0039) loss 0.8538 (0.8325) grad_norm 7.5757 (8.4269/1.7415) mem 68106MB [2022-12-20 04:20:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1180/1519] eta 0:05:40 lr 0.000015 time 0.9224 (1.0046) model_time 0.9222 (1.0039) loss 0.6896 (0.8320) grad_norm 8.0847 (8.4130/1.7296) mem 68106MB [2022-12-20 04:20:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1190/1519] eta 0:05:30 lr 0.000015 time 0.9331 (1.0048) model_time 0.9330 (1.0041) loss 0.7813 (0.8320) grad_norm 7.7779 (8.4061/1.7363) mem 68106MB [2022-12-20 04:20:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1200/1519] eta 0:05:20 lr 0.000015 time 0.9334 (1.0048) model_time 0.9332 (1.0041) loss 0.6655 (0.8320) grad_norm 8.2824 (8.3895/1.7468) mem 68106MB [2022-12-20 04:21:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1210/1519] eta 0:05:10 lr 0.000015 time 0.9242 (1.0048) model_time 0.9240 (1.0041) loss 0.7447 (0.8325) grad_norm 6.7631 (8.3904/1.7516) mem 68106MB [2022-12-20 04:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1220/1519] eta 0:05:00 lr 0.000015 time 0.9358 (1.0048) model_time 0.9357 (1.0041) loss 0.6922 (0.8324) grad_norm 8.5349 (8.3948/1.7560) mem 68106MB [2022-12-20 04:21:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1230/1519] eta 0:04:50 lr 0.000015 time 0.9202 (1.0050) model_time 0.9201 (1.0043) loss 0.7054 (0.8325) grad_norm 12.7909 (8.3996/1.7429) mem 68106MB [2022-12-20 04:21:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1240/1519] eta 0:04:40 lr 0.000015 time 0.9235 (1.0050) model_time 0.9233 (1.0043) loss 0.7624 (0.8321) grad_norm 6.3821 (8.3971/1.7628) mem 68106MB [2022-12-20 04:21:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1250/1519] eta 0:04:30 lr 0.000015 time 0.9433 (1.0050) model_time 0.9432 (1.0043) loss 0.6929 (0.8323) grad_norm 8.5436 (8.3882/1.7704) mem 68106MB [2022-12-20 04:21:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1260/1519] eta 0:04:20 lr 0.000015 time 0.9155 (1.0050) model_time 0.9154 (1.0043) loss 0.9785 (0.8321) grad_norm 10.4007 (8.4418/1.8428) mem 68106MB [2022-12-20 04:22:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1270/1519] eta 0:04:10 lr 0.000015 time 0.9718 (1.0050) model_time 0.9717 (1.0043) loss 0.6706 (0.8315) grad_norm 9.7688 (8.4467/1.8420) mem 68106MB [2022-12-20 04:22:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1280/1519] eta 0:04:00 lr 0.000015 time 0.9232 (1.0049) model_time 0.9231 (1.0042) loss 0.8402 (0.8313) grad_norm 10.4798 (8.4535/1.8439) mem 68106MB [2022-12-20 04:22:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1290/1519] eta 0:03:50 lr 0.000015 time 0.9290 (1.0049) model_time 0.9288 (1.0042) loss 0.7072 (0.8313) grad_norm 10.8256 (8.4572/1.8447) mem 68106MB [2022-12-20 04:22:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1300/1519] eta 0:03:40 lr 0.000015 time 0.9315 (1.0049) model_time 0.9314 (1.0042) loss 0.6649 (0.8313) grad_norm 8.8177 (8.4685/1.8423) mem 68106MB [2022-12-20 04:22:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1310/1519] eta 0:03:30 lr 0.000015 time 0.9289 (1.0049) model_time 0.9288 (1.0042) loss 0.7574 (0.8310) grad_norm 7.8884 (8.4587/1.8190) mem 68106MB [2022-12-20 04:22:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1320/1519] eta 0:03:19 lr 0.000014 time 0.9235 (1.0048) model_time 0.9234 (1.0042) loss 0.8387 (0.8316) grad_norm 10.1871 (8.4900/1.8223) mem 68106MB [2022-12-20 04:23:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1330/1519] eta 0:03:09 lr 0.000014 time 0.9242 (1.0048) model_time 0.9241 (1.0041) loss 0.6879 (0.8315) grad_norm 6.8448 (8.4852/1.8220) mem 68106MB [2022-12-20 04:23:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1340/1519] eta 0:02:59 lr 0.000014 time 0.9260 (1.0048) model_time 0.9258 (1.0041) loss 0.9174 (0.8318) grad_norm 8.8447 (8.4699/1.7967) mem 68106MB [2022-12-20 04:23:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1350/1519] eta 0:02:49 lr 0.000014 time 0.9206 (1.0048) model_time 0.9204 (1.0042) loss 0.7084 (0.8316) grad_norm 6.9761 (8.4888/1.8083) mem 68106MB [2022-12-20 04:23:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1360/1519] eta 0:02:39 lr 0.000014 time 0.9238 (1.0048) model_time 0.9237 (1.0041) loss 0.6869 (0.8316) grad_norm 6.1554 (8.5002/1.8215) mem 68106MB [2022-12-20 04:23:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1370/1519] eta 0:02:29 lr 0.000014 time 0.9892 (1.0050) model_time 0.9890 (1.0043) loss 0.7156 (0.8318) grad_norm 10.0838 (8.5097/1.8185) mem 68106MB [2022-12-20 04:23:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1380/1519] eta 0:02:19 lr 0.000014 time 0.9476 (1.0050) model_time 0.9475 (1.0044) loss 0.7492 (0.8312) grad_norm 8.5738 (8.5086/1.8133) mem 68106MB [2022-12-20 04:24:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1390/1519] eta 0:02:09 lr 0.000014 time 0.9211 (1.0050) model_time 0.9209 (1.0043) loss 0.7539 (0.8314) grad_norm 9.4892 (8.5134/1.7945) mem 68106MB [2022-12-20 04:24:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1400/1519] eta 0:01:59 lr 0.000014 time 0.9281 (1.0050) model_time 0.9279 (1.0043) loss 0.7263 (0.8321) grad_norm 7.7699 (8.4986/1.7614) mem 68106MB [2022-12-20 04:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1410/1519] eta 0:01:49 lr 0.000014 time 0.9339 (1.0050) model_time 0.9338 (1.0043) loss 1.0157 (0.8320) grad_norm 8.5653 (8.4608/1.7023) mem 68106MB [2022-12-20 04:24:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1420/1519] eta 0:01:39 lr 0.000014 time 0.9238 (1.0050) model_time 0.9236 (1.0044) loss 0.7163 (0.8314) grad_norm 7.5931 (8.4477/1.6909) mem 68106MB [2022-12-20 04:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1430/1519] eta 0:01:29 lr 0.000014 time 0.9217 (1.0050) model_time 0.9196 (1.0044) loss 0.8029 (0.8308) grad_norm 8.8798 (8.4425/1.6872) mem 68106MB [2022-12-20 04:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1440/1519] eta 0:01:19 lr 0.000014 time 0.9216 (1.0050) model_time 0.9215 (1.0043) loss 0.7544 (0.8305) grad_norm 7.7082 (8.4350/1.6815) mem 68106MB [2022-12-20 04:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1450/1519] eta 0:01:09 lr 0.000014 time 0.9206 (1.0049) model_time 0.9204 (1.0043) loss 0.9007 (0.8308) grad_norm 7.7829 (8.4540/1.6766) mem 68106MB [2022-12-20 04:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1460/1519] eta 0:00:59 lr 0.000014 time 0.9265 (1.0049) model_time 0.9264 (1.0043) loss 0.7131 (0.8304) grad_norm 6.6219 (8.4475/1.6527) mem 68106MB [2022-12-20 04:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1470/1519] eta 0:00:49 lr 0.000014 time 0.9259 (1.0049) model_time 0.9258 (1.0043) loss 1.0938 (0.8302) grad_norm 6.5177 (8.4385/1.6595) mem 68106MB [2022-12-20 04:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1480/1519] eta 0:00:39 lr 0.000014 time 0.9113 (1.0049) model_time 0.9112 (1.0043) loss 0.7346 (0.8301) grad_norm 5.5559 (8.4350/1.6628) mem 68106MB [2022-12-20 04:25:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1490/1519] eta 0:00:29 lr 0.000014 time 0.9245 (1.0049) model_time 0.9244 (1.0043) loss 1.0699 (0.8308) grad_norm 7.4007 (8.4417/1.6686) mem 68106MB [2022-12-20 04:25:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1500/1519] eta 0:00:19 lr 0.000014 time 0.9243 (1.0049) model_time 0.9242 (1.0043) loss 0.6997 (0.8307) grad_norm 6.6861 (8.4224/1.6630) mem 68106MB [2022-12-20 04:26:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [54/100][1510/1519] eta 0:00:09 lr 0.000014 time 0.9161 (1.0049) model_time 0.9160 (1.0043) loss 0.8881 (0.8307) grad_norm 10.1338 (8.4121/1.6549) mem 68106MB [2022-12-20 04:26:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 54 training takes 0:25:26 [2022-12-20 04:26:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_54.pth saving...... [2022-12-20 04:26:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_54.pth saved !!! [2022-12-20 04:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.679 (0.679) Loss 0.5258 (0.5258) Acc@1 90.972 (90.972) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 04:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.332) Loss 0.5019 (0.4877) Acc@1 92.361 (92.361) Acc@5 98.264 (98.485) Mem 68106MB [2022-12-20 04:26:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.304 (0.317) Loss 0.4544 (0.4846) Acc@1 92.361 (92.477) Acc@5 98.958 (98.380) Mem 68106MB [2022-12-20 04:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.294 (0.311) Loss 0.6165 (0.4913) Acc@1 88.889 (92.204) Acc@5 97.569 (98.376) Mem 68106MB [2022-12-20 04:26:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.306 (0.308) Loss 0.4459 (0.4829) Acc@1 93.750 (92.319) Acc@5 99.306 (98.450) Mem 68106MB [2022-12-20 04:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.306) Loss 0.4829 (0.4810) Acc@1 90.625 (92.313) Acc@5 99.306 (98.509) Mem 68106MB [2022-12-20 04:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.4978 (0.4806) Acc@1 91.319 (92.321) Acc@5 97.917 (98.480) Mem 68106MB [2022-12-20 04:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5311 (0.4820) Acc@1 91.667 (92.239) Acc@5 98.611 (98.513) Mem 68106MB [2022-12-20 04:27:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.303) Loss 0.4114 (0.4803) Acc@1 94.097 (92.254) Acc@5 98.264 (98.555) Mem 68106MB [2022-12-20 04:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:54] * Acc@1 92.236 Acc@5 98.559 [2022-12-20 04:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.2% [2022-12-20 04:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 04:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 04:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.24% [2022-12-20 04:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][0/1519] eta 0:34:41 lr 0.000014 time 1.3701 (1.3701) model_time 0.9237 (0.9237) loss 0.6887 (0.6887) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 04:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][10/1519] eta 0:25:56 lr 0.000014 time 0.9309 (1.0317) model_time 0.9308 (0.9909) loss 0.8088 (0.8042) grad_norm 7.6770 (7.7644/1.0335) mem 68106MB [2022-12-20 04:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][20/1519] eta 0:25:26 lr 0.000014 time 0.9230 (1.0182) model_time 0.9228 (0.9967) loss 0.8823 (0.8295) grad_norm 8.7352 (8.3022/1.2267) mem 68106MB [2022-12-20 04:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][30/1519] eta 0:25:09 lr 0.000014 time 0.9256 (1.0139) model_time 0.9254 (0.9992) loss 1.0204 (0.8333) grad_norm 8.5587 (8.4046/1.3034) mem 68106MB [2022-12-20 04:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][40/1519] eta 0:24:55 lr 0.000014 time 0.9220 (1.0110) model_time 0.9219 (0.9998) loss 0.8658 (0.8348) grad_norm 6.7241 (8.4338/1.2714) mem 68106MB [2022-12-20 04:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][50/1519] eta 0:24:45 lr 0.000014 time 0.9559 (1.0114) model_time 0.9558 (1.0023) loss 0.6675 (0.8302) grad_norm 6.6885 (8.1631/1.3294) mem 68106MB [2022-12-20 04:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][60/1519] eta 0:24:34 lr 0.000014 time 0.9827 (1.0106) model_time 0.9825 (1.0030) loss 0.7251 (0.8349) grad_norm 8.1229 (8.3286/1.6316) mem 68106MB [2022-12-20 04:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][70/1519] eta 0:24:26 lr 0.000014 time 0.9832 (1.0123) model_time 0.9830 (1.0057) loss 1.1732 (0.8421) grad_norm 8.1234 (8.3403/1.5539) mem 68106MB [2022-12-20 04:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][80/1519] eta 0:24:14 lr 0.000014 time 0.9332 (1.0106) model_time 0.9330 (1.0048) loss 1.0818 (0.8393) grad_norm 11.5984 (8.5913/1.8312) mem 68106MB [2022-12-20 04:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][90/1519] eta 0:24:03 lr 0.000014 time 0.9260 (1.0102) model_time 0.9259 (1.0050) loss 0.8351 (0.8491) grad_norm 10.9225 (8.5772/1.7825) mem 68106MB [2022-12-20 04:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][100/1519] eta 0:23:52 lr 0.000014 time 0.9225 (1.0098) model_time 0.9223 (1.0051) loss 0.7051 (0.8467) grad_norm 10.5573 (8.6057/1.7291) mem 68106MB [2022-12-20 04:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][110/1519] eta 0:23:41 lr 0.000014 time 0.9246 (1.0088) model_time 0.9244 (1.0045) loss 0.8929 (0.8425) grad_norm 7.9096 (8.6589/1.7997) mem 68106MB [2022-12-20 04:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][120/1519] eta 0:23:30 lr 0.000014 time 0.9374 (1.0079) model_time 0.9372 (1.0039) loss 0.7667 (0.8443) grad_norm 9.0243 (8.7922/2.0681) mem 68106MB [2022-12-20 04:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][130/1519] eta 0:23:19 lr 0.000014 time 0.9212 (1.0074) model_time 0.9210 (1.0037) loss 1.0543 (0.8430) grad_norm 9.5612 (8.8563/2.0489) mem 68106MB [2022-12-20 04:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][140/1519] eta 0:23:09 lr 0.000014 time 0.9230 (1.0074) model_time 0.9227 (1.0040) loss 0.8026 (0.8378) grad_norm 7.7870 (8.7558/2.0300) mem 68106MB [2022-12-20 04:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][150/1519] eta 0:23:00 lr 0.000014 time 0.9775 (1.0081) model_time 0.9774 (1.0048) loss 0.8420 (0.8432) grad_norm 6.3041 (8.6311/2.0291) mem 68106MB [2022-12-20 04:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][160/1519] eta 0:22:48 lr 0.000014 time 0.9202 (1.0074) model_time 0.9201 (1.0043) loss 0.9596 (0.8422) grad_norm 8.5185 (8.6798/1.9800) mem 68106MB [2022-12-20 04:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][170/1519] eta 0:22:38 lr 0.000014 time 0.9184 (1.0073) model_time 0.9183 (1.0043) loss 0.7034 (0.8386) grad_norm 8.4170 (8.6818/1.9368) mem 68106MB [2022-12-20 04:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][180/1519] eta 0:22:28 lr 0.000014 time 0.9288 (1.0068) model_time 0.9287 (1.0041) loss 1.0223 (0.8384) grad_norm 6.4680 (8.6814/1.9405) mem 68106MB [2022-12-20 04:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][190/1519] eta 0:22:17 lr 0.000014 time 0.9305 (1.0063) model_time 0.9302 (1.0037) loss 0.6745 (0.8367) grad_norm 8.7315 (8.6661/1.9502) mem 68106MB [2022-12-20 04:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][200/1519] eta 0:22:07 lr 0.000014 time 0.9216 (1.0061) model_time 0.9214 (1.0036) loss 0.7118 (0.8359) grad_norm 6.8699 (8.6024/1.9300) mem 68106MB [2022-12-20 04:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][210/1519] eta 0:21:57 lr 0.000014 time 0.9220 (1.0063) model_time 0.9218 (1.0039) loss 0.8241 (0.8369) grad_norm 7.2990 (8.6263/1.9211) mem 68106MB [2022-12-20 04:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][220/1519] eta 0:21:47 lr 0.000014 time 0.9215 (1.0062) model_time 0.9213 (1.0039) loss 0.6691 (0.8356) grad_norm 6.8495 (8.5901/1.9040) mem 68106MB [2022-12-20 04:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][230/1519] eta 0:21:36 lr 0.000014 time 0.9206 (1.0062) model_time 0.9205 (1.0039) loss 0.9365 (0.8346) grad_norm 10.1604 (8.5810/1.8816) mem 68106MB [2022-12-20 04:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][240/1519] eta 0:21:26 lr 0.000014 time 0.9245 (1.0058) model_time 0.9243 (1.0037) loss 0.8103 (0.8335) grad_norm 11.8134 (8.5995/1.8862) mem 68106MB [2022-12-20 04:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][250/1519] eta 0:21:16 lr 0.000014 time 0.9247 (1.0056) model_time 0.9246 (1.0035) loss 0.7850 (0.8316) grad_norm 8.5150 (8.5763/1.8571) mem 68106MB [2022-12-20 04:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][260/1519] eta 0:21:05 lr 0.000014 time 0.9208 (1.0054) model_time 0.9207 (1.0034) loss 0.6793 (0.8308) grad_norm 8.1876 (8.6439/1.9130) mem 68106MB [2022-12-20 04:32:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][270/1519] eta 0:20:55 lr 0.000014 time 0.9319 (1.0051) model_time 0.9317 (1.0031) loss 0.8008 (0.8308) grad_norm 11.2388 (8.6325/1.9189) mem 68106MB [2022-12-20 04:32:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][280/1519] eta 0:20:45 lr 0.000014 time 0.9256 (1.0050) model_time 0.9255 (1.0031) loss 0.7353 (0.8331) grad_norm 9.6403 (8.6880/1.9478) mem 68106MB [2022-12-20 04:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][290/1519] eta 0:20:35 lr 0.000014 time 0.9246 (1.0050) model_time 0.9245 (1.0032) loss 1.0746 (0.8314) grad_norm 13.4286 (8.7621/2.0122) mem 68106MB [2022-12-20 04:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][300/1519] eta 0:20:24 lr 0.000014 time 0.9232 (1.0048) model_time 0.9231 (1.0030) loss 0.9431 (0.8328) grad_norm 6.6094 (8.7544/2.0121) mem 68106MB [2022-12-20 04:32:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][310/1519] eta 0:20:14 lr 0.000014 time 0.9243 (1.0046) model_time 0.9242 (1.0028) loss 0.9221 (0.8349) grad_norm 6.4267 (8.7285/2.0032) mem 68106MB [2022-12-20 04:32:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][320/1519] eta 0:20:04 lr 0.000014 time 0.9354 (1.0048) model_time 0.9353 (1.0031) loss 0.6938 (0.8365) grad_norm 6.5330 (8.7056/1.9831) mem 68106MB [2022-12-20 04:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][330/1519] eta 0:19:54 lr 0.000014 time 0.9025 (1.0049) model_time 0.9023 (1.0032) loss 0.9584 (0.8349) grad_norm 6.6890 (8.6684/1.9657) mem 68106MB [2022-12-20 04:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][340/1519] eta 0:19:44 lr 0.000014 time 0.9845 (1.0049) model_time 0.9843 (1.0033) loss 1.0087 (0.8348) grad_norm 7.1313 (8.6260/1.9596) mem 68106MB [2022-12-20 04:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][350/1519] eta 0:19:34 lr 0.000014 time 0.9227 (1.0048) model_time 0.9225 (1.0032) loss 0.6994 (0.8347) grad_norm 6.7818 (8.6304/1.9443) mem 68106MB [2022-12-20 04:33:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][360/1519] eta 0:19:24 lr 0.000014 time 0.9285 (1.0046) model_time 0.9284 (1.0031) loss 0.8561 (0.8328) grad_norm 6.1265 (8.5935/1.9407) mem 68106MB [2022-12-20 04:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][370/1519] eta 0:19:14 lr 0.000014 time 0.9226 (1.0046) model_time 0.9224 (1.0031) loss 0.8371 (0.8337) grad_norm 9.8119 (8.6019/1.9304) mem 68106MB [2022-12-20 04:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][380/1519] eta 0:19:04 lr 0.000014 time 0.9268 (1.0045) model_time 0.9266 (1.0031) loss 1.0224 (0.8343) grad_norm 6.4528 (8.5793/1.9462) mem 68106MB [2022-12-20 04:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][390/1519] eta 0:18:54 lr 0.000014 time 0.9268 (1.0051) model_time 0.9267 (1.0037) loss 0.7180 (0.8328) grad_norm 9.7059 (8.5851/1.9232) mem 68106MB [2022-12-20 04:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][400/1519] eta 0:18:44 lr 0.000014 time 0.9270 (1.0052) model_time 0.9268 (1.0037) loss 0.7228 (0.8333) grad_norm 8.2154 (8.5919/1.9128) mem 68106MB [2022-12-20 04:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][410/1519] eta 0:18:34 lr 0.000014 time 0.9126 (1.0053) model_time 0.9124 (1.0039) loss 1.0349 (0.8343) grad_norm 10.5248 (8.5936/1.9049) mem 68106MB [2022-12-20 04:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][420/1519] eta 0:18:24 lr 0.000014 time 0.9241 (1.0052) model_time 0.9240 (1.0038) loss 0.6724 (0.8339) grad_norm 9.5431 (8.6017/1.9239) mem 68106MB [2022-12-20 04:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][430/1519] eta 0:18:14 lr 0.000014 time 0.9246 (1.0050) model_time 0.9245 (1.0037) loss 1.0030 (0.8345) grad_norm 6.8754 (8.5926/1.9082) mem 68106MB [2022-12-20 04:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][440/1519] eta 0:18:04 lr 0.000014 time 0.9641 (1.0050) model_time 0.9640 (1.0037) loss 1.1255 (0.8340) grad_norm 12.8821 (8.6140/1.9131) mem 68106MB [2022-12-20 04:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][450/1519] eta 0:17:54 lr 0.000014 time 0.9331 (1.0048) model_time 0.9329 (1.0035) loss 0.8432 (0.8328) grad_norm 10.3169 (8.6090/1.9208) mem 68106MB [2022-12-20 04:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][460/1519] eta 0:17:44 lr 0.000014 time 0.9261 (1.0050) model_time 0.9260 (1.0037) loss 0.9015 (0.8332) grad_norm 13.7332 (8.6297/1.9317) mem 68106MB [2022-12-20 04:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][470/1519] eta 0:17:34 lr 0.000014 time 0.9200 (1.0050) model_time 0.9199 (1.0038) loss 0.9603 (0.8328) grad_norm 9.1496 (8.6307/1.9162) mem 68106MB [2022-12-20 04:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][480/1519] eta 0:17:24 lr 0.000014 time 0.9295 (1.0050) model_time 0.9294 (1.0038) loss 1.1052 (0.8334) grad_norm 6.8369 (8.6225/1.9017) mem 68106MB [2022-12-20 04:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][490/1519] eta 0:17:14 lr 0.000014 time 0.9269 (1.0050) model_time 0.9267 (1.0038) loss 0.7113 (0.8341) grad_norm 8.8606 (8.6332/1.9021) mem 68106MB [2022-12-20 04:35:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][500/1519] eta 0:17:03 lr 0.000014 time 0.9240 (1.0049) model_time 0.9239 (1.0037) loss 0.8580 (0.8339) grad_norm 7.5309 (8.6326/1.8877) mem 68106MB [2022-12-20 04:36:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][510/1519] eta 0:16:53 lr 0.000014 time 1.0000 (1.0049) model_time 0.9998 (1.0037) loss 0.8453 (0.8334) grad_norm 9.6725 (8.6413/1.8849) mem 68106MB [2022-12-20 04:36:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][520/1519] eta 0:16:43 lr 0.000014 time 1.0309 (1.0049) model_time 1.0308 (1.0038) loss 0.6781 (0.8335) grad_norm 8.2205 (8.6272/1.8742) mem 68106MB [2022-12-20 04:36:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][530/1519] eta 0:16:33 lr 0.000014 time 0.9233 (1.0049) model_time 0.9231 (1.0037) loss 1.1000 (0.8344) grad_norm 9.4963 (8.6199/1.8721) mem 68106MB [2022-12-20 04:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][540/1519] eta 0:16:23 lr 0.000014 time 0.9062 (1.0048) model_time 0.9060 (1.0037) loss 0.8477 (0.8339) grad_norm 6.9190 (8.6404/1.8941) mem 68106MB [2022-12-20 04:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][550/1519] eta 0:16:13 lr 0.000014 time 0.9298 (1.0047) model_time 0.9296 (1.0036) loss 0.6591 (0.8343) grad_norm 5.6576 (8.6294/1.8903) mem 68106MB [2022-12-20 04:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][560/1519] eta 0:16:03 lr 0.000014 time 0.9233 (1.0048) model_time 0.9232 (1.0037) loss 0.6874 (0.8343) grad_norm 6.9729 (8.6184/1.8817) mem 68106MB [2022-12-20 04:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][570/1519] eta 0:15:53 lr 0.000014 time 0.9484 (1.0049) model_time 0.9483 (1.0038) loss 0.9170 (0.8342) grad_norm 9.4537 (8.6148/1.8822) mem 68106MB [2022-12-20 04:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][580/1519] eta 0:15:43 lr 0.000014 time 0.9203 (1.0047) model_time 0.9201 (1.0037) loss 0.8533 (0.8336) grad_norm 8.6398 (8.6081/1.8723) mem 68106MB [2022-12-20 04:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][590/1519] eta 0:15:33 lr 0.000014 time 0.9265 (1.0047) model_time 0.9264 (1.0036) loss 0.7395 (0.8328) grad_norm 8.0172 (8.6119/1.8624) mem 68106MB [2022-12-20 04:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][600/1519] eta 0:15:23 lr 0.000014 time 0.9761 (1.0047) model_time 0.9758 (1.0037) loss 0.7714 (0.8321) grad_norm 11.7309 (8.6193/1.8714) mem 68106MB [2022-12-20 04:37:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][610/1519] eta 0:15:13 lr 0.000014 time 0.9274 (1.0046) model_time 0.9273 (1.0036) loss 0.6925 (0.8324) grad_norm 9.9961 (8.6252/1.8741) mem 68106MB [2022-12-20 04:37:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][620/1519] eta 0:15:03 lr 0.000014 time 0.9754 (1.0046) model_time 0.9753 (1.0036) loss 0.7695 (0.8322) grad_norm 6.0648 (8.6459/1.9220) mem 68106MB [2022-12-20 04:38:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][630/1519] eta 0:14:53 lr 0.000014 time 0.9213 (1.0046) model_time 0.9211 (1.0036) loss 0.9376 (0.8319) grad_norm 7.4112 (8.6412/1.9262) mem 68106MB [2022-12-20 04:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][640/1519] eta 0:14:42 lr 0.000014 time 0.9334 (1.0045) model_time 0.9332 (1.0035) loss 0.7735 (0.8319) grad_norm 6.3624 (8.6312/1.9353) mem 68106MB [2022-12-20 04:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][650/1519] eta 0:14:32 lr 0.000014 time 0.9379 (1.0046) model_time 0.9377 (1.0036) loss 0.7570 (0.8319) grad_norm 8.2148 (8.6612/1.9435) mem 68106MB [2022-12-20 04:38:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][660/1519] eta 0:14:22 lr 0.000014 time 0.9283 (1.0045) model_time 0.9281 (1.0036) loss 1.0410 (0.8321) grad_norm 6.6485 (8.6449/1.9321) mem 68106MB [2022-12-20 04:38:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][670/1519] eta 0:14:12 lr 0.000014 time 0.9351 (1.0045) model_time 0.9349 (1.0035) loss 0.8022 (0.8334) grad_norm 10.2949 (8.6442/1.9378) mem 68106MB [2022-12-20 04:38:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][680/1519] eta 0:14:02 lr 0.000014 time 0.9254 (1.0045) model_time 0.9252 (1.0035) loss 0.7399 (0.8330) grad_norm 8.0296 (8.5988/1.9101) mem 68106MB [2022-12-20 04:39:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][690/1519] eta 0:13:52 lr 0.000014 time 0.9107 (1.0044) model_time 0.9106 (1.0035) loss 0.7964 (0.8324) grad_norm 7.9982 (8.5956/1.9151) mem 68106MB [2022-12-20 04:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][700/1519] eta 0:13:42 lr 0.000014 time 0.9208 (1.0045) model_time 0.9207 (1.0035) loss 0.7305 (0.8329) grad_norm 9.8799 (8.5866/1.9204) mem 68106MB [2022-12-20 04:39:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][710/1519] eta 0:13:32 lr 0.000014 time 0.9264 (1.0044) model_time 0.9263 (1.0035) loss 0.8460 (0.8336) grad_norm 10.1657 (8.5626/1.9064) mem 68106MB [2022-12-20 04:39:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][720/1519] eta 0:13:22 lr 0.000014 time 0.9266 (1.0045) model_time 0.9265 (1.0036) loss 0.6891 (0.8337) grad_norm 9.2812 (8.5243/1.8408) mem 68106MB [2022-12-20 04:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][730/1519] eta 0:13:12 lr 0.000014 time 0.9220 (1.0044) model_time 0.9219 (1.0035) loss 0.7011 (0.8337) grad_norm 12.9732 (8.5062/1.8459) mem 68106MB [2022-12-20 04:39:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][740/1519] eta 0:13:02 lr 0.000014 time 0.9300 (1.0044) model_time 0.9299 (1.0035) loss 0.7694 (0.8328) grad_norm 7.1772 (8.5200/1.8373) mem 68106MB [2022-12-20 04:40:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][750/1519] eta 0:12:52 lr 0.000014 time 0.9279 (1.0043) model_time 0.9278 (1.0034) loss 0.6948 (0.8334) grad_norm 8.8622 (8.5588/1.8809) mem 68106MB [2022-12-20 04:40:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][760/1519] eta 0:12:42 lr 0.000014 time 0.9348 (1.0043) model_time 0.9347 (1.0034) loss 0.7033 (0.8330) grad_norm 13.3312 (8.5656/1.9056) mem 68106MB [2022-12-20 04:40:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][770/1519] eta 0:12:32 lr 0.000014 time 0.9370 (1.0043) model_time 0.9369 (1.0035) loss 1.0445 (0.8331) grad_norm 8.5944 (8.5550/1.9042) mem 68106MB [2022-12-20 04:40:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][780/1519] eta 0:12:22 lr 0.000014 time 0.9983 (1.0045) model_time 0.9981 (1.0036) loss 0.7584 (0.8326) grad_norm 8.8677 (8.5509/1.8950) mem 68106MB [2022-12-20 04:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][790/1519] eta 0:12:12 lr 0.000014 time 0.9260 (1.0044) model_time 0.9259 (1.0035) loss 0.7856 (0.8317) grad_norm 12.4302 (8.5478/1.8961) mem 68106MB [2022-12-20 04:40:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][800/1519] eta 0:12:02 lr 0.000014 time 0.9232 (1.0043) model_time 0.9230 (1.0035) loss 0.6773 (0.8311) grad_norm 6.3999 (8.5566/1.8928) mem 68106MB [2022-12-20 04:41:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][810/1519] eta 0:11:52 lr 0.000014 time 0.9296 (1.0043) model_time 0.9294 (1.0034) loss 0.7074 (0.8309) grad_norm 9.6800 (8.5811/1.9271) mem 68106MB [2022-12-20 04:41:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][820/1519] eta 0:11:41 lr 0.000014 time 0.9249 (1.0042) model_time 0.9247 (1.0033) loss 0.8244 (0.8318) grad_norm 7.5935 (8.5890/1.9236) mem 68106MB [2022-12-20 04:41:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][830/1519] eta 0:11:31 lr 0.000014 time 0.9320 (1.0043) model_time 0.9319 (1.0034) loss 0.8629 (0.8317) grad_norm 7.4065 (8.5707/1.9308) mem 68106MB [2022-12-20 04:41:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][840/1519] eta 0:11:21 lr 0.000014 time 0.9262 (1.0043) model_time 0.9261 (1.0035) loss 0.6883 (0.8321) grad_norm 7.6222 (8.5576/1.9158) mem 68106MB [2022-12-20 04:41:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][850/1519] eta 0:11:11 lr 0.000014 time 0.9262 (1.0042) model_time 0.9260 (1.0034) loss 0.7820 (0.8325) grad_norm 8.1115 (8.5540/1.9175) mem 68106MB [2022-12-20 04:41:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][860/1519] eta 0:11:01 lr 0.000014 time 0.9205 (1.0042) model_time 0.9203 (1.0034) loss 0.8946 (0.8324) grad_norm 6.6052 (8.5257/1.8834) mem 68106MB [2022-12-20 04:42:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][870/1519] eta 0:10:51 lr 0.000014 time 0.9044 (1.0042) model_time 0.9042 (1.0034) loss 0.6869 (0.8325) grad_norm 8.0530 (8.5176/1.8668) mem 68106MB [2022-12-20 04:42:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][880/1519] eta 0:10:41 lr 0.000014 time 0.9208 (1.0043) model_time 0.9206 (1.0035) loss 0.9030 (0.8324) grad_norm 8.2329 (8.4828/1.8415) mem 68106MB [2022-12-20 04:42:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][890/1519] eta 0:10:31 lr 0.000014 time 0.9277 (1.0043) model_time 0.9276 (1.0035) loss 0.7783 (0.8323) grad_norm 7.4752 (8.4405/1.7909) mem 68106MB [2022-12-20 04:42:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][900/1519] eta 0:10:21 lr 0.000014 time 0.9216 (1.0042) model_time 0.9215 (1.0034) loss 0.7104 (0.8321) grad_norm 10.2752 (8.4437/1.7836) mem 68106MB [2022-12-20 04:42:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][910/1519] eta 0:10:11 lr 0.000014 time 0.9312 (1.0041) model_time 0.9310 (1.0034) loss 0.8004 (0.8317) grad_norm 6.5401 (8.4350/1.7853) mem 68106MB [2022-12-20 04:42:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][920/1519] eta 0:10:01 lr 0.000014 time 0.9183 (1.0041) model_time 0.9182 (1.0033) loss 0.8295 (0.8313) grad_norm 9.1961 (8.4675/1.7939) mem 68106MB [2022-12-20 04:43:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][930/1519] eta 0:09:51 lr 0.000014 time 0.9272 (1.0040) model_time 0.9271 (1.0032) loss 0.7631 (0.8308) grad_norm 10.3104 (8.4790/1.7987) mem 68106MB [2022-12-20 04:43:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][940/1519] eta 0:09:41 lr 0.000014 time 0.9346 (1.0040) model_time 0.9344 (1.0032) loss 0.6685 (0.8307) grad_norm 7.7491 (8.5040/1.7983) mem 68106MB [2022-12-20 04:43:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][950/1519] eta 0:09:31 lr 0.000014 time 0.9235 (1.0040) model_time 0.9233 (1.0032) loss 0.6683 (0.8300) grad_norm 7.0933 (8.5045/1.8223) mem 68106MB [2022-12-20 04:43:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][960/1519] eta 0:09:21 lr 0.000014 time 0.9239 (1.0040) model_time 0.9238 (1.0032) loss 1.1451 (0.8300) grad_norm 8.5523 (8.5329/1.8131) mem 68106MB [2022-12-20 04:43:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][970/1519] eta 0:09:11 lr 0.000014 time 0.9323 (1.0039) model_time 0.9321 (1.0031) loss 0.9116 (0.8299) grad_norm 10.4811 (8.5202/1.8119) mem 68106MB [2022-12-20 04:43:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][980/1519] eta 0:09:01 lr 0.000014 time 0.9302 (1.0039) model_time 0.9301 (1.0031) loss 0.7887 (0.8299) grad_norm 7.4445 (8.5266/1.8000) mem 68106MB [2022-12-20 04:44:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][990/1519] eta 0:08:51 lr 0.000014 time 0.9381 (1.0039) model_time 0.9379 (1.0031) loss 0.6927 (0.8296) grad_norm 11.9299 (8.5271/1.8161) mem 68106MB [2022-12-20 04:44:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1000/1519] eta 0:08:40 lr 0.000014 time 0.9217 (1.0038) model_time 0.9216 (1.0031) loss 0.7608 (0.8293) grad_norm 7.7756 (8.5191/1.8268) mem 68106MB [2022-12-20 04:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1010/1519] eta 0:08:30 lr 0.000014 time 0.9234 (1.0038) model_time 0.9233 (1.0031) loss 0.7105 (0.8289) grad_norm 6.0656 (8.5326/1.8427) mem 68106MB [2022-12-20 04:44:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1020/1519] eta 0:08:20 lr 0.000014 time 0.9203 (1.0039) model_time 0.9201 (1.0031) loss 0.8277 (0.8292) grad_norm 10.1356 (8.5413/1.8266) mem 68106MB [2022-12-20 04:44:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1030/1519] eta 0:08:10 lr 0.000014 time 0.9301 (1.0039) model_time 0.9300 (1.0032) loss 0.6869 (0.8294) grad_norm 9.5582 (8.5497/1.8401) mem 68106MB [2022-12-20 04:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1040/1519] eta 0:08:00 lr 0.000014 time 0.9318 (1.0039) model_time 0.9317 (1.0032) loss 0.7514 (0.8295) grad_norm 6.3900 (8.5386/1.8487) mem 68106MB [2022-12-20 04:45:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1050/1519] eta 0:07:50 lr 0.000014 time 0.9264 (1.0039) model_time 0.9262 (1.0031) loss 0.9937 (0.8297) grad_norm 6.4673 (8.5409/1.8327) mem 68106MB [2022-12-20 04:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1060/1519] eta 0:07:40 lr 0.000014 time 0.9194 (1.0038) model_time 0.9192 (1.0031) loss 0.6740 (0.8294) grad_norm 10.6784 (8.5320/1.8149) mem 68106MB [2022-12-20 04:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1070/1519] eta 0:07:30 lr 0.000014 time 0.9203 (1.0038) model_time 0.9201 (1.0031) loss 1.3381 (0.8296) grad_norm 7.3198 (8.5235/1.8163) mem 68106MB [2022-12-20 04:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1080/1519] eta 0:07:20 lr 0.000014 time 0.9989 (1.0039) model_time 0.9987 (1.0031) loss 0.6653 (0.8298) grad_norm 10.1073 (8.5349/1.8227) mem 68106MB [2022-12-20 04:45:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1090/1519] eta 0:07:10 lr 0.000014 time 0.9220 (1.0038) model_time 0.9218 (1.0031) loss 0.6831 (0.8293) grad_norm 7.5369 (8.5187/1.8127) mem 68106MB [2022-12-20 04:45:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1100/1519] eta 0:07:00 lr 0.000014 time 0.9363 (1.0039) model_time 0.9361 (1.0032) loss 0.9392 (0.8296) grad_norm 7.0975 (8.5091/1.8183) mem 68106MB [2022-12-20 04:46:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1110/1519] eta 0:06:50 lr 0.000014 time 0.9276 (1.0039) model_time 0.9274 (1.0032) loss 0.6725 (0.8291) grad_norm 10.6414 (8.5045/1.8123) mem 68106MB [2022-12-20 04:46:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1120/1519] eta 0:06:40 lr 0.000014 time 0.9377 (1.0038) model_time 0.9375 (1.0031) loss 0.7044 (0.8292) grad_norm 8.3563 (8.5028/1.8098) mem 68106MB [2022-12-20 04:46:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1130/1519] eta 0:06:30 lr 0.000014 time 0.9239 (1.0038) model_time 0.9238 (1.0031) loss 0.7256 (0.8295) grad_norm 8.2413 (8.4979/1.7995) mem 68106MB [2022-12-20 04:46:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1140/1519] eta 0:06:20 lr 0.000014 time 0.9195 (1.0037) model_time 0.9193 (1.0030) loss 0.6881 (0.8291) grad_norm 7.5141 (8.4671/1.7672) mem 68106MB [2022-12-20 04:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1150/1519] eta 0:06:10 lr 0.000014 time 0.9202 (1.0037) model_time 0.9200 (1.0030) loss 0.7760 (0.8286) grad_norm 12.3479 (8.5054/1.7785) mem 68106MB [2022-12-20 04:46:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1160/1519] eta 0:06:00 lr 0.000014 time 0.9240 (1.0037) model_time 0.9239 (1.0030) loss 0.9141 (0.8283) grad_norm 12.4993 (8.5070/1.7998) mem 68106MB [2022-12-20 04:47:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1170/1519] eta 0:05:50 lr 0.000014 time 0.9220 (1.0038) model_time 0.9219 (1.0031) loss 0.9864 (0.8284) grad_norm 8.0482 (8.5003/1.7848) mem 68106MB [2022-12-20 04:47:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1180/1519] eta 0:05:40 lr 0.000014 time 0.9232 (1.0038) model_time 0.9231 (1.0031) loss 0.7225 (0.8285) grad_norm 7.9267 (8.5234/1.8018) mem 68106MB [2022-12-20 04:47:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1190/1519] eta 0:05:30 lr 0.000014 time 0.9224 (1.0041) model_time 0.9222 (1.0034) loss 0.7942 (0.8288) grad_norm 8.7361 (8.5184/1.8040) mem 68106MB [2022-12-20 04:47:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1200/1519] eta 0:05:20 lr 0.000014 time 0.9333 (1.0040) model_time 0.9331 (1.0033) loss 0.7474 (0.8287) grad_norm 10.3844 (8.5088/1.7894) mem 68106MB [2022-12-20 04:47:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1210/1519] eta 0:05:10 lr 0.000014 time 0.9153 (1.0040) model_time 0.9152 (1.0033) loss 1.0439 (0.8290) grad_norm 9.0741 (8.5141/1.8029) mem 68106MB [2022-12-20 04:47:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1220/1519] eta 0:05:00 lr 0.000014 time 0.9245 (1.0040) model_time 0.9244 (1.0033) loss 1.1076 (0.8288) grad_norm 8.2971 (8.4980/1.7513) mem 68106MB [2022-12-20 04:48:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1230/1519] eta 0:04:50 lr 0.000014 time 0.9186 (1.0039) model_time 0.9184 (1.0033) loss 0.8371 (0.8291) grad_norm 7.3217 (8.4774/1.7521) mem 68106MB [2022-12-20 04:48:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1240/1519] eta 0:04:40 lr 0.000014 time 0.9159 (1.0039) model_time 0.9158 (1.0032) loss 0.7123 (0.8292) grad_norm 6.8087 (8.4571/1.7562) mem 68106MB [2022-12-20 04:48:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1250/1519] eta 0:04:30 lr 0.000014 time 0.9313 (1.0039) model_time 0.9312 (1.0032) loss 0.6885 (0.8288) grad_norm 10.4607 (8.4637/1.7458) mem 68106MB [2022-12-20 04:48:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1260/1519] eta 0:04:20 lr 0.000014 time 1.0061 (1.0040) model_time 1.0059 (1.0033) loss 0.7459 (0.8282) grad_norm 7.0187 (8.4695/1.7324) mem 68106MB [2022-12-20 04:48:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1270/1519] eta 0:04:10 lr 0.000014 time 0.9220 (1.0040) model_time 0.9218 (1.0034) loss 0.7697 (0.8278) grad_norm 6.5922 (8.4745/1.7299) mem 68106MB [2022-12-20 04:48:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1280/1519] eta 0:03:59 lr 0.000014 time 0.9276 (1.0040) model_time 0.9275 (1.0033) loss 0.9752 (0.8285) grad_norm 8.2218 (8.4811/1.7233) mem 68106MB [2022-12-20 04:49:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1290/1519] eta 0:03:49 lr 0.000014 time 0.9196 (1.0040) model_time 0.9195 (1.0033) loss 0.8130 (0.8293) grad_norm 7.6866 (8.4720/1.7138) mem 68106MB [2022-12-20 04:49:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1300/1519] eta 0:03:39 lr 0.000014 time 0.9354 (1.0039) model_time 0.9352 (1.0033) loss 0.7048 (0.8295) grad_norm 7.6514 (8.4591/1.7099) mem 68106MB [2022-12-20 04:49:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1310/1519] eta 0:03:29 lr 0.000014 time 0.9219 (1.0039) model_time 0.9218 (1.0033) loss 0.7798 (0.8294) grad_norm 8.5502 (8.4878/1.7520) mem 68106MB [2022-12-20 04:49:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1320/1519] eta 0:03:19 lr 0.000014 time 0.9304 (1.0039) model_time 0.9303 (1.0032) loss 1.0518 (0.8291) grad_norm 7.4974 (8.4820/1.7489) mem 68106MB [2022-12-20 04:49:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1330/1519] eta 0:03:09 lr 0.000014 time 0.9210 (1.0039) model_time 0.9208 (1.0033) loss 0.7571 (0.8291) grad_norm 8.0918 (8.4869/1.7403) mem 68106MB [2022-12-20 04:49:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1340/1519] eta 0:02:59 lr 0.000014 time 0.9304 (1.0040) model_time 0.9302 (1.0033) loss 0.7494 (0.8292) grad_norm 8.7680 (8.4873/1.7404) mem 68106MB [2022-12-20 04:50:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1350/1519] eta 0:02:49 lr 0.000014 time 0.9407 (1.0040) model_time 0.9406 (1.0034) loss 0.7752 (0.8289) grad_norm 7.1459 (8.4465/1.6937) mem 68106MB [2022-12-20 04:50:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1360/1519] eta 0:02:39 lr 0.000014 time 0.9252 (1.0040) model_time 0.9251 (1.0033) loss 0.6882 (0.8288) grad_norm 7.2115 (8.4203/1.6769) mem 68106MB [2022-12-20 04:50:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1370/1519] eta 0:02:29 lr 0.000014 time 0.9316 (1.0039) model_time 0.9314 (1.0033) loss 1.4327 (0.8290) grad_norm 9.4802 (8.4268/1.6876) mem 68106MB [2022-12-20 04:50:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1380/1519] eta 0:02:19 lr 0.000014 time 0.9224 (1.0039) model_time 0.9223 (1.0033) loss 0.9205 (0.8284) grad_norm 8.2875 (8.4387/1.6833) mem 68106MB [2022-12-20 04:50:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1390/1519] eta 0:02:09 lr 0.000014 time 1.0027 (1.0039) model_time 1.0026 (1.0033) loss 0.9345 (0.8290) grad_norm 9.8265 (8.4542/1.6697) mem 68106MB [2022-12-20 04:50:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1400/1519] eta 0:01:59 lr 0.000014 time 0.9344 (1.0039) model_time 0.9343 (1.0033) loss 0.8030 (0.8291) grad_norm 7.5824 (8.4956/1.7224) mem 68106MB [2022-12-20 04:51:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1410/1519] eta 0:01:49 lr 0.000014 time 0.9214 (1.0039) model_time 0.9212 (1.0033) loss 0.7363 (0.8292) grad_norm 10.0184 (8.4740/1.6832) mem 68106MB [2022-12-20 04:51:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1420/1519] eta 0:01:39 lr 0.000014 time 0.9289 (1.0039) model_time 0.9288 (1.0033) loss 0.7857 (0.8295) grad_norm 7.2162 (8.4685/1.6797) mem 68106MB [2022-12-20 04:51:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1430/1519] eta 0:01:29 lr 0.000014 time 0.9318 (1.0039) model_time 0.9317 (1.0032) loss 0.8622 (0.8296) grad_norm 7.9761 (8.5121/1.6827) mem 68106MB [2022-12-20 04:51:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1440/1519] eta 0:01:19 lr 0.000014 time 0.9237 (1.0039) model_time 0.9236 (1.0033) loss 0.6955 (0.8296) grad_norm 7.4904 (8.5200/1.6954) mem 68106MB [2022-12-20 04:51:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1450/1519] eta 0:01:09 lr 0.000014 time 0.9286 (1.0039) model_time 0.9285 (1.0033) loss 0.7107 (0.8296) grad_norm 6.5310 (8.5226/1.6983) mem 68106MB [2022-12-20 04:51:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1460/1519] eta 0:00:59 lr 0.000014 time 0.9278 (1.0039) model_time 0.9276 (1.0033) loss 1.0077 (0.8297) grad_norm 10.2048 (8.5085/1.7062) mem 68106MB [2022-12-20 04:52:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1470/1519] eta 0:00:49 lr 0.000014 time 0.9456 (1.0039) model_time 0.9454 (1.0033) loss 0.9161 (0.8300) grad_norm 9.9603 (8.5156/1.7084) mem 68106MB [2022-12-20 04:52:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1480/1519] eta 0:00:39 lr 0.000014 time 0.9042 (1.0039) model_time 0.9040 (1.0033) loss 0.7593 (0.8299) grad_norm 14.3712 (8.5598/1.7641) mem 68106MB [2022-12-20 04:52:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1490/1519] eta 0:00:29 lr 0.000014 time 0.9369 (1.0038) model_time 0.9368 (1.0032) loss 0.8334 (0.8297) grad_norm 9.0218 (8.5641/1.7591) mem 68106MB [2022-12-20 04:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1500/1519] eta 0:00:19 lr 0.000014 time 1.0325 (1.0039) model_time 1.0323 (1.0033) loss 0.7376 (0.8295) grad_norm 7.9520 (8.5439/1.7554) mem 68106MB [2022-12-20 04:52:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [55/100][1510/1519] eta 0:00:09 lr 0.000014 time 0.9193 (1.0038) model_time 0.9192 (1.0032) loss 0.6772 (0.8290) grad_norm 8.3961 (8.5565/1.7452) mem 68106MB [2022-12-20 04:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 55 training takes 0:25:24 [2022-12-20 04:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_55.pth saving...... [2022-12-20 04:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_55.pth saved !!! [2022-12-20 04:53:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.631 (0.631) Loss 0.5183 (0.5183) Acc@1 90.625 (90.625) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 04:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.327) Loss 0.4954 (0.4855) Acc@1 93.403 (92.771) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 04:53:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4514 (0.4846) Acc@1 93.403 (92.659) Acc@5 98.958 (98.429) Mem 68106MB [2022-12-20 04:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.6126 (0.4920) Acc@1 89.236 (92.428) Acc@5 97.569 (98.398) Mem 68106MB [2022-12-20 04:53:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.308) Loss 0.4465 (0.4840) Acc@1 93.056 (92.471) Acc@5 98.958 (98.467) Mem 68106MB [2022-12-20 04:53:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.307) Loss 0.4826 (0.4829) Acc@1 90.625 (92.456) Acc@5 99.653 (98.536) Mem 68106MB [2022-12-20 04:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.5088 (0.4834) Acc@1 91.667 (92.401) Acc@5 97.917 (98.486) Mem 68106MB [2022-12-20 04:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5274 (0.4839) Acc@1 92.014 (92.361) Acc@5 98.264 (98.484) Mem 68106MB [2022-12-20 04:53:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.303) Loss 0.4132 (0.4828) Acc@1 93.403 (92.344) Acc@5 98.958 (98.530) Mem 68106MB [2022-12-20 04:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:55] * Acc@1 92.301 Acc@5 98.527 [2022-12-20 04:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.3% [2022-12-20 04:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 04:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 04:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.30% [2022-12-20 04:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][0/1519] eta 0:35:47 lr 0.000014 time 1.4138 (1.4138) model_time 0.9890 (0.9890) loss 0.7996 (0.7996) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 04:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][10/1519] eta 0:26:13 lr 0.000014 time 0.9250 (1.0428) model_time 0.9248 (1.0039) loss 0.6738 (0.7808) grad_norm 7.4501 (9.6926/1.2696) mem 68106MB [2022-12-20 04:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][20/1519] eta 0:25:34 lr 0.000014 time 0.9735 (1.0236) model_time 0.9734 (1.0031) loss 0.7336 (0.7860) grad_norm 10.2511 (8.9936/1.4267) mem 68106MB [2022-12-20 04:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][30/1519] eta 0:25:12 lr 0.000014 time 0.9273 (1.0157) model_time 0.9272 (1.0018) loss 0.6747 (0.7864) grad_norm 7.9199 (8.7897/1.3395) mem 68106MB [2022-12-20 04:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][40/1519] eta 0:24:55 lr 0.000014 time 0.9256 (1.0111) model_time 0.9255 (1.0005) loss 0.7400 (0.8171) grad_norm 7.0713 (8.8129/1.5673) mem 68106MB [2022-12-20 04:55:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][50/1519] eta 0:24:41 lr 0.000014 time 0.9207 (1.0084) model_time 0.9205 (0.9998) loss 1.1060 (0.8268) grad_norm 5.7716 (8.5173/1.6369) mem 68106MB [2022-12-20 04:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][60/1519] eta 0:24:31 lr 0.000014 time 0.9955 (1.0085) model_time 0.9953 (1.0012) loss 0.6710 (0.8199) grad_norm 8.2074 (8.3547/1.6045) mem 68106MB [2022-12-20 04:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][70/1519] eta 0:24:18 lr 0.000014 time 0.9199 (1.0069) model_time 0.9198 (1.0006) loss 0.8986 (0.8198) grad_norm 7.9522 (8.6082/1.9388) mem 68106MB [2022-12-20 04:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][80/1519] eta 0:24:07 lr 0.000014 time 0.9560 (1.0059) model_time 0.9558 (1.0003) loss 0.8182 (0.8180) grad_norm 14.9326 (8.6882/2.1277) mem 68106MB [2022-12-20 04:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][90/1519] eta 0:24:00 lr 0.000014 time 0.9187 (1.0077) model_time 0.9186 (1.0027) loss 0.6766 (0.8209) grad_norm 7.2718 (8.6341/2.0278) mem 68106MB [2022-12-20 04:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][100/1519] eta 0:23:48 lr 0.000014 time 0.9223 (1.0069) model_time 0.9222 (1.0024) loss 0.9029 (0.8187) grad_norm 7.2148 (8.6658/2.0349) mem 68106MB [2022-12-20 04:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][110/1519] eta 0:23:37 lr 0.000014 time 0.9277 (1.0059) model_time 0.9276 (1.0018) loss 0.7223 (0.8159) grad_norm 8.0375 (8.5878/1.9850) mem 68106MB [2022-12-20 04:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][120/1519] eta 0:23:28 lr 0.000014 time 0.9123 (1.0068) model_time 0.9122 (1.0030) loss 0.7090 (0.8183) grad_norm 9.1949 (8.6911/2.1012) mem 68106MB [2022-12-20 04:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][130/1519] eta 0:23:17 lr 0.000014 time 0.9249 (1.0062) model_time 0.9248 (1.0027) loss 1.2445 (0.8222) grad_norm 7.6787 (8.6532/2.0409) mem 68106MB [2022-12-20 04:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][140/1519] eta 0:23:08 lr 0.000014 time 0.9433 (1.0066) model_time 0.9431 (1.0033) loss 0.6785 (0.8234) grad_norm 8.9825 (8.7913/2.2051) mem 68106MB [2022-12-20 04:56:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][150/1519] eta 0:22:57 lr 0.000014 time 0.9207 (1.0063) model_time 0.9205 (1.0032) loss 0.7871 (0.8235) grad_norm 8.1877 (8.7567/2.1411) mem 68106MB [2022-12-20 04:56:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][160/1519] eta 0:22:46 lr 0.000014 time 0.9211 (1.0057) model_time 0.9210 (1.0028) loss 0.6703 (0.8208) grad_norm 7.4891 (8.7415/2.0983) mem 68106MB [2022-12-20 04:57:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][170/1519] eta 0:22:37 lr 0.000014 time 0.9385 (1.0062) model_time 0.9383 (1.0035) loss 0.6813 (0.8209) grad_norm 11.5906 (8.7395/2.0891) mem 68106MB [2022-12-20 04:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][180/1519] eta 0:22:27 lr 0.000014 time 0.9344 (1.0060) model_time 0.9343 (1.0034) loss 0.7496 (0.8197) grad_norm 7.2815 (8.7313/2.0877) mem 68106MB [2022-12-20 04:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][190/1519] eta 0:22:16 lr 0.000014 time 0.9253 (1.0058) model_time 0.9252 (1.0033) loss 0.8844 (0.8239) grad_norm 11.3351 (8.7120/2.0935) mem 68106MB [2022-12-20 04:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][200/1519] eta 0:22:06 lr 0.000014 time 0.9313 (1.0056) model_time 0.9311 (1.0032) loss 0.7357 (0.8242) grad_norm 8.9980 (8.7172/2.0461) mem 68106MB [2022-12-20 04:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][210/1519] eta 0:21:55 lr 0.000014 time 0.9318 (1.0053) model_time 0.9317 (1.0030) loss 0.7128 (0.8251) grad_norm 9.0629 (8.7087/2.0259) mem 68106MB [2022-12-20 04:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][220/1519] eta 0:21:45 lr 0.000014 time 0.9382 (1.0051) model_time 0.9380 (1.0028) loss 0.7356 (0.8225) grad_norm 6.2478 (8.6532/2.0060) mem 68106MB [2022-12-20 04:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][230/1519] eta 0:21:35 lr 0.000014 time 0.9295 (1.0051) model_time 0.9294 (1.0029) loss 0.6856 (0.8236) grad_norm 8.7073 (8.6248/1.9904) mem 68106MB [2022-12-20 04:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][240/1519] eta 0:21:25 lr 0.000014 time 0.9248 (1.0048) model_time 0.9246 (1.0027) loss 0.8541 (0.8259) grad_norm 7.0899 (8.6091/1.9549) mem 68106MB [2022-12-20 04:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][250/1519] eta 0:21:14 lr 0.000014 time 0.9323 (1.0046) model_time 0.9321 (1.0026) loss 0.9596 (0.8262) grad_norm 8.7324 (8.5811/1.9264) mem 68106MB [2022-12-20 04:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][260/1519] eta 0:21:04 lr 0.000014 time 0.9239 (1.0046) model_time 0.9237 (1.0027) loss 0.7191 (0.8264) grad_norm 11.8143 (8.5750/1.9279) mem 68106MB [2022-12-20 04:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][270/1519] eta 0:20:54 lr 0.000014 time 0.9013 (1.0045) model_time 0.9011 (1.0026) loss 1.0139 (0.8288) grad_norm 7.4974 (8.6283/2.0044) mem 68106MB [2022-12-20 04:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][280/1519] eta 0:20:44 lr 0.000014 time 0.9177 (1.0044) model_time 0.9175 (1.0026) loss 0.6862 (0.8283) grad_norm 9.2625 (8.6335/1.9950) mem 68106MB [2022-12-20 04:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][290/1519] eta 0:20:34 lr 0.000014 time 0.9188 (1.0047) model_time 0.9187 (1.0029) loss 0.7524 (0.8280) grad_norm 6.7725 (8.6101/1.9758) mem 68106MB [2022-12-20 04:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][300/1519] eta 0:20:25 lr 0.000014 time 0.9248 (1.0052) model_time 0.9246 (1.0035) loss 0.8606 (0.8284) grad_norm 8.2759 (8.5976/1.9523) mem 68106MB [2022-12-20 04:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][310/1519] eta 0:20:15 lr 0.000014 time 0.9855 (1.0053) model_time 0.9854 (1.0036) loss 0.8166 (0.8300) grad_norm 10.5720 (8.5962/1.9311) mem 68106MB [2022-12-20 04:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][320/1519] eta 0:20:05 lr 0.000014 time 0.9881 (1.0054) model_time 0.9880 (1.0037) loss 0.6943 (0.8290) grad_norm 7.6560 (8.6325/1.9658) mem 68106MB [2022-12-20 04:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][330/1519] eta 0:19:55 lr 0.000014 time 0.9248 (1.0051) model_time 0.9247 (1.0035) loss 0.8491 (0.8291) grad_norm 6.2515 (8.6087/1.9610) mem 68106MB [2022-12-20 04:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][340/1519] eta 0:19:44 lr 0.000014 time 0.9202 (1.0050) model_time 0.9200 (1.0035) loss 0.9754 (0.8304) grad_norm 10.4153 (8.5982/1.9545) mem 68106MB [2022-12-20 05:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][350/1519] eta 0:19:34 lr 0.000014 time 0.9260 (1.0050) model_time 0.9259 (1.0034) loss 0.9030 (0.8318) grad_norm 11.6979 (8.6053/1.9433) mem 68106MB [2022-12-20 05:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][360/1519] eta 0:19:24 lr 0.000014 time 0.9232 (1.0048) model_time 0.9230 (1.0033) loss 0.8462 (0.8297) grad_norm 9.5330 (8.6055/1.9227) mem 68106MB [2022-12-20 05:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][370/1519] eta 0:19:14 lr 0.000014 time 0.9193 (1.0047) model_time 0.9191 (1.0032) loss 1.0675 (0.8303) grad_norm 7.8177 (8.5955/1.9033) mem 68106MB [2022-12-20 05:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][380/1519] eta 0:19:04 lr 0.000014 time 0.9274 (1.0048) model_time 0.9272 (1.0033) loss 0.6906 (0.8291) grad_norm 8.2746 (8.5795/1.8850) mem 68106MB [2022-12-20 05:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][390/1519] eta 0:18:54 lr 0.000014 time 0.9417 (1.0047) model_time 0.9416 (1.0033) loss 0.8850 (0.8306) grad_norm 9.4163 (8.5579/1.8756) mem 68106MB [2022-12-20 05:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][400/1519] eta 0:18:44 lr 0.000014 time 0.9791 (1.0049) model_time 0.9789 (1.0035) loss 0.8947 (0.8309) grad_norm 7.0699 (8.5588/1.8637) mem 68106MB [2022-12-20 05:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][410/1519] eta 0:18:34 lr 0.000014 time 0.9195 (1.0046) model_time 0.9194 (1.0033) loss 0.9588 (0.8312) grad_norm 9.5123 (8.5844/1.8850) mem 68106MB [2022-12-20 05:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][420/1519] eta 0:18:23 lr 0.000014 time 0.9261 (1.0045) model_time 0.9259 (1.0032) loss 0.7801 (0.8316) grad_norm 8.9965 (8.6148/1.8908) mem 68106MB [2022-12-20 05:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][430/1519] eta 0:18:13 lr 0.000014 time 0.9252 (1.0045) model_time 0.9251 (1.0032) loss 0.6626 (0.8300) grad_norm 7.2253 (8.6026/1.8717) mem 68106MB [2022-12-20 05:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][440/1519] eta 0:18:03 lr 0.000014 time 0.9204 (1.0045) model_time 0.9203 (1.0032) loss 1.0724 (0.8297) grad_norm 8.5447 (8.6024/1.8531) mem 68106MB [2022-12-20 05:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][450/1519] eta 0:17:53 lr 0.000014 time 0.9743 (1.0045) model_time 0.9742 (1.0032) loss 0.8432 (0.8300) grad_norm 7.8755 (8.5952/1.8445) mem 68106MB [2022-12-20 05:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][460/1519] eta 0:17:44 lr 0.000014 time 0.9244 (1.0047) model_time 0.9243 (1.0035) loss 0.7824 (0.8302) grad_norm 6.8417 (8.5884/1.8426) mem 68106MB [2022-12-20 05:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][470/1519] eta 0:17:33 lr 0.000014 time 0.9286 (1.0046) model_time 0.9285 (1.0034) loss 0.7883 (0.8302) grad_norm 13.2546 (8.5900/1.8628) mem 68106MB [2022-12-20 05:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][480/1519] eta 0:17:23 lr 0.000014 time 0.9225 (1.0047) model_time 0.9224 (1.0035) loss 0.8330 (0.8303) grad_norm 7.1140 (8.5692/1.8580) mem 68106MB [2022-12-20 05:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][490/1519] eta 0:17:13 lr 0.000014 time 0.9282 (1.0045) model_time 0.9280 (1.0034) loss 0.8704 (0.8295) grad_norm 8.3763 (8.5528/1.8521) mem 68106MB [2022-12-20 05:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][500/1519] eta 0:17:03 lr 0.000014 time 1.0107 (1.0047) model_time 1.0106 (1.0035) loss 0.9330 (0.8294) grad_norm 8.2285 (8.5527/1.8344) mem 68106MB [2022-12-20 05:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][510/1519] eta 0:16:53 lr 0.000014 time 0.9296 (1.0046) model_time 0.9295 (1.0034) loss 0.8233 (0.8296) grad_norm 7.3599 (8.5392/1.8224) mem 68106MB [2022-12-20 05:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][520/1519] eta 0:16:43 lr 0.000014 time 0.9205 (1.0044) model_time 0.9203 (1.0033) loss 0.8915 (0.8292) grad_norm 8.0814 (8.5531/1.8160) mem 68106MB [2022-12-20 05:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][530/1519] eta 0:16:33 lr 0.000014 time 0.9286 (1.0043) model_time 0.9285 (1.0032) loss 0.7347 (0.8292) grad_norm 5.8666 (8.5586/1.8187) mem 68106MB [2022-12-20 05:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][540/1519] eta 0:16:23 lr 0.000014 time 1.0771 (1.0044) model_time 1.0769 (1.0033) loss 0.7392 (0.8289) grad_norm 8.8885 (8.5412/1.8100) mem 68106MB [2022-12-20 05:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][550/1519] eta 0:16:13 lr 0.000014 time 0.9384 (1.0045) model_time 0.9383 (1.0034) loss 1.3102 (0.8289) grad_norm 6.9676 (8.5466/1.8129) mem 68106MB [2022-12-20 05:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][560/1519] eta 0:16:03 lr 0.000014 time 0.9249 (1.0049) model_time 0.9248 (1.0039) loss 0.9788 (0.8300) grad_norm 6.2653 (8.5317/1.8082) mem 68106MB [2022-12-20 05:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][570/1519] eta 0:15:53 lr 0.000014 time 0.9336 (1.0048) model_time 0.9335 (1.0038) loss 0.9785 (0.8304) grad_norm 7.7158 (8.5297/1.7951) mem 68106MB [2022-12-20 05:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][580/1519] eta 0:15:43 lr 0.000014 time 0.9231 (1.0053) model_time 0.9229 (1.0042) loss 0.8585 (0.8315) grad_norm 10.7088 (8.5488/1.8037) mem 68106MB [2022-12-20 05:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][590/1519] eta 0:15:33 lr 0.000014 time 0.9283 (1.0053) model_time 0.9281 (1.0043) loss 0.7180 (0.8310) grad_norm 6.6759 (8.5359/1.7963) mem 68106MB [2022-12-20 05:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][600/1519] eta 0:15:23 lr 0.000014 time 0.9608 (1.0054) model_time 0.9606 (1.0044) loss 0.9914 (0.8306) grad_norm 8.0454 (8.5351/1.7863) mem 68106MB [2022-12-20 05:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][610/1519] eta 0:15:13 lr 0.000014 time 0.9374 (1.0053) model_time 0.9372 (1.0043) loss 0.7426 (0.8306) grad_norm 7.3857 (8.5261/1.7915) mem 68106MB [2022-12-20 05:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][620/1519] eta 0:15:03 lr 0.000014 time 0.9253 (1.0054) model_time 0.9251 (1.0044) loss 0.9694 (0.8323) grad_norm 9.3757 (8.5343/1.7863) mem 68106MB [2022-12-20 05:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][630/1519] eta 0:14:53 lr 0.000014 time 0.9972 (1.0055) model_time 0.9970 (1.0045) loss 0.8232 (0.8320) grad_norm 7.4793 (8.5400/1.8018) mem 68106MB [2022-12-20 05:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][640/1519] eta 0:14:43 lr 0.000014 time 0.9201 (1.0054) model_time 0.9200 (1.0044) loss 0.6762 (0.8319) grad_norm 10.3490 (8.5399/1.7947) mem 68106MB [2022-12-20 05:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][650/1519] eta 0:14:33 lr 0.000014 time 0.9241 (1.0053) model_time 0.9240 (1.0043) loss 1.0644 (0.8323) grad_norm 6.2989 (8.5895/1.8163) mem 68106MB [2022-12-20 05:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][660/1519] eta 0:14:23 lr 0.000014 time 0.9246 (1.0052) model_time 0.9244 (1.0043) loss 0.9886 (0.8324) grad_norm 9.3901 (8.5980/1.8226) mem 68106MB [2022-12-20 05:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][670/1519] eta 0:14:13 lr 0.000014 time 0.9254 (1.0052) model_time 0.9253 (1.0042) loss 0.7831 (0.8314) grad_norm 6.9088 (8.5864/1.7998) mem 68106MB [2022-12-20 05:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][680/1519] eta 0:14:03 lr 0.000014 time 0.9166 (1.0051) model_time 0.9165 (1.0042) loss 0.7988 (0.8314) grad_norm 6.5305 (8.5614/1.7626) mem 68106MB [2022-12-20 05:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][690/1519] eta 0:13:53 lr 0.000014 time 0.9354 (1.0050) model_time 0.9353 (1.0041) loss 0.7077 (0.8309) grad_norm 7.4910 (8.5457/1.7707) mem 68106MB [2022-12-20 05:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][700/1519] eta 0:13:43 lr 0.000014 time 0.9258 (1.0050) model_time 0.9256 (1.0040) loss 0.8229 (0.8296) grad_norm 8.7235 (8.5481/1.7595) mem 68106MB [2022-12-20 05:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][710/1519] eta 0:13:32 lr 0.000014 time 0.9190 (1.0049) model_time 0.9189 (1.0039) loss 0.8510 (0.8294) grad_norm 9.6718 (8.5768/1.7576) mem 68106MB [2022-12-20 05:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][720/1519] eta 0:13:22 lr 0.000014 time 0.9810 (1.0050) model_time 0.9809 (1.0040) loss 0.7806 (0.8291) grad_norm 11.0474 (8.5768/1.7209) mem 68106MB [2022-12-20 05:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][730/1519] eta 0:13:12 lr 0.000014 time 0.9348 (1.0049) model_time 0.9347 (1.0040) loss 0.8008 (0.8293) grad_norm 8.1507 (8.5787/1.7206) mem 68106MB [2022-12-20 05:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][740/1519] eta 0:13:02 lr 0.000014 time 0.9469 (1.0049) model_time 0.9468 (1.0040) loss 0.7179 (0.8289) grad_norm 8.7629 (8.5546/1.6538) mem 68106MB [2022-12-20 05:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][750/1519] eta 0:12:52 lr 0.000014 time 0.9328 (1.0049) model_time 0.9326 (1.0041) loss 0.9850 (0.8303) grad_norm 7.0194 (8.5561/1.6578) mem 68106MB [2022-12-20 05:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][760/1519] eta 0:12:42 lr 0.000014 time 0.9243 (1.0049) model_time 0.9241 (1.0040) loss 0.6937 (0.8300) grad_norm 11.3243 (8.5611/1.6599) mem 68106MB [2022-12-20 05:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][770/1519] eta 0:12:32 lr 0.000014 time 0.9243 (1.0050) model_time 0.9242 (1.0042) loss 0.7069 (0.8298) grad_norm 6.3325 (8.5299/1.6594) mem 68106MB [2022-12-20 05:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][780/1519] eta 0:12:22 lr 0.000014 time 0.9650 (1.0051) model_time 0.9649 (1.0042) loss 0.6792 (0.8286) grad_norm 15.4469 (8.5498/1.6909) mem 68106MB [2022-12-20 05:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][790/1519] eta 0:12:12 lr 0.000014 time 0.9247 (1.0050) model_time 0.9245 (1.0042) loss 1.0023 (0.8286) grad_norm 8.2404 (8.5489/1.6706) mem 68106MB [2022-12-20 05:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][800/1519] eta 0:12:02 lr 0.000014 time 0.9247 (1.0052) model_time 0.9246 (1.0043) loss 0.7092 (0.8287) grad_norm 7.2186 (8.5518/1.6767) mem 68106MB [2022-12-20 05:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][810/1519] eta 0:11:52 lr 0.000014 time 0.9267 (1.0052) model_time 0.9266 (1.0044) loss 0.6709 (0.8284) grad_norm 9.3030 (8.5611/1.6768) mem 68106MB [2022-12-20 05:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][820/1519] eta 0:11:42 lr 0.000014 time 0.9256 (1.0052) model_time 0.9255 (1.0044) loss 1.0053 (0.8284) grad_norm 8.5125 (8.5633/1.6722) mem 68106MB [2022-12-20 05:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][830/1519] eta 0:11:32 lr 0.000014 time 0.9281 (1.0051) model_time 0.9280 (1.0043) loss 0.6869 (0.8278) grad_norm 8.9218 (8.5892/1.6786) mem 68106MB [2022-12-20 05:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][840/1519] eta 0:11:22 lr 0.000014 time 0.9241 (1.0051) model_time 0.9240 (1.0043) loss 0.8110 (0.8280) grad_norm 8.5376 (8.5925/1.6763) mem 68106MB [2022-12-20 05:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][850/1519] eta 0:11:12 lr 0.000014 time 0.9226 (1.0050) model_time 0.9225 (1.0042) loss 1.1745 (0.8278) grad_norm 7.1507 (8.6038/1.6964) mem 68106MB [2022-12-20 05:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][860/1519] eta 0:11:02 lr 0.000014 time 0.9074 (1.0050) model_time 0.9072 (1.0042) loss 0.7536 (0.8276) grad_norm 6.7780 (8.6008/1.6909) mem 68106MB [2022-12-20 05:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][870/1519] eta 0:10:52 lr 0.000014 time 0.9329 (1.0050) model_time 0.9328 (1.0042) loss 0.9080 (0.8274) grad_norm 9.1229 (8.5715/1.6345) mem 68106MB [2022-12-20 05:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][880/1519] eta 0:10:42 lr 0.000014 time 0.9250 (1.0050) model_time 0.9249 (1.0042) loss 0.6809 (0.8273) grad_norm 8.4555 (8.5528/1.6298) mem 68106MB [2022-12-20 05:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][890/1519] eta 0:10:32 lr 0.000014 time 0.9283 (1.0050) model_time 0.9281 (1.0042) loss 0.6796 (0.8266) grad_norm 7.3820 (8.5609/1.6339) mem 68106MB [2022-12-20 05:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][900/1519] eta 0:10:22 lr 0.000014 time 0.9267 (1.0053) model_time 0.9265 (1.0045) loss 0.8570 (0.8271) grad_norm 7.7898 (8.5694/1.6319) mem 68106MB [2022-12-20 05:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][910/1519] eta 0:10:12 lr 0.000014 time 0.9272 (1.0054) model_time 0.9270 (1.0046) loss 0.8671 (0.8273) grad_norm 7.4871 (8.5847/1.6639) mem 68106MB [2022-12-20 05:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][920/1519] eta 0:10:02 lr 0.000014 time 0.9228 (1.0053) model_time 0.9227 (1.0045) loss 0.8855 (0.8277) grad_norm 9.8225 (8.5782/1.6284) mem 68106MB [2022-12-20 05:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][930/1519] eta 0:09:52 lr 0.000014 time 0.9297 (1.0054) model_time 0.9295 (1.0046) loss 0.6661 (0.8273) grad_norm 9.0098 (8.5953/1.6332) mem 68106MB [2022-12-20 05:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][940/1519] eta 0:09:42 lr 0.000014 time 0.9217 (1.0054) model_time 0.9215 (1.0046) loss 1.1873 (0.8278) grad_norm 6.8016 (8.6013/1.6304) mem 68106MB [2022-12-20 05:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][950/1519] eta 0:09:32 lr 0.000014 time 0.9324 (1.0053) model_time 0.9323 (1.0046) loss 0.6894 (0.8280) grad_norm 9.8041 (8.6083/1.6576) mem 68106MB [2022-12-20 05:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][960/1519] eta 0:09:21 lr 0.000014 time 0.9300 (1.0053) model_time 0.9299 (1.0046) loss 0.6800 (0.8278) grad_norm 8.9017 (8.6240/1.6640) mem 68106MB [2022-12-20 05:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][970/1519] eta 0:09:11 lr 0.000014 time 0.9273 (1.0053) model_time 0.9271 (1.0045) loss 0.6734 (0.8273) grad_norm 11.8646 (8.6551/1.6945) mem 68106MB [2022-12-20 05:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][980/1519] eta 0:09:01 lr 0.000014 time 0.9292 (1.0053) model_time 0.9291 (1.0045) loss 0.8070 (0.8278) grad_norm 10.0059 (8.6595/1.7012) mem 68106MB [2022-12-20 05:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][990/1519] eta 0:08:51 lr 0.000014 time 0.9205 (1.0052) model_time 0.9204 (1.0045) loss 0.6797 (0.8279) grad_norm 6.8392 (8.6781/1.7039) mem 68106MB [2022-12-20 05:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1000/1519] eta 0:08:41 lr 0.000014 time 0.9246 (1.0052) model_time 0.9245 (1.0044) loss 0.6726 (0.8274) grad_norm 9.4240 (8.6806/1.7010) mem 68106MB [2022-12-20 05:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1010/1519] eta 0:08:31 lr 0.000014 time 0.9250 (1.0051) model_time 0.9249 (1.0044) loss 0.6824 (0.8270) grad_norm 8.9394 (8.6672/1.6743) mem 68106MB [2022-12-20 05:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1020/1519] eta 0:08:21 lr 0.000014 time 0.9209 (1.0051) model_time 0.9207 (1.0043) loss 0.6848 (0.8270) grad_norm 8.7781 (8.6336/1.6605) mem 68106MB [2022-12-20 05:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1030/1519] eta 0:08:11 lr 0.000014 time 0.9354 (1.0051) model_time 0.9353 (1.0043) loss 0.7717 (0.8270) grad_norm 11.3950 (8.6739/1.7031) mem 68106MB [2022-12-20 05:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1040/1519] eta 0:08:01 lr 0.000014 time 0.9303 (1.0051) model_time 0.9302 (1.0044) loss 0.6736 (0.8268) grad_norm 7.5081 (8.6578/1.7140) mem 68106MB [2022-12-20 05:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1050/1519] eta 0:07:51 lr 0.000014 time 0.9660 (1.0051) model_time 0.9659 (1.0044) loss 0.7026 (0.8270) grad_norm 10.5127 (8.6653/1.7108) mem 68106MB [2022-12-20 05:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1060/1519] eta 0:07:41 lr 0.000014 time 0.8907 (1.0052) model_time 0.8906 (1.0045) loss 0.6835 (0.8265) grad_norm 9.5197 (8.6731/1.7034) mem 68106MB [2022-12-20 05:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1070/1519] eta 0:07:31 lr 0.000014 time 0.9256 (1.0051) model_time 0.9254 (1.0044) loss 1.0410 (0.8267) grad_norm 7.6958 (8.6440/1.6899) mem 68106MB [2022-12-20 05:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1080/1519] eta 0:07:21 lr 0.000014 time 0.9873 (1.0052) model_time 0.9871 (1.0045) loss 0.7942 (0.8268) grad_norm 8.3969 (8.6792/1.7108) mem 68106MB [2022-12-20 05:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1090/1519] eta 0:07:11 lr 0.000014 time 0.9366 (1.0053) model_time 0.9364 (1.0046) loss 0.6992 (0.8269) grad_norm 7.2199 (8.6827/1.7099) mem 68106MB [2022-12-20 05:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1100/1519] eta 0:07:01 lr 0.000014 time 0.9280 (1.0053) model_time 0.9278 (1.0046) loss 1.0440 (0.8270) grad_norm 8.4263 (8.6595/1.7265) mem 68106MB [2022-12-20 05:12:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1110/1519] eta 0:06:51 lr 0.000014 time 0.9288 (1.0053) model_time 0.9286 (1.0047) loss 0.8518 (0.8268) grad_norm 7.8292 (8.6573/1.7445) mem 68106MB [2022-12-20 05:13:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1120/1519] eta 0:06:41 lr 0.000014 time 0.9183 (1.0054) model_time 0.9181 (1.0047) loss 1.0658 (0.8273) grad_norm 8.3336 (8.6381/1.7405) mem 68106MB [2022-12-20 05:13:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1130/1519] eta 0:06:31 lr 0.000014 time 0.9249 (1.0054) model_time 0.9248 (1.0047) loss 0.8732 (0.8280) grad_norm 9.0590 (8.6409/1.7310) mem 68106MB [2022-12-20 05:13:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1140/1519] eta 0:06:21 lr 0.000014 time 0.9202 (1.0054) model_time 0.9201 (1.0047) loss 0.7660 (0.8275) grad_norm 11.1057 (8.6569/1.7343) mem 68106MB [2022-12-20 05:13:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1150/1519] eta 0:06:10 lr 0.000014 time 0.9302 (1.0053) model_time 0.9301 (1.0046) loss 0.6774 (0.8272) grad_norm 8.9737 (8.6413/1.7275) mem 68106MB [2022-12-20 05:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1160/1519] eta 0:06:00 lr 0.000014 time 0.9324 (1.0053) model_time 0.9322 (1.0046) loss 0.7638 (0.8264) grad_norm 5.8359 (8.6408/1.7270) mem 68106MB [2022-12-20 05:13:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1170/1519] eta 0:05:50 lr 0.000014 time 0.9266 (1.0052) model_time 0.9265 (1.0045) loss 0.6972 (0.8260) grad_norm 6.3228 (8.6284/1.7349) mem 68106MB [2022-12-20 05:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1180/1519] eta 0:05:40 lr 0.000014 time 0.9262 (1.0052) model_time 0.9260 (1.0046) loss 0.7683 (0.8264) grad_norm 8.0620 (8.5903/1.7201) mem 68106MB [2022-12-20 05:14:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1190/1519] eta 0:05:30 lr 0.000014 time 0.9297 (1.0052) model_time 0.9296 (1.0045) loss 0.7375 (0.8259) grad_norm 7.0203 (8.5801/1.7217) mem 68106MB [2022-12-20 05:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1200/1519] eta 0:05:20 lr 0.000014 time 0.9371 (1.0053) model_time 0.9370 (1.0046) loss 0.7073 (0.8266) grad_norm 9.4840 (8.5770/1.7250) mem 68106MB [2022-12-20 05:14:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1210/1519] eta 0:05:10 lr 0.000014 time 0.9252 (1.0053) model_time 0.9250 (1.0046) loss 1.0734 (0.8268) grad_norm 6.4021 (8.5486/1.7213) mem 68106MB [2022-12-20 05:14:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1220/1519] eta 0:05:00 lr 0.000014 time 0.9337 (1.0052) model_time 0.9336 (1.0046) loss 1.0028 (0.8271) grad_norm 8.0429 (8.5384/1.7216) mem 68106MB [2022-12-20 05:14:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1230/1519] eta 0:04:50 lr 0.000013 time 0.9821 (1.0052) model_time 0.9819 (1.0046) loss 0.6670 (0.8267) grad_norm 9.8675 (8.5352/1.7055) mem 68106MB [2022-12-20 05:15:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1240/1519] eta 0:04:40 lr 0.000013 time 0.9257 (1.0055) model_time 0.9256 (1.0048) loss 0.9788 (0.8268) grad_norm 7.8796 (8.5289/1.6944) mem 68106MB [2022-12-20 05:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1250/1519] eta 0:04:30 lr 0.000013 time 0.9192 (1.0055) model_time 0.9191 (1.0048) loss 0.6933 (0.8272) grad_norm 22.4600 (8.5709/1.8644) mem 68106MB [2022-12-20 05:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1260/1519] eta 0:04:20 lr 0.000013 time 1.0720 (1.0056) model_time 1.0718 (1.0049) loss 0.6794 (0.8269) grad_norm 7.2339 (8.5870/1.8820) mem 68106MB [2022-12-20 05:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1270/1519] eta 0:04:10 lr 0.000013 time 0.9338 (1.0056) model_time 0.9337 (1.0049) loss 1.0474 (0.8272) grad_norm 7.7301 (8.5722/1.8658) mem 68106MB [2022-12-20 05:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1280/1519] eta 0:04:00 lr 0.000013 time 0.9271 (1.0055) model_time 0.9270 (1.0049) loss 1.0114 (0.8272) grad_norm 7.4393 (8.5723/1.8634) mem 68106MB [2022-12-20 05:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1290/1519] eta 0:03:50 lr 0.000013 time 0.9229 (1.0056) model_time 0.9227 (1.0049) loss 0.6903 (0.8272) grad_norm 8.6317 (8.5867/1.8588) mem 68106MB [2022-12-20 05:16:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1300/1519] eta 0:03:40 lr 0.000013 time 0.9228 (1.0055) model_time 0.9227 (1.0049) loss 0.7286 (0.8272) grad_norm 7.0564 (8.5875/1.8644) mem 68106MB [2022-12-20 05:16:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1310/1519] eta 0:03:30 lr 0.000013 time 0.9219 (1.0055) model_time 0.9217 (1.0048) loss 0.7560 (0.8268) grad_norm 11.0021 (8.5671/1.8711) mem 68106MB [2022-12-20 05:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1320/1519] eta 0:03:20 lr 0.000013 time 0.9220 (1.0054) model_time 0.9218 (1.0048) loss 0.8466 (0.8264) grad_norm 8.2063 (8.5465/1.8634) mem 68106MB [2022-12-20 05:16:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1330/1519] eta 0:03:10 lr 0.000013 time 0.9297 (1.0054) model_time 0.9295 (1.0048) loss 0.8052 (0.8262) grad_norm 9.4396 (8.5536/1.8616) mem 68106MB [2022-12-20 05:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1340/1519] eta 0:02:59 lr 0.000013 time 0.9225 (1.0053) model_time 0.9223 (1.0047) loss 1.0667 (0.8262) grad_norm 7.5582 (8.5254/1.8659) mem 68106MB [2022-12-20 05:16:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1350/1519] eta 0:02:49 lr 0.000013 time 0.9282 (1.0055) model_time 0.9281 (1.0048) loss 1.0415 (0.8261) grad_norm 7.6112 (8.5297/1.8748) mem 68106MB [2022-12-20 05:17:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1360/1519] eta 0:02:39 lr 0.000013 time 0.9565 (1.0054) model_time 0.9564 (1.0048) loss 1.2109 (0.8265) grad_norm 7.0136 (8.5173/1.8683) mem 68106MB [2022-12-20 05:17:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1370/1519] eta 0:02:29 lr 0.000013 time 0.9220 (1.0054) model_time 0.9219 (1.0047) loss 0.9081 (0.8267) grad_norm 8.7657 (8.5548/1.8777) mem 68106MB [2022-12-20 05:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1380/1519] eta 0:02:19 lr 0.000013 time 0.9332 (1.0053) model_time 0.9331 (1.0047) loss 0.6825 (0.8263) grad_norm 9.6382 (8.5726/1.8808) mem 68106MB [2022-12-20 05:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1390/1519] eta 0:02:09 lr 0.000013 time 0.9771 (1.0053) model_time 0.9770 (1.0047) loss 0.6810 (0.8264) grad_norm 9.2957 (8.5570/1.8387) mem 68106MB [2022-12-20 05:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1400/1519] eta 0:01:59 lr 0.000013 time 0.9211 (1.0053) model_time 0.9209 (1.0047) loss 0.7819 (0.8262) grad_norm 8.7420 (8.5378/1.8391) mem 68106MB [2022-12-20 05:17:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1410/1519] eta 0:01:49 lr 0.000013 time 0.9264 (1.0053) model_time 0.9263 (1.0047) loss 0.7858 (0.8267) grad_norm 8.3490 (8.5250/1.8320) mem 68106MB [2022-12-20 05:18:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1420/1519] eta 0:01:39 lr 0.000013 time 0.9315 (1.0053) model_time 0.9314 (1.0047) loss 0.8482 (0.8267) grad_norm 12.3218 (8.5319/1.8485) mem 68106MB [2022-12-20 05:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1430/1519] eta 0:01:29 lr 0.000013 time 0.9664 (1.0053) model_time 0.9662 (1.0047) loss 0.8205 (0.8267) grad_norm 8.6338 (8.5107/1.8360) mem 68106MB [2022-12-20 05:18:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1440/1519] eta 0:01:19 lr 0.000013 time 0.9259 (1.0053) model_time 0.9258 (1.0047) loss 0.7292 (0.8265) grad_norm 8.5359 (8.5148/1.8413) mem 68106MB [2022-12-20 05:18:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1450/1519] eta 0:01:09 lr 0.000013 time 0.9235 (1.0053) model_time 0.9233 (1.0047) loss 1.0766 (0.8265) grad_norm 8.2466 (8.5026/1.8238) mem 68106MB [2022-12-20 05:18:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1460/1519] eta 0:00:59 lr 0.000013 time 0.9282 (1.0053) model_time 0.9281 (1.0047) loss 0.7300 (0.8265) grad_norm 9.7129 (8.4970/1.8174) mem 68106MB [2022-12-20 05:18:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1470/1519] eta 0:00:49 lr 0.000013 time 0.9225 (1.0053) model_time 0.9223 (1.0047) loss 0.7256 (0.8264) grad_norm 7.2791 (8.4915/1.8214) mem 68106MB [2022-12-20 05:19:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1480/1519] eta 0:00:39 lr 0.000013 time 0.9201 (1.0052) model_time 0.9200 (1.0046) loss 0.9552 (0.8264) grad_norm 7.8799 (8.5063/1.8201) mem 68106MB [2022-12-20 05:19:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1490/1519] eta 0:00:29 lr 0.000013 time 0.9978 (1.0053) model_time 0.9976 (1.0047) loss 1.1894 (0.8268) grad_norm 7.1087 (8.4987/1.8144) mem 68106MB [2022-12-20 05:19:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1500/1519] eta 0:00:19 lr 0.000013 time 0.9215 (1.0052) model_time 0.9214 (1.0046) loss 0.7295 (0.8269) grad_norm 7.7680 (8.5202/1.8583) mem 68106MB [2022-12-20 05:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [56/100][1510/1519] eta 0:00:09 lr 0.000013 time 0.9877 (1.0053) model_time 0.9876 (1.0047) loss 0.6961 (0.8267) grad_norm 6.5591 (8.5335/1.8984) mem 68106MB [2022-12-20 05:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 56 training takes 0:25:26 [2022-12-20 05:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_56.pth saving...... [2022-12-20 05:20:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_56.pth saved !!! [2022-12-20 05:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.679 (0.679) Loss 0.5009 (0.5009) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 05:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.332) Loss 0.5263 (0.4974) Acc@1 92.361 (92.361) Acc@5 97.917 (98.422) Mem 68106MB [2022-12-20 05:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4662 (0.4925) Acc@1 92.361 (92.411) Acc@5 98.958 (98.429) Mem 68106MB [2022-12-20 05:20:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6191 (0.5018) Acc@1 89.931 (92.070) Acc@5 97.917 (98.387) Mem 68106MB [2022-12-20 05:20:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4432 (0.4926) Acc@1 92.708 (92.158) Acc@5 98.958 (98.459) Mem 68106MB [2022-12-20 05:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 0.4782 (0.4901) Acc@1 91.667 (92.252) Acc@5 99.653 (98.509) Mem 68106MB [2022-12-20 05:20:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.5130 (0.4907) Acc@1 90.972 (92.173) Acc@5 97.917 (98.469) Mem 68106MB [2022-12-20 05:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.300 (0.303) Loss 0.5350 (0.4912) Acc@1 92.014 (92.136) Acc@5 98.264 (98.464) Mem 68106MB [2022-12-20 05:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4211 (0.4902) Acc@1 93.056 (92.142) Acc@5 98.958 (98.504) Mem 68106MB [2022-12-20 05:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:56] * Acc@1 92.101 Acc@5 98.510 [2022-12-20 05:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.1% [2022-12-20 05:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.30% [2022-12-20 05:20:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][0/1519] eta 0:47:05 lr 0.000013 time 1.8603 (1.8603) model_time 1.0863 (1.0863) loss 0.8108 (0.8108) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 05:20:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][10/1519] eta 0:27:05 lr 0.000013 time 0.9282 (1.0773) model_time 0.9281 (1.0066) loss 0.7891 (0.7864) grad_norm 9.5166 (7.9428/0.8791) mem 68106MB [2022-12-20 05:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][20/1519] eta 0:26:01 lr 0.000013 time 0.9249 (1.0418) model_time 0.9248 (1.0047) loss 0.6881 (0.8036) grad_norm 9.9326 (8.4489/1.3933) mem 68106MB [2022-12-20 05:21:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][30/1519] eta 0:25:38 lr 0.000013 time 0.9789 (1.0335) model_time 0.9788 (1.0083) loss 0.6939 (0.8157) grad_norm 8.1619 (8.2678/1.1941) mem 68106MB [2022-12-20 05:21:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][40/1519] eta 0:25:16 lr 0.000013 time 0.9300 (1.0251) model_time 0.9299 (1.0059) loss 0.6892 (0.8078) grad_norm 6.9252 (8.0083/1.2079) mem 68106MB [2022-12-20 05:21:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][50/1519] eta 0:24:57 lr 0.000013 time 0.9238 (1.0196) model_time 0.9237 (1.0041) loss 0.7876 (0.8141) grad_norm 8.0809 (8.0941/1.2904) mem 68106MB [2022-12-20 05:21:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][60/1519] eta 0:24:43 lr 0.000013 time 0.9288 (1.0170) model_time 0.9287 (1.0040) loss 0.6771 (0.8174) grad_norm 9.7075 (8.4472/1.6247) mem 68106MB [2022-12-20 05:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][70/1519] eta 0:24:29 lr 0.000013 time 0.9277 (1.0143) model_time 0.9275 (1.0031) loss 0.7134 (0.8309) grad_norm 8.5222 (8.7235/2.1493) mem 68106MB [2022-12-20 05:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][80/1519] eta 0:24:19 lr 0.000013 time 0.9427 (1.0139) model_time 0.9426 (1.0041) loss 0.7395 (0.8343) grad_norm 7.7564 (8.5681/2.0561) mem 68106MB [2022-12-20 05:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][90/1519] eta 0:24:09 lr 0.000013 time 0.9211 (1.0140) model_time 0.9210 (1.0052) loss 0.7154 (0.8326) grad_norm 7.4192 (8.5427/2.1030) mem 68106MB [2022-12-20 05:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][100/1519] eta 0:23:56 lr 0.000013 time 0.9184 (1.0122) model_time 0.9183 (1.0043) loss 0.9576 (0.8337) grad_norm 7.1832 (8.5191/2.0804) mem 68106MB [2022-12-20 05:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][110/1519] eta 0:23:45 lr 0.000013 time 0.9246 (1.0115) model_time 0.9245 (1.0042) loss 0.7717 (0.8360) grad_norm 6.0181 (8.5575/2.2676) mem 68106MB [2022-12-20 05:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][120/1519] eta 0:23:34 lr 0.000013 time 0.9275 (1.0108) model_time 0.9274 (1.0041) loss 0.7598 (0.8372) grad_norm 6.5295 (8.4169/2.2225) mem 68106MB [2022-12-20 05:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][130/1519] eta 0:23:23 lr 0.000013 time 0.9228 (1.0103) model_time 0.9227 (1.0041) loss 0.6879 (0.8330) grad_norm 6.5919 (8.3863/2.1678) mem 68106MB [2022-12-20 05:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][140/1519] eta 0:23:12 lr 0.000013 time 0.9238 (1.0094) model_time 0.9236 (1.0037) loss 0.7044 (0.8268) grad_norm 8.5343 (8.3395/2.1048) mem 68106MB [2022-12-20 05:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][150/1519] eta 0:23:00 lr 0.000013 time 0.9187 (1.0086) model_time 0.9186 (1.0032) loss 0.9731 (0.8304) grad_norm 8.0932 (8.3166/2.0517) mem 68106MB [2022-12-20 05:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][160/1519] eta 0:22:50 lr 0.000013 time 0.9374 (1.0086) model_time 0.9372 (1.0034) loss 0.9747 (0.8277) grad_norm 7.0639 (8.2651/2.0279) mem 68106MB [2022-12-20 05:23:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][170/1519] eta 0:22:40 lr 0.000013 time 0.9242 (1.0082) model_time 0.9241 (1.0034) loss 0.8252 (0.8270) grad_norm 5.7054 (8.2450/2.0010) mem 68106MB [2022-12-20 05:23:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][180/1519] eta 0:22:30 lr 0.000013 time 0.9667 (1.0086) model_time 0.9666 (1.0040) loss 0.6830 (0.8270) grad_norm 7.5694 (8.2349/1.9650) mem 68106MB [2022-12-20 05:23:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][190/1519] eta 0:22:19 lr 0.000013 time 0.9231 (1.0080) model_time 0.9229 (1.0036) loss 0.6718 (0.8323) grad_norm 7.5565 (8.1773/1.9419) mem 68106MB [2022-12-20 05:23:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][200/1519] eta 0:22:09 lr 0.000013 time 0.9243 (1.0080) model_time 0.9242 (1.0039) loss 1.0753 (0.8320) grad_norm 12.1306 (8.2793/2.0078) mem 68106MB [2022-12-20 05:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][210/1519] eta 0:22:01 lr 0.000013 time 1.0401 (1.0092) model_time 1.0399 (1.0053) loss 0.6742 (0.8348) grad_norm 10.4692 (8.3228/2.0419) mem 68106MB [2022-12-20 05:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][220/1519] eta 0:21:50 lr 0.000013 time 0.9226 (1.0087) model_time 0.9225 (1.0049) loss 0.7871 (0.8318) grad_norm 7.5481 (8.2698/2.0208) mem 68106MB [2022-12-20 05:24:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][230/1519] eta 0:21:39 lr 0.000013 time 0.9266 (1.0083) model_time 0.9265 (1.0047) loss 0.8663 (0.8291) grad_norm 8.6314 (8.2634/1.9813) mem 68106MB [2022-12-20 05:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][240/1519] eta 0:21:29 lr 0.000013 time 0.9335 (1.0079) model_time 0.9333 (1.0044) loss 0.9982 (0.8312) grad_norm 7.7762 (8.2518/1.9635) mem 68106MB [2022-12-20 05:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][250/1519] eta 0:21:18 lr 0.000013 time 0.9272 (1.0078) model_time 0.9270 (1.0044) loss 0.8388 (0.8284) grad_norm 7.3079 (8.2466/1.9284) mem 68106MB [2022-12-20 05:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][260/1519] eta 0:21:09 lr 0.000013 time 1.0002 (1.0085) model_time 1.0000 (1.0052) loss 0.6964 (0.8266) grad_norm 7.3577 (8.2210/1.8963) mem 68106MB [2022-12-20 05:25:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][270/1519] eta 0:21:00 lr 0.000013 time 0.9285 (1.0091) model_time 0.9284 (1.0059) loss 0.6619 (0.8252) grad_norm 6.7837 (8.2231/1.8904) mem 68106MB [2022-12-20 05:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][280/1519] eta 0:20:49 lr 0.000013 time 0.9260 (1.0088) model_time 0.9258 (1.0057) loss 0.6820 (0.8261) grad_norm 7.5456 (8.3218/2.0316) mem 68106MB [2022-12-20 05:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][290/1519] eta 0:20:39 lr 0.000013 time 0.9218 (1.0085) model_time 0.9217 (1.0055) loss 0.9386 (0.8276) grad_norm 8.0594 (8.2993/2.0031) mem 68106MB [2022-12-20 05:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][300/1519] eta 0:20:29 lr 0.000013 time 0.9255 (1.0083) model_time 0.9254 (1.0054) loss 0.8827 (0.8255) grad_norm 6.1589 (8.2905/1.9802) mem 68106MB [2022-12-20 05:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][310/1519] eta 0:20:18 lr 0.000013 time 0.9197 (1.0080) model_time 0.9196 (1.0052) loss 0.7461 (0.8297) grad_norm 8.6574 (8.2774/1.9636) mem 68106MB [2022-12-20 05:25:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][320/1519] eta 0:20:08 lr 0.000013 time 0.9210 (1.0077) model_time 0.9208 (1.0050) loss 0.7559 (0.8294) grad_norm 13.4363 (8.3688/2.0199) mem 68106MB [2022-12-20 05:26:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][330/1519] eta 0:19:58 lr 0.000013 time 0.9409 (1.0076) model_time 0.9407 (1.0050) loss 0.7629 (0.8303) grad_norm 9.1077 (8.3460/2.0040) mem 68106MB [2022-12-20 05:26:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][340/1519] eta 0:19:47 lr 0.000013 time 0.9366 (1.0075) model_time 0.9365 (1.0050) loss 0.8951 (0.8297) grad_norm 8.7046 (8.3267/1.9830) mem 68106MB [2022-12-20 05:26:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][350/1519] eta 0:19:37 lr 0.000013 time 0.9297 (1.0074) model_time 0.9296 (1.0049) loss 0.7682 (0.8283) grad_norm 6.6112 (8.3337/1.9803) mem 68106MB [2022-12-20 05:26:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][360/1519] eta 0:19:27 lr 0.000013 time 0.9737 (1.0073) model_time 0.9735 (1.0049) loss 0.9879 (0.8306) grad_norm 7.1372 (8.3424/1.9672) mem 68106MB [2022-12-20 05:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][370/1519] eta 0:19:17 lr 0.000013 time 0.9231 (1.0076) model_time 0.9230 (1.0052) loss 0.9361 (0.8297) grad_norm 8.0171 (8.3567/1.9542) mem 68106MB [2022-12-20 05:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][380/1519] eta 0:19:07 lr 0.000013 time 0.9283 (1.0076) model_time 0.9282 (1.0052) loss 0.8964 (0.8301) grad_norm 9.8973 (8.3322/1.9486) mem 68106MB [2022-12-20 05:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][390/1519] eta 0:18:57 lr 0.000013 time 0.9211 (1.0073) model_time 0.9210 (1.0050) loss 0.7703 (0.8308) grad_norm 10.0825 (8.3429/1.9278) mem 68106MB [2022-12-20 05:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][400/1519] eta 0:18:47 lr 0.000013 time 0.9615 (1.0076) model_time 0.9613 (1.0053) loss 1.0375 (0.8318) grad_norm 9.1817 (8.3482/1.9483) mem 68106MB [2022-12-20 05:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][410/1519] eta 0:18:37 lr 0.000013 time 0.9228 (1.0073) model_time 0.9227 (1.0051) loss 0.7294 (0.8328) grad_norm 7.4732 (8.3670/1.9439) mem 68106MB [2022-12-20 05:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][420/1519] eta 0:18:26 lr 0.000013 time 0.9284 (1.0071) model_time 0.9282 (1.0049) loss 0.7309 (0.8331) grad_norm 6.9137 (8.3636/1.9475) mem 68106MB [2022-12-20 05:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][430/1519] eta 0:18:16 lr 0.000013 time 0.9219 (1.0072) model_time 0.9217 (1.0051) loss 0.8493 (0.8334) grad_norm 6.6699 (8.3564/1.9388) mem 68106MB [2022-12-20 05:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][440/1519] eta 0:18:06 lr 0.000013 time 0.9336 (1.0073) model_time 0.9335 (1.0052) loss 0.7565 (0.8322) grad_norm 9.2770 (8.3536/1.9307) mem 68106MB [2022-12-20 05:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][450/1519] eta 0:17:56 lr 0.000013 time 0.9273 (1.0070) model_time 0.9271 (1.0050) loss 1.3378 (0.8323) grad_norm 7.7940 (8.3357/1.9298) mem 68106MB [2022-12-20 05:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][460/1519] eta 0:17:46 lr 0.000013 time 0.9261 (1.0068) model_time 0.9260 (1.0049) loss 0.7093 (0.8322) grad_norm 11.5134 (8.3445/1.9287) mem 68106MB [2022-12-20 05:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][470/1519] eta 0:17:36 lr 0.000013 time 0.9490 (1.0067) model_time 0.9488 (1.0047) loss 1.0580 (0.8317) grad_norm 5.7946 (8.3283/1.9258) mem 68106MB [2022-12-20 05:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][480/1519] eta 0:17:25 lr 0.000013 time 0.9243 (1.0065) model_time 0.9241 (1.0046) loss 0.9462 (0.8322) grad_norm 9.6164 (8.3460/1.9166) mem 68106MB [2022-12-20 05:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][490/1519] eta 0:17:15 lr 0.000013 time 0.9313 (1.0065) model_time 0.9312 (1.0046) loss 0.7540 (0.8329) grad_norm 6.2562 (8.3532/1.9404) mem 68106MB [2022-12-20 05:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][500/1519] eta 0:17:05 lr 0.000013 time 0.9243 (1.0064) model_time 0.9241 (1.0046) loss 0.6914 (0.8325) grad_norm 8.1841 (8.3440/1.9270) mem 68106MB [2022-12-20 05:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][510/1519] eta 0:16:55 lr 0.000013 time 0.9337 (1.0062) model_time 0.9335 (1.0044) loss 0.8333 (0.8321) grad_norm 7.6078 (8.3115/1.9238) mem 68106MB [2022-12-20 05:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][520/1519] eta 0:16:45 lr 0.000013 time 0.9191 (1.0062) model_time 0.9189 (1.0044) loss 0.8372 (0.8321) grad_norm 6.5912 (8.3236/1.9412) mem 68106MB [2022-12-20 05:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][530/1519] eta 0:16:35 lr 0.000013 time 0.9186 (1.0063) model_time 0.9185 (1.0045) loss 0.8706 (0.8318) grad_norm 7.3146 (8.3245/1.9287) mem 68106MB [2022-12-20 05:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][540/1519] eta 0:16:24 lr 0.000013 time 0.9276 (1.0061) model_time 0.9274 (1.0044) loss 1.0185 (0.8323) grad_norm 5.1994 (8.3320/1.9424) mem 68106MB [2022-12-20 05:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][550/1519] eta 0:16:14 lr 0.000013 time 0.9264 (1.0060) model_time 0.9263 (1.0043) loss 0.8899 (0.8322) grad_norm 7.3895 (8.3294/1.9328) mem 68106MB [2022-12-20 05:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][560/1519] eta 0:16:04 lr 0.000013 time 0.9252 (1.0059) model_time 0.9250 (1.0042) loss 0.8855 (0.8331) grad_norm 7.8935 (8.3507/1.9640) mem 68106MB [2022-12-20 05:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][570/1519] eta 0:15:54 lr 0.000013 time 0.8875 (1.0060) model_time 0.8874 (1.0043) loss 0.7902 (0.8343) grad_norm 8.3685 (8.3575/1.9574) mem 68106MB [2022-12-20 05:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][580/1519] eta 0:15:44 lr 0.000013 time 0.9157 (1.0062) model_time 0.9156 (1.0045) loss 0.6739 (0.8337) grad_norm 7.1388 (8.3808/1.9629) mem 68106MB [2022-12-20 05:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][590/1519] eta 0:15:34 lr 0.000013 time 0.9230 (1.0060) model_time 0.9229 (1.0044) loss 0.7053 (0.8339) grad_norm 6.3831 (8.3910/1.9777) mem 68106MB [2022-12-20 05:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][600/1519] eta 0:15:24 lr 0.000013 time 0.9377 (1.0059) model_time 0.9375 (1.0043) loss 0.6650 (0.8338) grad_norm 5.3528 (8.3715/1.9767) mem 68106MB [2022-12-20 05:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][610/1519] eta 0:15:14 lr 0.000013 time 1.1153 (1.0061) model_time 1.1152 (1.0045) loss 0.8128 (0.8340) grad_norm 6.8988 (8.3610/1.9800) mem 68106MB [2022-12-20 05:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][620/1519] eta 0:15:04 lr 0.000013 time 0.9476 (1.0061) model_time 0.9475 (1.0045) loss 0.7055 (0.8350) grad_norm 8.5648 (8.3484/1.9815) mem 68106MB [2022-12-20 05:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][630/1519] eta 0:14:54 lr 0.000013 time 0.8890 (1.0060) model_time 0.8888 (1.0044) loss 0.7612 (0.8339) grad_norm 11.0128 (8.3551/1.9957) mem 68106MB [2022-12-20 05:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][640/1519] eta 0:14:44 lr 0.000013 time 0.9251 (1.0058) model_time 0.9250 (1.0043) loss 0.7611 (0.8331) grad_norm 12.6478 (8.3681/2.0131) mem 68106MB [2022-12-20 05:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][650/1519] eta 0:14:33 lr 0.000013 time 0.9236 (1.0057) model_time 0.9235 (1.0042) loss 0.8919 (0.8349) grad_norm 8.8323 (8.3600/2.0081) mem 68106MB [2022-12-20 05:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][660/1519] eta 0:14:23 lr 0.000013 time 0.9299 (1.0057) model_time 0.9297 (1.0042) loss 0.9843 (0.8361) grad_norm 8.4858 (8.3391/1.9866) mem 68106MB [2022-12-20 05:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][670/1519] eta 0:14:13 lr 0.000013 time 0.9331 (1.0056) model_time 0.9329 (1.0041) loss 0.6668 (0.8354) grad_norm 8.1165 (8.2847/1.9208) mem 68106MB [2022-12-20 05:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][680/1519] eta 0:14:03 lr 0.000013 time 0.9388 (1.0057) model_time 0.9386 (1.0042) loss 1.0678 (0.8372) grad_norm 6.7758 (8.2764/1.9306) mem 68106MB [2022-12-20 05:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][690/1519] eta 0:13:53 lr 0.000013 time 0.9291 (1.0057) model_time 0.9289 (1.0043) loss 0.6690 (0.8369) grad_norm 7.1051 (8.2671/1.9095) mem 68106MB [2022-12-20 05:32:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][700/1519] eta 0:13:43 lr 0.000013 time 0.9252 (1.0056) model_time 0.9250 (1.0042) loss 0.7620 (0.8365) grad_norm 8.7514 (8.2904/2.0088) mem 68106MB [2022-12-20 05:32:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][710/1519] eta 0:13:33 lr 0.000013 time 0.9904 (1.0058) model_time 0.9903 (1.0044) loss 0.7970 (0.8365) grad_norm 9.5742 (8.2963/1.9826) mem 68106MB [2022-12-20 05:32:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][720/1519] eta 0:13:23 lr 0.000013 time 0.9448 (1.0059) model_time 0.9446 (1.0045) loss 0.8877 (0.8357) grad_norm 6.5060 (8.3001/1.9838) mem 68106MB [2022-12-20 05:32:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][730/1519] eta 0:13:13 lr 0.000013 time 0.9294 (1.0058) model_time 0.9293 (1.0044) loss 0.6840 (0.8348) grad_norm 8.3578 (8.3115/2.0027) mem 68106MB [2022-12-20 05:32:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][740/1519] eta 0:13:03 lr 0.000013 time 0.9477 (1.0057) model_time 0.9476 (1.0044) loss 0.8160 (0.8351) grad_norm 7.1551 (8.2971/2.0086) mem 68106MB [2022-12-20 05:33:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][750/1519] eta 0:12:53 lr 0.000013 time 0.9948 (1.0058) model_time 0.9946 (1.0044) loss 0.7082 (0.8349) grad_norm 12.6306 (8.2963/2.0305) mem 68106MB [2022-12-20 05:33:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][760/1519] eta 0:12:43 lr 0.000013 time 0.9199 (1.0057) model_time 0.9197 (1.0044) loss 0.7467 (0.8348) grad_norm 8.8742 (8.3190/2.0790) mem 68106MB [2022-12-20 05:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][770/1519] eta 0:12:33 lr 0.000013 time 0.9387 (1.0057) model_time 0.9385 (1.0044) loss 0.8734 (0.8347) grad_norm 10.4606 (8.3551/2.1196) mem 68106MB [2022-12-20 05:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][780/1519] eta 0:12:23 lr 0.000013 time 0.8986 (1.0056) model_time 0.8984 (1.0043) loss 1.2052 (0.8350) grad_norm 10.7103 (8.3631/2.1267) mem 68106MB [2022-12-20 05:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][790/1519] eta 0:12:13 lr 0.000013 time 0.9351 (1.0056) model_time 0.9349 (1.0043) loss 0.9826 (0.8352) grad_norm 8.9772 (8.3973/2.1282) mem 68106MB [2022-12-20 05:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][800/1519] eta 0:12:02 lr 0.000013 time 0.9193 (1.0056) model_time 0.9192 (1.0043) loss 0.7237 (0.8349) grad_norm 20.7218 (8.4114/2.2143) mem 68106MB [2022-12-20 05:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][810/1519] eta 0:11:52 lr 0.000013 time 0.9396 (1.0056) model_time 0.9394 (1.0043) loss 0.9940 (0.8345) grad_norm 9.7120 (8.4140/2.1978) mem 68106MB [2022-12-20 05:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][820/1519] eta 0:11:42 lr 0.000013 time 0.9219 (1.0054) model_time 0.9217 (1.0042) loss 0.9263 (0.8339) grad_norm 6.1325 (8.4228/2.1970) mem 68106MB [2022-12-20 05:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][830/1519] eta 0:11:32 lr 0.000013 time 0.9031 (1.0054) model_time 0.9030 (1.0042) loss 0.9399 (0.8342) grad_norm 8.6121 (8.4255/2.1957) mem 68106MB [2022-12-20 05:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][840/1519] eta 0:11:22 lr 0.000013 time 0.9281 (1.0054) model_time 0.9279 (1.0042) loss 1.0995 (0.8348) grad_norm 8.5122 (8.4423/2.2044) mem 68106MB [2022-12-20 05:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][850/1519] eta 0:11:12 lr 0.000013 time 0.9278 (1.0053) model_time 0.9277 (1.0041) loss 1.0294 (0.8346) grad_norm 8.7753 (8.4437/2.2060) mem 68106MB [2022-12-20 05:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][860/1519] eta 0:11:02 lr 0.000013 time 0.9122 (1.0053) model_time 0.9120 (1.0041) loss 0.9479 (0.8350) grad_norm 10.3494 (8.4819/2.2216) mem 68106MB [2022-12-20 05:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][870/1519] eta 0:10:52 lr 0.000013 time 0.9238 (1.0052) model_time 0.9237 (1.0040) loss 0.7941 (0.8350) grad_norm 6.6890 (8.4839/2.2165) mem 68106MB [2022-12-20 05:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][880/1519] eta 0:10:42 lr 0.000013 time 0.9244 (1.0051) model_time 0.9243 (1.0039) loss 0.7097 (0.8340) grad_norm 8.2029 (8.4419/2.1622) mem 68106MB [2022-12-20 05:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][890/1519] eta 0:10:32 lr 0.000013 time 1.1660 (1.0056) model_time 1.1659 (1.0044) loss 0.8349 (0.8339) grad_norm 7.3596 (8.4682/2.1748) mem 68106MB [2022-12-20 05:35:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][900/1519] eta 0:10:22 lr 0.000013 time 0.9314 (1.0055) model_time 0.9313 (1.0044) loss 0.7638 (0.8342) grad_norm 5.7325 (8.4844/2.1892) mem 68106MB [2022-12-20 05:35:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][910/1519] eta 0:10:12 lr 0.000013 time 0.9269 (1.0055) model_time 0.9268 (1.0043) loss 1.0125 (0.8337) grad_norm 10.7046 (8.4946/2.1908) mem 68106MB [2022-12-20 05:35:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][920/1519] eta 0:10:02 lr 0.000013 time 0.9241 (1.0054) model_time 0.9240 (1.0042) loss 0.8139 (0.8335) grad_norm 8.8419 (8.4645/2.1575) mem 68106MB [2022-12-20 05:36:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][930/1519] eta 0:09:52 lr 0.000013 time 0.9203 (1.0055) model_time 0.9202 (1.0043) loss 0.6781 (0.8335) grad_norm 8.1975 (8.4834/2.1630) mem 68106MB [2022-12-20 05:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][940/1519] eta 0:09:42 lr 0.000013 time 0.9202 (1.0054) model_time 0.9200 (1.0042) loss 0.6878 (0.8333) grad_norm 7.2238 (8.4890/2.1614) mem 68106MB [2022-12-20 05:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][950/1519] eta 0:09:32 lr 0.000013 time 0.9310 (1.0053) model_time 0.9309 (1.0042) loss 0.8277 (0.8328) grad_norm 8.6352 (8.5031/2.1801) mem 68106MB [2022-12-20 05:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][960/1519] eta 0:09:21 lr 0.000013 time 0.9204 (1.0053) model_time 0.9202 (1.0042) loss 0.8229 (0.8334) grad_norm 9.2360 (8.5134/2.1818) mem 68106MB [2022-12-20 05:36:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][970/1519] eta 0:09:11 lr 0.000013 time 0.9237 (1.0053) model_time 0.9236 (1.0042) loss 1.0471 (0.8334) grad_norm 7.6563 (8.4900/2.1792) mem 68106MB [2022-12-20 05:36:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][980/1519] eta 0:09:01 lr 0.000013 time 0.9342 (1.0053) model_time 0.9339 (1.0042) loss 0.7258 (0.8332) grad_norm 6.7369 (8.5300/2.2315) mem 68106MB [2022-12-20 05:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][990/1519] eta 0:08:51 lr 0.000013 time 0.9229 (1.0053) model_time 0.9227 (1.0042) loss 0.6732 (0.8338) grad_norm 10.2525 (8.5206/2.2348) mem 68106MB [2022-12-20 05:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1000/1519] eta 0:08:41 lr 0.000013 time 0.9269 (1.0053) model_time 0.9268 (1.0042) loss 0.7834 (0.8337) grad_norm 9.4229 (8.5073/2.2143) mem 68106MB [2022-12-20 05:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1010/1519] eta 0:08:31 lr 0.000013 time 0.9224 (1.0052) model_time 0.9223 (1.0041) loss 0.6761 (0.8335) grad_norm 13.9481 (8.5381/2.2657) mem 68106MB [2022-12-20 05:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1020/1519] eta 0:08:21 lr 0.000013 time 0.9248 (1.0052) model_time 0.9247 (1.0041) loss 0.6740 (0.8328) grad_norm 8.5814 (8.5373/2.2655) mem 68106MB [2022-12-20 05:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1030/1519] eta 0:08:11 lr 0.000013 time 0.9348 (1.0053) model_time 0.9347 (1.0043) loss 0.7140 (0.8326) grad_norm 6.7401 (8.5362/2.2651) mem 68106MB [2022-12-20 05:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1040/1519] eta 0:08:01 lr 0.000013 time 0.9275 (1.0053) model_time 0.9274 (1.0042) loss 1.1972 (0.8332) grad_norm 11.9431 (8.5516/2.2671) mem 68106MB [2022-12-20 05:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1050/1519] eta 0:07:51 lr 0.000013 time 0.9644 (1.0052) model_time 0.9642 (1.0042) loss 0.6772 (0.8327) grad_norm 6.6027 (8.5538/2.2666) mem 68106MB [2022-12-20 05:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1060/1519] eta 0:07:41 lr 0.000013 time 0.9034 (1.0053) model_time 0.9033 (1.0042) loss 0.7918 (0.8327) grad_norm 17.9081 (8.5657/2.3263) mem 68106MB [2022-12-20 05:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1070/1519] eta 0:07:31 lr 0.000013 time 0.9803 (1.0054) model_time 0.9802 (1.0043) loss 0.8392 (0.8327) grad_norm 10.3362 (8.6066/2.3326) mem 68106MB [2022-12-20 05:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1080/1519] eta 0:07:21 lr 0.000013 time 0.9246 (1.0053) model_time 0.9245 (1.0042) loss 0.8829 (0.8330) grad_norm 9.4786 (8.5920/2.3289) mem 68106MB [2022-12-20 05:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1090/1519] eta 0:07:11 lr 0.000013 time 0.9225 (1.0052) model_time 0.9222 (1.0042) loss 0.7068 (0.8326) grad_norm 11.2680 (8.5829/2.3104) mem 68106MB [2022-12-20 05:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1100/1519] eta 0:07:01 lr 0.000013 time 0.9289 (1.0051) model_time 0.9287 (1.0041) loss 0.8262 (0.8333) grad_norm 5.6051 (8.5765/2.3161) mem 68106MB [2022-12-20 05:39:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1110/1519] eta 0:06:51 lr 0.000013 time 0.9254 (1.0053) model_time 0.9252 (1.0043) loss 0.7049 (0.8336) grad_norm 8.0126 (8.6096/2.3074) mem 68106MB [2022-12-20 05:39:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1120/1519] eta 0:06:41 lr 0.000013 time 0.9233 (1.0053) model_time 0.9232 (1.0043) loss 0.7219 (0.8338) grad_norm 8.0561 (8.6100/2.2928) mem 68106MB [2022-12-20 05:39:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1130/1519] eta 0:06:31 lr 0.000013 time 0.9399 (1.0052) model_time 0.9398 (1.0042) loss 1.0447 (0.8337) grad_norm 11.8187 (8.6361/2.3116) mem 68106MB [2022-12-20 05:39:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1140/1519] eta 0:06:20 lr 0.000013 time 0.9244 (1.0052) model_time 0.9242 (1.0042) loss 0.7032 (0.8333) grad_norm 6.5009 (8.6284/2.2958) mem 68106MB [2022-12-20 05:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1150/1519] eta 0:06:10 lr 0.000013 time 0.9209 (1.0054) model_time 0.9207 (1.0044) loss 0.8785 (0.8334) grad_norm 7.3784 (8.6555/2.3191) mem 68106MB [2022-12-20 05:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1160/1519] eta 0:06:00 lr 0.000013 time 0.9213 (1.0054) model_time 0.9212 (1.0044) loss 0.7118 (0.8329) grad_norm 8.7705 (8.6738/2.3091) mem 68106MB [2022-12-20 05:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1170/1519] eta 0:05:50 lr 0.000013 time 0.9269 (1.0054) model_time 0.9268 (1.0044) loss 0.6882 (0.8327) grad_norm 11.0640 (8.6920/2.3209) mem 68106MB [2022-12-20 05:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1180/1519] eta 0:05:40 lr 0.000013 time 0.9230 (1.0053) model_time 0.9228 (1.0044) loss 0.6796 (0.8323) grad_norm 7.5343 (8.6695/2.3131) mem 68106MB [2022-12-20 05:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1190/1519] eta 0:05:30 lr 0.000013 time 0.9310 (1.0053) model_time 0.9308 (1.0044) loss 0.8926 (0.8328) grad_norm 11.6783 (8.6970/2.3079) mem 68106MB [2022-12-20 05:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1200/1519] eta 0:05:20 lr 0.000013 time 0.9193 (1.0053) model_time 0.9192 (1.0044) loss 0.7030 (0.8328) grad_norm 8.1277 (8.7110/2.2956) mem 68106MB [2022-12-20 05:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1210/1519] eta 0:05:10 lr 0.000013 time 0.9309 (1.0054) model_time 0.9308 (1.0045) loss 1.0342 (0.8329) grad_norm 9.9131 (8.7278/2.2895) mem 68106MB [2022-12-20 05:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1220/1519] eta 0:05:00 lr 0.000013 time 0.9250 (1.0054) model_time 0.9249 (1.0045) loss 1.3071 (0.8336) grad_norm 11.0147 (8.7993/2.3982) mem 68106MB [2022-12-20 05:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1230/1519] eta 0:04:50 lr 0.000013 time 0.9215 (1.0053) model_time 0.9214 (1.0044) loss 0.8149 (0.8331) grad_norm 10.1562 (8.7930/2.3937) mem 68106MB [2022-12-20 05:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1240/1519] eta 0:04:40 lr 0.000013 time 0.9223 (1.0054) model_time 0.9220 (1.0044) loss 1.0083 (0.8339) grad_norm 4.8992 (8.8368/2.4265) mem 68106MB [2022-12-20 05:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1250/1519] eta 0:04:30 lr 0.000013 time 0.9287 (1.0054) model_time 0.9285 (1.0044) loss 0.7137 (0.8333) grad_norm 9.0846 (8.8541/2.4302) mem 68106MB [2022-12-20 05:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1260/1519] eta 0:04:20 lr 0.000013 time 0.9229 (1.0053) model_time 0.9228 (1.0044) loss 0.7370 (0.8325) grad_norm 6.8945 (8.8490/2.4641) mem 68106MB [2022-12-20 05:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1270/1519] eta 0:04:10 lr 0.000013 time 0.9290 (1.0053) model_time 0.9288 (1.0043) loss 0.9443 (0.8327) grad_norm 9.7535 (8.8705/2.4643) mem 68106MB [2022-12-20 05:41:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1280/1519] eta 0:04:00 lr 0.000013 time 0.9277 (1.0053) model_time 0.9276 (1.0044) loss 0.7076 (0.8323) grad_norm 7.3607 (8.8848/2.4530) mem 68106MB [2022-12-20 05:42:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1290/1519] eta 0:03:50 lr 0.000013 time 0.9514 (1.0054) model_time 0.9512 (1.0044) loss 0.8456 (0.8324) grad_norm 9.6886 (8.8983/2.4526) mem 68106MB [2022-12-20 05:42:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1300/1519] eta 0:03:40 lr 0.000013 time 0.9269 (1.0054) model_time 0.9267 (1.0045) loss 0.8166 (0.8329) grad_norm 7.8836 (8.8777/2.3717) mem 68106MB [2022-12-20 05:42:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1310/1519] eta 0:03:30 lr 0.000013 time 1.0161 (1.0054) model_time 1.0160 (1.0045) loss 0.9181 (0.8338) grad_norm 8.7353 (8.8926/2.3618) mem 68106MB [2022-12-20 05:42:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1320/1519] eta 0:03:20 lr 0.000013 time 0.9349 (1.0054) model_time 0.9347 (1.0045) loss 1.0537 (0.8340) grad_norm 8.4090 (8.9161/2.3531) mem 68106MB [2022-12-20 05:42:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1330/1519] eta 0:03:10 lr 0.000013 time 0.9379 (1.0055) model_time 0.9378 (1.0046) loss 0.8221 (0.8344) grad_norm 9.2600 (8.9294/2.3488) mem 68106MB [2022-12-20 05:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1340/1519] eta 0:02:59 lr 0.000013 time 0.9255 (1.0055) model_time 0.9254 (1.0046) loss 0.7066 (0.8349) grad_norm 10.4821 (8.9657/2.3367) mem 68106MB [2022-12-20 05:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1350/1519] eta 0:02:49 lr 0.000013 time 0.9188 (1.0054) model_time 0.9186 (1.0045) loss 0.7037 (0.8347) grad_norm 8.7976 (9.0129/2.3612) mem 68106MB [2022-12-20 05:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1360/1519] eta 0:02:39 lr 0.000013 time 0.9211 (1.0054) model_time 0.9209 (1.0045) loss 0.7635 (0.8344) grad_norm 6.9523 (8.9988/2.3181) mem 68106MB [2022-12-20 05:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1370/1519] eta 0:02:29 lr 0.000013 time 0.9302 (1.0054) model_time 0.9301 (1.0045) loss 0.6587 (0.8341) grad_norm 8.9185 (8.9787/2.3121) mem 68106MB [2022-12-20 05:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1380/1519] eta 0:02:19 lr 0.000013 time 0.9206 (1.0055) model_time 0.9205 (1.0046) loss 0.6875 (0.8336) grad_norm 9.1734 (9.0070/2.3286) mem 68106MB [2022-12-20 05:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1390/1519] eta 0:02:09 lr 0.000013 time 0.9221 (1.0055) model_time 0.9220 (1.0046) loss 0.6784 (0.8335) grad_norm 6.8778 (8.9886/2.3296) mem 68106MB [2022-12-20 05:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1400/1519] eta 0:01:59 lr 0.000013 time 0.9484 (1.0055) model_time 0.9483 (1.0046) loss 0.6852 (0.8337) grad_norm 7.9139 (8.9437/2.2303) mem 68106MB [2022-12-20 05:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1410/1519] eta 0:01:49 lr 0.000013 time 0.9375 (1.0055) model_time 0.9374 (1.0046) loss 0.9832 (0.8342) grad_norm 8.5746 (8.9378/2.2525) mem 68106MB [2022-12-20 05:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1420/1519] eta 0:01:39 lr 0.000013 time 0.9286 (1.0054) model_time 0.9285 (1.0046) loss 1.0688 (0.8343) grad_norm 9.2245 (8.9595/2.2745) mem 68106MB [2022-12-20 05:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1430/1519] eta 0:01:29 lr 0.000013 time 0.9235 (1.0054) model_time 0.9233 (1.0045) loss 0.6605 (0.8345) grad_norm 7.9323 (8.9437/2.2919) mem 68106MB [2022-12-20 05:44:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1440/1519] eta 0:01:19 lr 0.000013 time 0.9286 (1.0053) model_time 0.9284 (1.0045) loss 0.8176 (0.8343) grad_norm 7.1315 (8.9380/2.2943) mem 68106MB [2022-12-20 05:44:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1450/1519] eta 0:01:09 lr 0.000013 time 0.9279 (1.0053) model_time 0.9277 (1.0045) loss 0.6965 (0.8344) grad_norm 10.0902 (8.9314/2.2998) mem 68106MB [2022-12-20 05:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1460/1519] eta 0:00:59 lr 0.000013 time 0.9268 (1.0053) model_time 0.9266 (1.0044) loss 0.9936 (0.8345) grad_norm 7.6991 (8.9056/2.3022) mem 68106MB [2022-12-20 05:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1470/1519] eta 0:00:49 lr 0.000013 time 0.9227 (1.0053) model_time 0.9225 (1.0045) loss 0.6898 (0.8344) grad_norm 8.1949 (8.9125/2.2984) mem 68106MB [2022-12-20 05:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1480/1519] eta 0:00:39 lr 0.000013 time 0.9298 (1.0053) model_time 0.9296 (1.0044) loss 0.8654 (0.8349) grad_norm 10.9874 (8.9163/2.2892) mem 68106MB [2022-12-20 05:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1490/1519] eta 0:00:29 lr 0.000013 time 1.0105 (1.0055) model_time 1.0104 (1.0046) loss 0.7563 (0.8348) grad_norm 8.0961 (8.9027/2.2911) mem 68106MB [2022-12-20 05:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1500/1519] eta 0:00:19 lr 0.000013 time 0.9198 (1.0054) model_time 0.9196 (1.0046) loss 0.8922 (0.8350) grad_norm 9.0120 (8.8964/2.2779) mem 68106MB [2022-12-20 05:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [57/100][1510/1519] eta 0:00:09 lr 0.000013 time 0.9222 (1.0053) model_time 0.9221 (1.0045) loss 0.8498 (0.8347) grad_norm 8.6258 (8.9083/2.2697) mem 68106MB [2022-12-20 05:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 57 training takes 0:25:27 [2022-12-20 05:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_57.pth saving...... [2022-12-20 05:46:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_57.pth saved !!! [2022-12-20 05:46:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.671 (0.671) Loss 0.5146 (0.5146) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 05:46:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.331) Loss 0.5167 (0.4995) Acc@1 92.708 (92.551) Acc@5 97.917 (98.611) Mem 68106MB [2022-12-20 05:46:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4540 (0.4936) Acc@1 92.361 (92.526) Acc@5 99.306 (98.462) Mem 68106MB [2022-12-20 05:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6151 (0.5016) Acc@1 89.236 (92.182) Acc@5 97.569 (98.398) Mem 68106MB [2022-12-20 05:46:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.308) Loss 0.4511 (0.4922) Acc@1 93.056 (92.327) Acc@5 99.306 (98.493) Mem 68106MB [2022-12-20 05:46:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.306) Loss 0.4753 (0.4891) Acc@1 90.972 (92.334) Acc@5 99.653 (98.557) Mem 68106MB [2022-12-20 05:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.5026 (0.4889) Acc@1 91.319 (92.247) Acc@5 97.569 (98.509) Mem 68106MB [2022-12-20 05:46:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5210 (0.4894) Acc@1 92.014 (92.249) Acc@5 98.264 (98.518) Mem 68106MB [2022-12-20 05:46:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.303) Loss 0.4270 (0.4878) Acc@1 93.056 (92.245) Acc@5 97.917 (98.555) Mem 68106MB [2022-12-20 05:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:57] * Acc@1 92.236 Acc@5 98.563 [2022-12-20 05:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.2% [2022-12-20 05:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.30% [2022-12-20 05:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][0/1519] eta 0:46:39 lr 0.000013 time 1.8432 (1.8432) model_time 1.0362 (1.0362) loss 0.7018 (0.7018) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 05:47:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][10/1519] eta 0:27:01 lr 0.000013 time 0.9239 (1.0744) model_time 0.9238 (1.0007) loss 0.7000 (0.8888) grad_norm 14.9786 (10.6627/2.5940) mem 68106MB [2022-12-20 05:47:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][20/1519] eta 0:26:00 lr 0.000013 time 0.8855 (1.0413) model_time 0.8854 (1.0025) loss 0.8690 (0.8578) grad_norm 8.3783 (9.7180/2.1647) mem 68106MB [2022-12-20 05:47:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][30/1519] eta 0:25:32 lr 0.000013 time 0.9280 (1.0290) model_time 0.9278 (1.0027) loss 0.9198 (0.8722) grad_norm 11.5183 (9.4382/2.0637) mem 68106MB [2022-12-20 05:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][40/1519] eta 0:25:15 lr 0.000013 time 0.9220 (1.0245) model_time 0.9219 (1.0045) loss 0.7588 (0.8658) grad_norm 9.2522 (9.1685/1.9021) mem 68106MB [2022-12-20 05:47:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][50/1519] eta 0:24:59 lr 0.000013 time 0.9271 (1.0207) model_time 0.9270 (1.0045) loss 0.9642 (0.8622) grad_norm 6.0806 (8.9382/1.8944) mem 68106MB [2022-12-20 05:47:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][60/1519] eta 0:24:43 lr 0.000013 time 0.9208 (1.0170) model_time 0.9207 (1.0035) loss 0.8623 (0.8609) grad_norm 7.1329 (8.8689/1.8290) mem 68106MB [2022-12-20 05:48:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][70/1519] eta 0:24:29 lr 0.000013 time 0.9213 (1.0145) model_time 0.9212 (1.0028) loss 1.4654 (0.8533) grad_norm 8.7575 (8.7833/1.7287) mem 68106MB [2022-12-20 05:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][80/1519] eta 0:24:17 lr 0.000013 time 0.9229 (1.0131) model_time 0.9227 (1.0028) loss 0.7101 (0.8580) grad_norm 8.9532 (8.7151/1.7274) mem 68106MB [2022-12-20 05:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][90/1519] eta 0:24:06 lr 0.000013 time 0.9796 (1.0122) model_time 0.9795 (1.0030) loss 0.7165 (0.8520) grad_norm 8.2813 (8.6662/1.6546) mem 68106MB [2022-12-20 05:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][100/1519] eta 0:23:55 lr 0.000013 time 0.9485 (1.0114) model_time 0.9484 (1.0031) loss 0.6673 (0.8543) grad_norm 8.5927 (8.6294/1.5848) mem 68106MB [2022-12-20 05:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][110/1519] eta 0:23:43 lr 0.000013 time 0.9199 (1.0103) model_time 0.9197 (1.0027) loss 0.6998 (0.8506) grad_norm 7.9619 (8.5278/1.5897) mem 68106MB [2022-12-20 05:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][120/1519] eta 0:23:32 lr 0.000013 time 0.9275 (1.0099) model_time 0.9274 (1.0030) loss 0.7823 (0.8442) grad_norm 11.3460 (8.5367/1.5927) mem 68106MB [2022-12-20 05:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][130/1519] eta 0:23:22 lr 0.000013 time 0.9296 (1.0097) model_time 0.9295 (1.0033) loss 0.6846 (0.8395) grad_norm 7.6434 (8.5339/1.6271) mem 68106MB [2022-12-20 05:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][140/1519] eta 0:23:11 lr 0.000013 time 0.9230 (1.0088) model_time 0.9228 (1.0028) loss 0.9431 (0.8418) grad_norm 9.6237 (8.6615/1.9576) mem 68106MB [2022-12-20 05:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][150/1519] eta 0:23:00 lr 0.000013 time 0.9241 (1.0084) model_time 0.9240 (1.0028) loss 0.7182 (0.8412) grad_norm 9.0352 (8.6483/1.9484) mem 68106MB [2022-12-20 05:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][160/1519] eta 0:22:49 lr 0.000013 time 0.9263 (1.0077) model_time 0.9262 (1.0024) loss 0.7313 (0.8391) grad_norm 9.2205 (8.6352/1.9273) mem 68106MB [2022-12-20 05:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][170/1519] eta 0:22:39 lr 0.000013 time 1.0263 (1.0081) model_time 1.0262 (1.0030) loss 1.1471 (0.8404) grad_norm 20.7050 (8.7565/2.2837) mem 68106MB [2022-12-20 05:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][180/1519] eta 0:22:29 lr 0.000013 time 0.9215 (1.0080) model_time 0.9214 (1.0032) loss 0.7622 (0.8440) grad_norm 8.9219 (8.7661/2.2294) mem 68106MB [2022-12-20 05:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][190/1519] eta 0:22:18 lr 0.000013 time 0.9195 (1.0075) model_time 0.9194 (1.0030) loss 1.0045 (0.8414) grad_norm 9.2391 (8.8666/2.2511) mem 68106MB [2022-12-20 05:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][200/1519] eta 0:22:08 lr 0.000013 time 0.9746 (1.0074) model_time 0.9745 (1.0031) loss 0.7315 (0.8425) grad_norm 6.0436 (8.8676/2.2722) mem 68106MB [2022-12-20 05:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][210/1519] eta 0:21:58 lr 0.000013 time 0.9235 (1.0072) model_time 0.9234 (1.0031) loss 0.9544 (0.8432) grad_norm 6.2892 (8.8384/2.2802) mem 68106MB [2022-12-20 05:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][220/1519] eta 0:21:48 lr 0.000013 time 0.9252 (1.0071) model_time 0.9250 (1.0031) loss 0.6868 (0.8422) grad_norm 8.1772 (8.8230/2.2344) mem 68106MB [2022-12-20 05:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][230/1519] eta 0:21:37 lr 0.000013 time 0.9248 (1.0068) model_time 0.9247 (1.0030) loss 0.7838 (0.8387) grad_norm 6.6796 (8.7534/2.2114) mem 68106MB [2022-12-20 05:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][240/1519] eta 0:21:27 lr 0.000013 time 0.9335 (1.0063) model_time 0.9333 (1.0027) loss 0.6658 (0.8350) grad_norm 10.7472 (8.7762/2.1950) mem 68106MB [2022-12-20 05:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][250/1519] eta 0:21:16 lr 0.000013 time 0.9266 (1.0061) model_time 0.9265 (1.0026) loss 0.8982 (0.8371) grad_norm 7.9740 (8.7729/2.1549) mem 68106MB [2022-12-20 05:51:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][260/1519] eta 0:21:06 lr 0.000013 time 0.9278 (1.0061) model_time 0.9277 (1.0027) loss 0.8351 (0.8347) grad_norm 10.5717 (8.7870/2.1262) mem 68106MB [2022-12-20 05:51:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][270/1519] eta 0:20:56 lr 0.000013 time 0.9786 (1.0061) model_time 0.9785 (1.0028) loss 0.7045 (0.8337) grad_norm 8.4855 (8.7775/2.0947) mem 68106MB [2022-12-20 05:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][280/1519] eta 0:20:46 lr 0.000013 time 0.9286 (1.0061) model_time 0.9285 (1.0030) loss 0.6944 (0.8333) grad_norm 10.0159 (8.7629/2.0819) mem 68106MB [2022-12-20 05:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][290/1519] eta 0:20:36 lr 0.000013 time 0.9263 (1.0063) model_time 0.9262 (1.0032) loss 0.6993 (0.8326) grad_norm 8.1802 (8.7370/2.0686) mem 68106MB [2022-12-20 05:51:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][300/1519] eta 0:20:26 lr 0.000013 time 0.9141 (1.0063) model_time 0.9139 (1.0033) loss 0.9414 (0.8325) grad_norm 7.5728 (8.6989/2.0499) mem 68106MB [2022-12-20 05:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][310/1519] eta 0:20:16 lr 0.000013 time 0.9693 (1.0063) model_time 0.9692 (1.0033) loss 0.9790 (0.8330) grad_norm 6.5727 (8.6641/2.0378) mem 68106MB [2022-12-20 05:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][320/1519] eta 0:20:06 lr 0.000013 time 0.9252 (1.0061) model_time 0.9251 (1.0032) loss 1.1673 (0.8324) grad_norm 10.2178 (8.6882/2.0568) mem 68106MB [2022-12-20 05:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][330/1519] eta 0:19:55 lr 0.000013 time 0.9301 (1.0059) model_time 0.9299 (1.0030) loss 0.7739 (0.8326) grad_norm 5.9500 (8.6982/2.0614) mem 68106MB [2022-12-20 05:52:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][340/1519] eta 0:19:45 lr 0.000013 time 0.9332 (1.0059) model_time 0.9331 (1.0031) loss 0.9915 (0.8322) grad_norm 8.2454 (8.6968/2.0477) mem 68106MB [2022-12-20 05:52:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][350/1519] eta 0:19:36 lr 0.000013 time 1.0261 (1.0061) model_time 1.0259 (1.0034) loss 0.6811 (0.8309) grad_norm 6.3652 (8.7064/2.0635) mem 68106MB [2022-12-20 05:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][360/1519] eta 0:19:26 lr 0.000013 time 0.9070 (1.0063) model_time 0.9068 (1.0037) loss 1.3580 (0.8329) grad_norm 7.8853 (8.7515/2.1017) mem 68106MB [2022-12-20 05:53:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][370/1519] eta 0:19:16 lr 0.000013 time 0.9223 (1.0062) model_time 0.9221 (1.0036) loss 0.9445 (0.8329) grad_norm 6.3111 (8.7573/2.0861) mem 68106MB [2022-12-20 05:53:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][380/1519] eta 0:19:05 lr 0.000013 time 0.9296 (1.0061) model_time 0.9295 (1.0035) loss 0.7858 (0.8336) grad_norm 7.4746 (8.7698/2.0701) mem 68106MB [2022-12-20 05:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][390/1519] eta 0:18:55 lr 0.000013 time 0.9392 (1.0059) model_time 0.9390 (1.0035) loss 0.7106 (0.8336) grad_norm 9.0980 (8.7650/2.0696) mem 68106MB [2022-12-20 05:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][400/1519] eta 0:18:45 lr 0.000013 time 0.9270 (1.0057) model_time 0.9268 (1.0033) loss 0.6907 (0.8327) grad_norm 9.0706 (8.7690/2.0461) mem 68106MB [2022-12-20 05:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][410/1519] eta 0:18:35 lr 0.000013 time 0.9241 (1.0056) model_time 0.9240 (1.0032) loss 0.9599 (0.8335) grad_norm 8.2674 (8.7568/2.0266) mem 68106MB [2022-12-20 05:53:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][420/1519] eta 0:18:25 lr 0.000013 time 0.9268 (1.0056) model_time 0.9267 (1.0033) loss 0.8266 (0.8337) grad_norm 7.7731 (8.7404/2.0130) mem 68106MB [2022-12-20 05:54:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][430/1519] eta 0:18:14 lr 0.000013 time 0.9333 (1.0054) model_time 0.9332 (1.0032) loss 0.7916 (0.8348) grad_norm 9.1450 (8.7397/2.0047) mem 68106MB [2022-12-20 05:54:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][440/1519] eta 0:18:05 lr 0.000013 time 0.9222 (1.0056) model_time 0.9220 (1.0034) loss 0.7038 (0.8337) grad_norm 8.2006 (8.7263/1.9847) mem 68106MB [2022-12-20 05:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][450/1519] eta 0:17:54 lr 0.000013 time 0.9249 (1.0054) model_time 0.9248 (1.0033) loss 0.9768 (0.8345) grad_norm 7.8837 (8.7065/1.9686) mem 68106MB [2022-12-20 05:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][460/1519] eta 0:17:44 lr 0.000013 time 0.9173 (1.0055) model_time 0.9171 (1.0034) loss 0.6780 (0.8344) grad_norm 7.1736 (8.6955/1.9695) mem 68106MB [2022-12-20 05:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][470/1519] eta 0:17:34 lr 0.000013 time 0.9240 (1.0055) model_time 0.9239 (1.0034) loss 0.7524 (0.8346) grad_norm 8.3584 (8.7120/1.9830) mem 68106MB [2022-12-20 05:54:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][480/1519] eta 0:17:24 lr 0.000013 time 0.9281 (1.0057) model_time 0.9279 (1.0036) loss 0.7410 (0.8343) grad_norm 12.0135 (8.7136/1.9851) mem 68106MB [2022-12-20 05:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][490/1519] eta 0:17:15 lr 0.000013 time 0.9781 (1.0059) model_time 0.9780 (1.0038) loss 0.6909 (0.8336) grad_norm 8.5602 (8.7235/1.9740) mem 68106MB [2022-12-20 05:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][500/1519] eta 0:17:04 lr 0.000013 time 0.9229 (1.0059) model_time 0.9227 (1.0039) loss 0.9674 (0.8334) grad_norm 9.0385 (8.7120/1.9640) mem 68106MB [2022-12-20 05:55:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][510/1519] eta 0:16:54 lr 0.000013 time 0.9232 (1.0057) model_time 0.9231 (1.0038) loss 0.7142 (0.8334) grad_norm 8.3993 (8.6893/1.9565) mem 68106MB [2022-12-20 05:55:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][520/1519] eta 0:16:44 lr 0.000013 time 0.9046 (1.0058) model_time 0.9044 (1.0038) loss 0.7525 (0.8335) grad_norm 8.5202 (8.6789/1.9415) mem 68106MB [2022-12-20 05:55:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][530/1519] eta 0:16:34 lr 0.000013 time 0.9096 (1.0058) model_time 0.9095 (1.0039) loss 0.7029 (0.8338) grad_norm 9.5088 (8.6846/1.9283) mem 68106MB [2022-12-20 05:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][540/1519] eta 0:16:24 lr 0.000013 time 0.9226 (1.0057) model_time 0.9225 (1.0038) loss 0.9334 (0.8332) grad_norm 7.4728 (8.6612/1.9227) mem 68106MB [2022-12-20 05:56:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][550/1519] eta 0:16:14 lr 0.000013 time 0.9254 (1.0055) model_time 0.9252 (1.0037) loss 0.7712 (0.8323) grad_norm 7.5486 (8.6653/1.9187) mem 68106MB [2022-12-20 05:56:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][560/1519] eta 0:16:04 lr 0.000013 time 0.9226 (1.0054) model_time 0.9225 (1.0036) loss 0.7863 (0.8325) grad_norm 6.2802 (8.6433/1.9116) mem 68106MB [2022-12-20 05:56:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][570/1519] eta 0:15:54 lr 0.000013 time 0.9260 (1.0053) model_time 0.9259 (1.0036) loss 0.7998 (0.8324) grad_norm 7.2201 (8.6178/1.9082) mem 68106MB [2022-12-20 05:56:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][580/1519] eta 0:15:43 lr 0.000013 time 0.9270 (1.0052) model_time 0.9269 (1.0035) loss 0.8324 (0.8317) grad_norm 10.8366 (8.6500/1.9283) mem 68106MB [2022-12-20 05:56:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][590/1519] eta 0:15:33 lr 0.000013 time 0.9201 (1.0052) model_time 0.9200 (1.0035) loss 0.9786 (0.8322) grad_norm 14.4119 (8.6963/1.9879) mem 68106MB [2022-12-20 05:56:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][600/1519] eta 0:15:23 lr 0.000013 time 0.9257 (1.0051) model_time 0.9256 (1.0034) loss 0.7020 (0.8315) grad_norm 11.2707 (8.7117/1.9827) mem 68106MB [2022-12-20 05:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][610/1519] eta 0:15:13 lr 0.000013 time 0.9224 (1.0050) model_time 0.9222 (1.0033) loss 0.9488 (0.8323) grad_norm 7.9533 (8.6654/1.9429) mem 68106MB [2022-12-20 05:57:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][620/1519] eta 0:15:03 lr 0.000013 time 0.9384 (1.0054) model_time 0.9383 (1.0038) loss 0.9966 (0.8322) grad_norm 7.8984 (8.6703/1.9461) mem 68106MB [2022-12-20 05:57:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][630/1519] eta 0:14:53 lr 0.000013 time 0.9274 (1.0053) model_time 0.9273 (1.0037) loss 0.7174 (0.8313) grad_norm 8.1934 (8.6616/1.9405) mem 68106MB [2022-12-20 05:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][640/1519] eta 0:14:43 lr 0.000013 time 0.9218 (1.0052) model_time 0.9217 (1.0036) loss 0.7656 (0.8311) grad_norm 12.1909 (8.6648/1.9536) mem 68106MB [2022-12-20 05:57:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][650/1519] eta 0:14:33 lr 0.000013 time 0.9256 (1.0051) model_time 0.9255 (1.0035) loss 1.0806 (0.8315) grad_norm 7.4004 (8.6741/1.9455) mem 68106MB [2022-12-20 05:57:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][660/1519] eta 0:14:23 lr 0.000013 time 0.9294 (1.0051) model_time 0.9292 (1.0035) loss 0.6792 (0.8325) grad_norm 6.7568 (8.6682/1.9435) mem 68106MB [2022-12-20 05:58:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][670/1519] eta 0:14:13 lr 0.000013 time 0.9317 (1.0052) model_time 0.9316 (1.0037) loss 0.6978 (0.8321) grad_norm 11.7189 (8.6707/1.9532) mem 68106MB [2022-12-20 05:58:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][680/1519] eta 0:14:03 lr 0.000013 time 0.9207 (1.0052) model_time 0.9206 (1.0037) loss 0.6822 (0.8314) grad_norm 8.0561 (8.6650/1.9519) mem 68106MB [2022-12-20 05:58:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][690/1519] eta 0:13:53 lr 0.000013 time 0.9351 (1.0052) model_time 0.9349 (1.0036) loss 0.8535 (0.8317) grad_norm 11.8358 (8.6778/1.9682) mem 68106MB [2022-12-20 05:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][700/1519] eta 0:13:43 lr 0.000013 time 0.9446 (1.0052) model_time 0.9445 (1.0037) loss 0.8690 (0.8323) grad_norm 7.0673 (8.6832/1.9764) mem 68106MB [2022-12-20 05:58:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][710/1519] eta 0:13:33 lr 0.000013 time 0.9351 (1.0051) model_time 0.9349 (1.0036) loss 0.6791 (0.8317) grad_norm 6.8949 (8.6930/1.9784) mem 68106MB [2022-12-20 05:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][720/1519] eta 0:13:22 lr 0.000013 time 0.9249 (1.0050) model_time 0.9248 (1.0035) loss 0.8754 (0.8325) grad_norm 9.9628 (8.6887/1.9717) mem 68106MB [2022-12-20 05:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][730/1519] eta 0:13:12 lr 0.000013 time 0.9274 (1.0050) model_time 0.9273 (1.0035) loss 0.6708 (0.8317) grad_norm 6.2817 (8.6794/1.9618) mem 68106MB [2022-12-20 05:59:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][740/1519] eta 0:13:02 lr 0.000013 time 0.9252 (1.0049) model_time 0.9250 (1.0034) loss 0.9779 (0.8319) grad_norm 8.6120 (8.6547/1.8801) mem 68106MB [2022-12-20 05:59:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][750/1519] eta 0:12:52 lr 0.000013 time 0.9302 (1.0051) model_time 0.9300 (1.0037) loss 0.7311 (0.8318) grad_norm 9.8237 (8.6815/1.8771) mem 68106MB [2022-12-20 05:59:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][760/1519] eta 0:12:42 lr 0.000013 time 0.9355 (1.0051) model_time 0.9349 (1.0037) loss 0.9358 (0.8314) grad_norm 11.9344 (8.6849/1.8810) mem 68106MB [2022-12-20 05:59:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][770/1519] eta 0:12:32 lr 0.000013 time 0.9524 (1.0051) model_time 0.9523 (1.0037) loss 0.8925 (0.8313) grad_norm 8.0972 (8.6602/1.7664) mem 68106MB [2022-12-20 05:59:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][780/1519] eta 0:12:22 lr 0.000013 time 0.9230 (1.0052) model_time 0.9227 (1.0038) loss 0.7623 (0.8310) grad_norm 6.3721 (8.6296/1.7746) mem 68106MB [2022-12-20 06:00:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][790/1519] eta 0:12:12 lr 0.000013 time 0.9735 (1.0052) model_time 0.9734 (1.0038) loss 0.6822 (0.8304) grad_norm 7.6166 (8.5858/1.7609) mem 68106MB [2022-12-20 06:00:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][800/1519] eta 0:12:02 lr 0.000013 time 0.9337 (1.0052) model_time 0.9335 (1.0038) loss 0.7118 (0.8307) grad_norm 7.6331 (8.5750/1.7327) mem 68106MB [2022-12-20 06:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][810/1519] eta 0:11:52 lr 0.000013 time 0.9312 (1.0053) model_time 0.9310 (1.0040) loss 0.7230 (0.8312) grad_norm 8.4444 (8.5754/1.7154) mem 68106MB [2022-12-20 06:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][820/1519] eta 0:11:42 lr 0.000013 time 0.9306 (1.0052) model_time 0.9305 (1.0039) loss 0.6868 (0.8316) grad_norm 5.4621 (8.5451/1.7346) mem 68106MB [2022-12-20 06:00:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][830/1519] eta 0:11:32 lr 0.000013 time 0.9293 (1.0052) model_time 0.9291 (1.0039) loss 0.7147 (0.8308) grad_norm 7.4757 (8.5493/1.7319) mem 68106MB [2022-12-20 06:00:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][840/1519] eta 0:11:22 lr 0.000013 time 0.9335 (1.0053) model_time 0.9334 (1.0040) loss 0.6712 (0.8307) grad_norm 8.2217 (8.5167/1.7264) mem 68106MB [2022-12-20 06:01:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][850/1519] eta 0:11:12 lr 0.000013 time 0.9262 (1.0054) model_time 0.9261 (1.0041) loss 0.6978 (0.8308) grad_norm 8.3309 (8.4996/1.7305) mem 68106MB [2022-12-20 06:01:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][860/1519] eta 0:11:02 lr 0.000013 time 0.9370 (1.0053) model_time 0.9362 (1.0040) loss 0.6756 (0.8300) grad_norm 8.5625 (8.4753/1.7300) mem 68106MB [2022-12-20 06:01:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][870/1519] eta 0:10:52 lr 0.000013 time 0.9372 (1.0053) model_time 0.9371 (1.0040) loss 1.0081 (0.8294) grad_norm 7.9960 (8.4813/1.7396) mem 68106MB [2022-12-20 06:01:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][880/1519] eta 0:10:42 lr 0.000013 time 0.9277 (1.0052) model_time 0.9276 (1.0039) loss 0.9376 (0.8293) grad_norm 7.1014 (8.4778/1.7330) mem 68106MB [2022-12-20 06:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][890/1519] eta 0:10:32 lr 0.000013 time 0.9369 (1.0051) model_time 0.9367 (1.0039) loss 0.7323 (0.8286) grad_norm 11.2402 (8.4928/1.7344) mem 68106MB [2022-12-20 06:01:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][900/1519] eta 0:10:22 lr 0.000013 time 0.9291 (1.0051) model_time 0.9289 (1.0038) loss 0.8956 (0.8284) grad_norm 5.9964 (8.5049/1.7377) mem 68106MB [2022-12-20 06:02:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][910/1519] eta 0:10:12 lr 0.000013 time 0.9223 (1.0050) model_time 0.9221 (1.0038) loss 0.8172 (0.8286) grad_norm 6.5907 (8.5221/1.7533) mem 68106MB [2022-12-20 06:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][920/1519] eta 0:10:01 lr 0.000013 time 0.9360 (1.0050) model_time 0.9357 (1.0037) loss 0.7751 (0.8278) grad_norm 5.2995 (8.5045/1.7389) mem 68106MB [2022-12-20 06:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][930/1519] eta 0:09:52 lr 0.000013 time 0.9319 (1.0051) model_time 0.9316 (1.0039) loss 1.0718 (0.8278) grad_norm 10.0802 (8.5229/1.8102) mem 68106MB [2022-12-20 06:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][940/1519] eta 0:09:41 lr 0.000013 time 0.9290 (1.0050) model_time 0.9289 (1.0038) loss 0.6781 (0.8277) grad_norm 11.0436 (8.5109/1.8194) mem 68106MB [2022-12-20 06:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][950/1519] eta 0:09:31 lr 0.000013 time 0.9293 (1.0050) model_time 0.9290 (1.0038) loss 0.6823 (0.8276) grad_norm 7.1973 (8.5037/1.7926) mem 68106MB [2022-12-20 06:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][960/1519] eta 0:09:21 lr 0.000013 time 0.9292 (1.0049) model_time 0.9289 (1.0037) loss 0.8760 (0.8289) grad_norm 7.7064 (8.4701/1.7502) mem 68106MB [2022-12-20 06:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][970/1519] eta 0:09:11 lr 0.000013 time 0.9953 (1.0049) model_time 0.9952 (1.0037) loss 0.7414 (0.8287) grad_norm 6.5074 (8.4577/1.7450) mem 68106MB [2022-12-20 06:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][980/1519] eta 0:09:01 lr 0.000013 time 0.9268 (1.0049) model_time 0.9266 (1.0038) loss 1.0076 (0.8285) grad_norm 13.6169 (8.4594/1.7669) mem 68106MB [2022-12-20 06:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][990/1519] eta 0:08:51 lr 0.000013 time 0.9279 (1.0049) model_time 0.9277 (1.0038) loss 0.6794 (0.8288) grad_norm 10.8674 (8.4854/1.7833) mem 68106MB [2022-12-20 06:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1000/1519] eta 0:08:41 lr 0.000013 time 0.9304 (1.0048) model_time 0.9303 (1.0037) loss 0.7670 (0.8286) grad_norm 6.8792 (8.4558/1.7904) mem 68106MB [2022-12-20 06:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1010/1519] eta 0:08:31 lr 0.000013 time 0.9319 (1.0049) model_time 0.9316 (1.0037) loss 0.8951 (0.8287) grad_norm 7.4987 (8.4608/1.8011) mem 68106MB [2022-12-20 06:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1020/1519] eta 0:08:21 lr 0.000013 time 0.9276 (1.0048) model_time 0.9274 (1.0036) loss 0.8840 (0.8295) grad_norm 7.7438 (8.4683/1.8030) mem 68106MB [2022-12-20 06:04:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1030/1519] eta 0:08:11 lr 0.000013 time 0.9278 (1.0047) model_time 0.9276 (1.0036) loss 0.7077 (0.8295) grad_norm 11.9859 (8.4737/1.8036) mem 68106MB [2022-12-20 06:04:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1040/1519] eta 0:08:01 lr 0.000013 time 0.9372 (1.0048) model_time 0.9371 (1.0036) loss 1.0044 (0.8292) grad_norm 8.8506 (8.4714/1.8050) mem 68106MB [2022-12-20 06:04:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1050/1519] eta 0:07:51 lr 0.000013 time 0.9443 (1.0047) model_time 0.9441 (1.0036) loss 0.6748 (0.8298) grad_norm 8.1601 (8.4891/1.8121) mem 68106MB [2022-12-20 06:04:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1060/1519] eta 0:07:41 lr 0.000013 time 0.9202 (1.0049) model_time 0.9199 (1.0038) loss 0.6881 (0.8295) grad_norm 9.7636 (8.5032/1.8194) mem 68106MB [2022-12-20 06:04:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1070/1519] eta 0:07:31 lr 0.000013 time 0.9352 (1.0049) model_time 0.9351 (1.0038) loss 0.6858 (0.8290) grad_norm 8.0215 (8.4770/1.7914) mem 68106MB [2022-12-20 06:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1080/1519] eta 0:07:21 lr 0.000013 time 0.9252 (1.0049) model_time 0.9251 (1.0037) loss 0.9964 (0.8288) grad_norm 10.4392 (8.4840/1.7772) mem 68106MB [2022-12-20 06:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1090/1519] eta 0:07:11 lr 0.000013 time 1.0133 (1.0049) model_time 1.0132 (1.0038) loss 0.7463 (0.8296) grad_norm 8.3185 (8.4737/1.7976) mem 68106MB [2022-12-20 06:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1100/1519] eta 0:07:01 lr 0.000013 time 0.9260 (1.0049) model_time 0.9258 (1.0038) loss 0.9442 (0.8288) grad_norm 7.8111 (8.4766/1.7995) mem 68106MB [2022-12-20 06:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1110/1519] eta 0:06:50 lr 0.000013 time 0.9302 (1.0048) model_time 0.9299 (1.0037) loss 0.6737 (0.8284) grad_norm 9.7937 (8.5094/1.8032) mem 68106MB [2022-12-20 06:05:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1120/1519] eta 0:06:40 lr 0.000013 time 0.9271 (1.0048) model_time 0.9269 (1.0037) loss 0.8583 (0.8283) grad_norm 8.1902 (8.5071/1.8066) mem 68106MB [2022-12-20 06:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1130/1519] eta 0:06:30 lr 0.000013 time 0.9280 (1.0047) model_time 0.9279 (1.0036) loss 0.8249 (0.8283) grad_norm 7.2148 (8.4936/1.8035) mem 68106MB [2022-12-20 06:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1140/1519] eta 0:06:20 lr 0.000013 time 0.9433 (1.0047) model_time 0.9430 (1.0037) loss 0.8197 (0.8285) grad_norm 7.5036 (8.5223/1.8116) mem 68106MB [2022-12-20 06:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1150/1519] eta 0:06:10 lr 0.000013 time 0.9301 (1.0047) model_time 0.9299 (1.0037) loss 0.6524 (0.8287) grad_norm 8.9841 (8.4962/1.8117) mem 68106MB [2022-12-20 06:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1160/1519] eta 0:06:00 lr 0.000013 time 0.9441 (1.0050) model_time 0.9439 (1.0039) loss 1.0874 (0.8290) grad_norm 8.3106 (8.5260/1.8167) mem 68106MB [2022-12-20 06:06:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1170/1519] eta 0:05:50 lr 0.000012 time 0.9293 (1.0050) model_time 0.9291 (1.0039) loss 0.7808 (0.8291) grad_norm 8.1839 (8.5327/1.8095) mem 68106MB [2022-12-20 06:06:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1180/1519] eta 0:05:40 lr 0.000012 time 0.9306 (1.0049) model_time 0.9304 (1.0039) loss 0.9555 (0.8291) grad_norm 9.5634 (8.5052/1.7735) mem 68106MB [2022-12-20 06:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1190/1519] eta 0:05:30 lr 0.000012 time 0.9212 (1.0049) model_time 0.9210 (1.0038) loss 0.9150 (0.8293) grad_norm 9.5135 (8.4749/1.7021) mem 68106MB [2022-12-20 06:06:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1200/1519] eta 0:05:20 lr 0.000012 time 0.9285 (1.0048) model_time 0.9284 (1.0038) loss 0.9258 (0.8292) grad_norm 7.4194 (8.4579/1.7078) mem 68106MB [2022-12-20 06:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1210/1519] eta 0:05:10 lr 0.000012 time 0.9369 (1.0048) model_time 0.9367 (1.0038) loss 0.7169 (0.8292) grad_norm 8.4391 (8.4870/1.7389) mem 68106MB [2022-12-20 06:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1220/1519] eta 0:05:00 lr 0.000012 time 0.9330 (1.0048) model_time 0.9328 (1.0038) loss 0.6585 (0.8297) grad_norm 10.4655 (8.4743/1.7422) mem 68106MB [2022-12-20 06:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1230/1519] eta 0:04:50 lr 0.000012 time 0.9324 (1.0048) model_time 0.9322 (1.0038) loss 0.6710 (0.8288) grad_norm 8.6945 (8.4962/1.7870) mem 68106MB [2022-12-20 06:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1240/1519] eta 0:04:40 lr 0.000012 time 0.9368 (1.0049) model_time 0.9367 (1.0039) loss 0.9410 (0.8293) grad_norm 6.0923 (8.4917/1.7788) mem 68106MB [2022-12-20 06:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1250/1519] eta 0:04:30 lr 0.000012 time 0.9385 (1.0049) model_time 0.9383 (1.0039) loss 0.8635 (0.8293) grad_norm 7.0105 (8.4753/1.7798) mem 68106MB [2022-12-20 06:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1260/1519] eta 0:04:20 lr 0.000012 time 0.9320 (1.0049) model_time 0.9317 (1.0039) loss 0.7431 (0.8287) grad_norm 7.3141 (8.4725/1.7850) mem 68106MB [2022-12-20 06:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1270/1519] eta 0:04:10 lr 0.000012 time 0.9303 (1.0050) model_time 0.9301 (1.0039) loss 0.6719 (0.8284) grad_norm 7.0805 (8.4600/1.7798) mem 68106MB [2022-12-20 06:08:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1280/1519] eta 0:04:00 lr 0.000012 time 0.9304 (1.0049) model_time 0.9302 (1.0039) loss 0.9241 (0.8290) grad_norm 10.8797 (8.4786/1.8261) mem 68106MB [2022-12-20 06:08:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1290/1519] eta 0:03:50 lr 0.000012 time 0.9168 (1.0050) model_time 0.9167 (1.0040) loss 0.7254 (0.8288) grad_norm 7.5187 (8.4569/1.8126) mem 68106MB [2022-12-20 06:08:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1300/1519] eta 0:03:40 lr 0.000012 time 0.9216 (1.0050) model_time 0.9214 (1.0040) loss 0.7318 (0.8286) grad_norm 8.8058 (8.4440/1.8061) mem 68106MB [2022-12-20 06:08:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1310/1519] eta 0:03:30 lr 0.000012 time 0.9306 (1.0049) model_time 0.9304 (1.0039) loss 0.9444 (0.8288) grad_norm 7.2227 (8.4389/1.7950) mem 68106MB [2022-12-20 06:08:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1320/1519] eta 0:03:19 lr 0.000012 time 0.9260 (1.0049) model_time 0.9258 (1.0039) loss 0.8316 (0.8287) grad_norm 7.2183 (8.4247/1.8025) mem 68106MB [2022-12-20 06:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1330/1519] eta 0:03:09 lr 0.000012 time 0.9038 (1.0049) model_time 0.9035 (1.0040) loss 0.9627 (0.8283) grad_norm 7.6514 (8.4307/1.8050) mem 68106MB [2022-12-20 06:09:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1340/1519] eta 0:02:59 lr 0.000012 time 0.9277 (1.0049) model_time 0.9275 (1.0039) loss 0.7092 (0.8286) grad_norm 10.1135 (8.4816/1.8912) mem 68106MB [2022-12-20 06:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1350/1519] eta 0:02:49 lr 0.000012 time 0.9306 (1.0049) model_time 0.9304 (1.0039) loss 0.8010 (0.8284) grad_norm 15.2353 (8.4883/1.9351) mem 68106MB [2022-12-20 06:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1360/1519] eta 0:02:39 lr 0.000012 time 0.9430 (1.0048) model_time 0.9427 (1.0039) loss 0.6777 (0.8280) grad_norm 8.3405 (8.4973/1.9407) mem 68106MB [2022-12-20 06:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1370/1519] eta 0:02:29 lr 0.000012 time 0.9308 (1.0048) model_time 0.9306 (1.0038) loss 0.8556 (0.8280) grad_norm 8.3087 (8.4855/1.9235) mem 68106MB [2022-12-20 06:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1380/1519] eta 0:02:19 lr 0.000012 time 0.9319 (1.0048) model_time 0.9316 (1.0038) loss 0.9155 (0.8284) grad_norm 8.9766 (8.5161/1.9173) mem 68106MB [2022-12-20 06:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1390/1519] eta 0:02:09 lr 0.000012 time 0.9678 (1.0050) model_time 0.9675 (1.0040) loss 1.1230 (0.8291) grad_norm 11.6439 (8.5329/1.9077) mem 68106MB [2022-12-20 06:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1400/1519] eta 0:01:59 lr 0.000012 time 0.9314 (1.0049) model_time 0.9311 (1.0040) loss 0.7515 (0.8290) grad_norm 7.2088 (8.5838/2.1401) mem 68106MB [2022-12-20 06:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1410/1519] eta 0:01:49 lr 0.000012 time 0.9282 (1.0050) model_time 0.9280 (1.0040) loss 0.9015 (0.8295) grad_norm 8.4746 (8.5989/2.1725) mem 68106MB [2022-12-20 06:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1420/1519] eta 0:01:39 lr 0.000012 time 0.9347 (1.0050) model_time 0.9345 (1.0040) loss 0.8132 (0.8295) grad_norm 5.5431 (8.6109/2.1659) mem 68106MB [2022-12-20 06:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1430/1519] eta 0:01:29 lr 0.000012 time 0.9312 (1.0050) model_time 0.9310 (1.0041) loss 0.7148 (0.8289) grad_norm 8.5067 (8.6107/2.1685) mem 68106MB [2022-12-20 06:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1440/1519] eta 0:01:19 lr 0.000012 time 0.9288 (1.0050) model_time 0.9286 (1.0040) loss 0.7416 (0.8292) grad_norm 9.2360 (8.6519/2.1704) mem 68106MB [2022-12-20 06:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1450/1519] eta 0:01:09 lr 0.000012 time 0.8808 (1.0051) model_time 0.8806 (1.0042) loss 0.9944 (0.8292) grad_norm 7.2748 (8.6433/2.1742) mem 68106MB [2022-12-20 06:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1460/1519] eta 0:00:59 lr 0.000012 time 0.9301 (1.0051) model_time 0.9300 (1.0042) loss 0.7272 (0.8288) grad_norm 7.9666 (8.6292/2.1816) mem 68106MB [2022-12-20 06:11:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1470/1519] eta 0:00:49 lr 0.000012 time 0.9146 (1.0052) model_time 0.9145 (1.0042) loss 0.8722 (0.8289) grad_norm 6.5203 (8.6175/2.1764) mem 68106MB [2022-12-20 06:11:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1480/1519] eta 0:00:39 lr 0.000012 time 0.9310 (1.0052) model_time 0.9308 (1.0042) loss 0.9434 (0.8289) grad_norm 8.5521 (8.6053/2.1810) mem 68106MB [2022-12-20 06:11:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1490/1519] eta 0:00:29 lr 0.000012 time 0.9402 (1.0051) model_time 0.9400 (1.0042) loss 0.7928 (0.8291) grad_norm 8.7396 (8.5996/2.1750) mem 68106MB [2022-12-20 06:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1500/1519] eta 0:00:19 lr 0.000012 time 0.9334 (1.0051) model_time 0.9333 (1.0042) loss 0.7115 (0.8288) grad_norm 7.0114 (8.6076/2.1831) mem 68106MB [2022-12-20 06:12:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [58/100][1510/1519] eta 0:00:09 lr 0.000012 time 0.9337 (1.0051) model_time 0.9336 (1.0042) loss 0.7221 (0.8287) grad_norm 9.8676 (8.6358/2.1800) mem 68106MB [2022-12-20 06:12:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 58 training takes 0:25:26 [2022-12-20 06:12:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_58.pth saving...... [2022-12-20 06:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_58.pth saved !!! [2022-12-20 06:12:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.647 (0.647) Loss 0.5133 (0.5133) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 06:12:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.329) Loss 0.5097 (0.5020) Acc@1 92.708 (92.487) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 06:12:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4694 (0.4981) Acc@1 91.319 (92.460) Acc@5 98.958 (98.313) Mem 68106MB [2022-12-20 06:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 0.6112 (0.5034) Acc@1 89.236 (92.204) Acc@5 98.264 (98.297) Mem 68106MB [2022-12-20 06:12:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.306) Loss 0.4485 (0.4925) Acc@1 93.056 (92.310) Acc@5 99.306 (98.433) Mem 68106MB [2022-12-20 06:12:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.304) Loss 0.4884 (0.4903) Acc@1 90.625 (92.259) Acc@5 99.306 (98.489) Mem 68106MB [2022-12-20 06:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.303) Loss 0.4956 (0.4906) Acc@1 91.319 (92.242) Acc@5 97.917 (98.463) Mem 68106MB [2022-12-20 06:13:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5315 (0.4910) Acc@1 93.056 (92.224) Acc@5 98.611 (98.469) Mem 68106MB [2022-12-20 06:13:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.302) Loss 0.4346 (0.4896) Acc@1 92.708 (92.215) Acc@5 98.264 (98.504) Mem 68106MB [2022-12-20 06:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:58] * Acc@1 92.174 Acc@5 98.502 [2022-12-20 06:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.2% [2022-12-20 06:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.30% [2022-12-20 06:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][0/1519] eta 0:46:25 lr 0.000012 time 1.8334 (1.8334) model_time 1.0703 (1.0703) loss 0.7582 (0.7582) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 06:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][10/1519] eta 0:27:15 lr 0.000012 time 0.9827 (1.0841) model_time 0.9822 (1.0142) loss 0.8008 (0.7879) grad_norm 6.6345 (8.3742/1.0550) mem 68106MB [2022-12-20 06:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][20/1519] eta 0:26:18 lr 0.000012 time 0.9669 (1.0530) model_time 0.9668 (1.0162) loss 0.7339 (0.7833) grad_norm 8.5012 (8.2439/1.4334) mem 68106MB [2022-12-20 06:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][30/1519] eta 0:25:46 lr 0.000012 time 1.0029 (1.0386) model_time 1.0028 (1.0135) loss 0.6761 (0.8146) grad_norm 7.2783 (8.6180/1.6413) mem 68106MB [2022-12-20 06:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][40/1519] eta 0:25:24 lr 0.000012 time 0.9885 (1.0305) model_time 0.9884 (1.0115) loss 0.8411 (0.8002) grad_norm 8.6241 (8.3022/1.5632) mem 68106MB [2022-12-20 06:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][50/1519] eta 0:25:05 lr 0.000012 time 0.9310 (1.0250) model_time 0.9309 (1.0097) loss 0.7670 (0.8106) grad_norm 8.5181 (8.5386/1.5789) mem 68106MB [2022-12-20 06:14:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][60/1519] eta 0:24:50 lr 0.000012 time 0.9460 (1.0216) model_time 0.9458 (1.0087) loss 0.7100 (0.8273) grad_norm 9.9609 (8.5028/1.5322) mem 68106MB [2022-12-20 06:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][70/1519] eta 0:24:35 lr 0.000012 time 0.9258 (1.0186) model_time 0.9256 (1.0074) loss 0.8187 (0.8166) grad_norm 7.6186 (8.3910/1.5209) mem 68106MB [2022-12-20 06:14:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][80/1519] eta 0:24:23 lr 0.000012 time 0.9387 (1.0169) model_time 0.9385 (1.0071) loss 0.8284 (0.8142) grad_norm 5.9712 (8.2151/1.5328) mem 68106MB [2022-12-20 06:14:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][90/1519] eta 0:24:11 lr 0.000012 time 0.9947 (1.0157) model_time 0.9945 (1.0070) loss 1.2253 (0.8150) grad_norm 9.0118 (8.3392/1.5282) mem 68106MB [2022-12-20 06:14:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][100/1519] eta 0:23:59 lr 0.000012 time 0.9302 (1.0146) model_time 0.9301 (1.0066) loss 0.6842 (0.8152) grad_norm 10.9627 (8.4483/1.5974) mem 68106MB [2022-12-20 06:14:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][110/1519] eta 0:23:48 lr 0.000012 time 0.9295 (1.0138) model_time 0.9293 (1.0065) loss 0.7599 (0.8119) grad_norm 7.9022 (8.4546/1.5562) mem 68106MB [2022-12-20 06:15:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][120/1519] eta 0:23:37 lr 0.000012 time 0.9334 (1.0132) model_time 0.9333 (1.0065) loss 0.8191 (0.8128) grad_norm 8.8047 (8.4397/1.5204) mem 68106MB [2022-12-20 06:15:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][130/1519] eta 0:23:26 lr 0.000012 time 0.9256 (1.0124) model_time 0.9252 (1.0062) loss 0.6797 (0.8134) grad_norm 7.0686 (8.3898/1.4865) mem 68106MB [2022-12-20 06:15:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][140/1519] eta 0:23:14 lr 0.000012 time 0.9340 (1.0116) model_time 0.9337 (1.0058) loss 0.7056 (0.8122) grad_norm 6.8925 (8.3626/1.5390) mem 68106MB [2022-12-20 06:15:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][150/1519] eta 0:23:04 lr 0.000012 time 0.9648 (1.0113) model_time 0.9646 (1.0059) loss 0.7683 (0.8123) grad_norm 6.7094 (8.3554/1.5485) mem 68106MB [2022-12-20 06:15:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][160/1519] eta 0:22:54 lr 0.000012 time 0.9450 (1.0116) model_time 0.9448 (1.0065) loss 0.7091 (0.8133) grad_norm 5.6555 (8.3489/1.5816) mem 68106MB [2022-12-20 06:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][170/1519] eta 0:22:45 lr 0.000012 time 0.9494 (1.0122) model_time 0.9492 (1.0074) loss 1.0802 (0.8156) grad_norm 15.6293 (8.4120/1.7286) mem 68106MB [2022-12-20 06:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][180/1519] eta 0:22:34 lr 0.000012 time 0.9363 (1.0116) model_time 0.9361 (1.0070) loss 0.6731 (0.8143) grad_norm 8.1206 (8.3727/1.7009) mem 68106MB [2022-12-20 06:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][190/1519] eta 0:22:24 lr 0.000012 time 1.0224 (1.0118) model_time 1.0221 (1.0075) loss 0.8493 (0.8143) grad_norm 8.3387 (8.3455/1.6853) mem 68106MB [2022-12-20 06:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][200/1519] eta 0:22:13 lr 0.000012 time 0.9338 (1.0111) model_time 0.9335 (1.0069) loss 0.9797 (0.8173) grad_norm 8.3758 (8.4922/1.9450) mem 68106MB [2022-12-20 06:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][210/1519] eta 0:22:03 lr 0.000012 time 1.0076 (1.0111) model_time 1.0075 (1.0071) loss 0.8291 (0.8193) grad_norm 7.6511 (8.4686/1.9071) mem 68106MB [2022-12-20 06:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][220/1519] eta 0:21:52 lr 0.000012 time 0.9404 (1.0106) model_time 0.9403 (1.0067) loss 0.8412 (0.8197) grad_norm 8.2218 (8.4423/1.8778) mem 68106MB [2022-12-20 06:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][230/1519] eta 0:21:42 lr 0.000012 time 0.9392 (1.0102) model_time 0.9389 (1.0065) loss 0.7601 (0.8175) grad_norm 8.6019 (8.4249/1.8439) mem 68106MB [2022-12-20 06:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][240/1519] eta 0:21:31 lr 0.000012 time 0.9326 (1.0099) model_time 0.9324 (1.0064) loss 0.6804 (0.8191) grad_norm 7.2549 (8.4687/1.9042) mem 68106MB [2022-12-20 06:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][250/1519] eta 0:21:20 lr 0.000012 time 0.9319 (1.0094) model_time 0.9317 (1.0060) loss 1.1511 (0.8208) grad_norm 6.8124 (8.4391/1.8826) mem 68106MB [2022-12-20 06:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][260/1519] eta 0:21:10 lr 0.000012 time 0.9354 (1.0090) model_time 0.9352 (1.0057) loss 0.9025 (0.8208) grad_norm 8.4852 (8.4175/1.8586) mem 68106MB [2022-12-20 06:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][270/1519] eta 0:21:00 lr 0.000012 time 0.9928 (1.0093) model_time 0.9924 (1.0061) loss 0.8911 (0.8220) grad_norm 6.6016 (8.3678/1.8480) mem 68106MB [2022-12-20 06:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][280/1519] eta 0:20:50 lr 0.000012 time 0.9294 (1.0090) model_time 0.9291 (1.0059) loss 0.6675 (0.8201) grad_norm 6.9170 (8.3650/1.8388) mem 68106MB [2022-12-20 06:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][290/1519] eta 0:20:39 lr 0.000012 time 0.9305 (1.0087) model_time 0.9302 (1.0056) loss 1.0048 (0.8219) grad_norm 7.9459 (8.3925/1.8303) mem 68106MB [2022-12-20 06:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][300/1519] eta 0:20:29 lr 0.000012 time 0.9321 (1.0084) model_time 0.9319 (1.0055) loss 0.6893 (0.8203) grad_norm 8.4254 (8.3719/1.8123) mem 68106MB [2022-12-20 06:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][310/1519] eta 0:20:19 lr 0.000012 time 0.9276 (1.0084) model_time 0.9274 (1.0055) loss 0.6758 (0.8200) grad_norm 7.4987 (8.3969/1.8002) mem 68106MB [2022-12-20 06:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][320/1519] eta 0:20:08 lr 0.000012 time 0.9298 (1.0081) model_time 0.9296 (1.0053) loss 0.9732 (0.8212) grad_norm 10.3955 (8.3917/1.7928) mem 68106MB [2022-12-20 06:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][330/1519] eta 0:19:59 lr 0.000012 time 0.8943 (1.0086) model_time 0.8941 (1.0059) loss 0.8751 (0.8214) grad_norm 10.3814 (8.3765/1.7904) mem 68106MB [2022-12-20 06:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][340/1519] eta 0:19:48 lr 0.000012 time 0.9261 (1.0083) model_time 0.9260 (1.0057) loss 0.8264 (0.8223) grad_norm 5.3527 (8.3628/1.7832) mem 68106MB [2022-12-20 06:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][350/1519] eta 0:19:38 lr 0.000012 time 0.9653 (1.0082) model_time 0.9652 (1.0056) loss 0.6655 (0.8233) grad_norm 8.3788 (8.3371/1.7715) mem 68106MB [2022-12-20 06:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][360/1519] eta 0:19:28 lr 0.000012 time 0.9206 (1.0084) model_time 0.9205 (1.0059) loss 0.7150 (0.8236) grad_norm 7.3189 (8.3434/1.7793) mem 68106MB [2022-12-20 06:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][370/1519] eta 0:19:19 lr 0.000012 time 0.9398 (1.0089) model_time 0.9397 (1.0065) loss 0.8179 (0.8232) grad_norm 12.2699 (8.3515/1.7830) mem 68106MB [2022-12-20 06:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][380/1519] eta 0:19:08 lr 0.000012 time 0.9484 (1.0087) model_time 0.9483 (1.0063) loss 0.8613 (0.8222) grad_norm 6.1193 (8.3232/1.7803) mem 68106MB [2022-12-20 06:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][390/1519] eta 0:18:58 lr 0.000012 time 0.9846 (1.0086) model_time 0.9845 (1.0063) loss 0.8178 (0.8228) grad_norm 9.2222 (8.3602/1.8169) mem 68106MB [2022-12-20 06:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][400/1519] eta 0:18:48 lr 0.000012 time 0.9723 (1.0084) model_time 0.9721 (1.0061) loss 0.8461 (0.8217) grad_norm 6.4025 (8.3716/1.8219) mem 68106MB [2022-12-20 06:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][410/1519] eta 0:18:38 lr 0.000012 time 0.9386 (1.0082) model_time 0.9385 (1.0060) loss 0.7986 (0.8215) grad_norm 5.9906 (8.3623/1.8138) mem 68106MB [2022-12-20 06:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][420/1519] eta 0:18:27 lr 0.000012 time 0.9274 (1.0082) model_time 0.9273 (1.0059) loss 0.9103 (0.8222) grad_norm 7.8792 (8.3478/1.8025) mem 68106MB [2022-12-20 06:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][430/1519] eta 0:18:17 lr 0.000012 time 0.9293 (1.0080) model_time 0.9291 (1.0058) loss 0.7247 (0.8234) grad_norm 10.4077 (8.3458/1.7942) mem 68106MB [2022-12-20 06:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][440/1519] eta 0:18:07 lr 0.000012 time 0.9287 (1.0078) model_time 0.9285 (1.0057) loss 0.6998 (0.8219) grad_norm 7.8375 (8.3266/1.7874) mem 68106MB [2022-12-20 06:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][450/1519] eta 0:17:57 lr 0.000012 time 0.9450 (1.0079) model_time 0.9449 (1.0058) loss 0.9577 (0.8235) grad_norm 4.3076 (8.3177/1.7901) mem 68106MB [2022-12-20 06:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][460/1519] eta 0:17:47 lr 0.000012 time 0.9408 (1.0078) model_time 0.9406 (1.0057) loss 0.7205 (0.8222) grad_norm 25.4537 (8.4056/2.1131) mem 68106MB [2022-12-20 06:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][470/1519] eta 0:17:37 lr 0.000012 time 0.9218 (1.0082) model_time 0.9217 (1.0062) loss 0.7291 (0.8223) grad_norm 6.9208 (8.3873/2.1012) mem 68106MB [2022-12-20 06:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][480/1519] eta 0:17:27 lr 0.000012 time 0.9725 (1.0082) model_time 0.9724 (1.0062) loss 0.6775 (0.8203) grad_norm 8.1078 (8.4035/2.1039) mem 68106MB [2022-12-20 06:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][490/1519] eta 0:17:17 lr 0.000012 time 0.9131 (1.0081) model_time 0.9129 (1.0061) loss 0.6840 (0.8193) grad_norm 10.8606 (8.3966/2.0958) mem 68106MB [2022-12-20 06:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][500/1519] eta 0:17:07 lr 0.000012 time 0.9359 (1.0079) model_time 0.9358 (1.0060) loss 0.7317 (0.8186) grad_norm 9.2885 (8.3951/2.0806) mem 68106MB [2022-12-20 06:21:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][510/1519] eta 0:16:56 lr 0.000012 time 0.9293 (1.0079) model_time 0.9292 (1.0060) loss 0.6995 (0.8192) grad_norm 6.9355 (8.3767/2.0674) mem 68106MB [2022-12-20 06:21:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][520/1519] eta 0:16:46 lr 0.000012 time 0.9326 (1.0078) model_time 0.9325 (1.0059) loss 0.7102 (0.8199) grad_norm 9.9421 (8.3797/2.0601) mem 68106MB [2022-12-20 06:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][530/1519] eta 0:16:36 lr 0.000012 time 0.9368 (1.0078) model_time 0.9366 (1.0059) loss 0.7138 (0.8200) grad_norm 8.4838 (8.3861/2.0459) mem 68106MB [2022-12-20 06:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][540/1519] eta 0:16:26 lr 0.000012 time 0.9295 (1.0077) model_time 0.9293 (1.0059) loss 0.8542 (0.8207) grad_norm 12.7535 (8.4273/2.0635) mem 68106MB [2022-12-20 06:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][550/1519] eta 0:16:16 lr 0.000012 time 0.9135 (1.0075) model_time 0.9133 (1.0057) loss 0.6786 (0.8220) grad_norm 6.8770 (8.4163/2.0537) mem 68106MB [2022-12-20 06:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][560/1519] eta 0:16:06 lr 0.000012 time 0.9293 (1.0075) model_time 0.9292 (1.0057) loss 0.6838 (0.8223) grad_norm 8.3596 (8.4074/2.0411) mem 68106MB [2022-12-20 06:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][570/1519] eta 0:15:55 lr 0.000012 time 0.9401 (1.0074) model_time 0.9399 (1.0056) loss 0.7776 (0.8233) grad_norm 7.1434 (8.3803/2.0357) mem 68106MB [2022-12-20 06:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][580/1519] eta 0:15:45 lr 0.000012 time 0.9820 (1.0073) model_time 0.9818 (1.0056) loss 0.8896 (0.8234) grad_norm 7.1637 (8.3772/2.0214) mem 68106MB [2022-12-20 06:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][590/1519] eta 0:15:35 lr 0.000012 time 0.9315 (1.0072) model_time 0.9312 (1.0055) loss 0.7942 (0.8220) grad_norm 9.8201 (8.3849/2.0128) mem 68106MB [2022-12-20 06:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][600/1519] eta 0:15:25 lr 0.000012 time 0.9319 (1.0072) model_time 0.9318 (1.0056) loss 0.9049 (0.8223) grad_norm 5.8453 (8.3701/2.0079) mem 68106MB [2022-12-20 06:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][610/1519] eta 0:15:15 lr 0.000012 time 0.9475 (1.0071) model_time 0.9474 (1.0055) loss 0.9199 (0.8215) grad_norm 6.2363 (8.3593/2.0121) mem 68106MB [2022-12-20 06:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][620/1519] eta 0:15:05 lr 0.000012 time 0.9323 (1.0071) model_time 0.9321 (1.0055) loss 1.2223 (0.8215) grad_norm 8.7145 (8.3651/2.0046) mem 68106MB [2022-12-20 06:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][630/1519] eta 0:14:55 lr 0.000012 time 0.9346 (1.0071) model_time 0.9345 (1.0055) loss 1.1008 (0.8225) grad_norm 9.4479 (8.3383/1.9924) mem 68106MB [2022-12-20 06:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][640/1519] eta 0:14:45 lr 0.000012 time 1.0429 (1.0072) model_time 1.0428 (1.0056) loss 0.7596 (0.8227) grad_norm 9.4162 (8.3575/1.9893) mem 68106MB [2022-12-20 06:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][650/1519] eta 0:14:35 lr 0.000012 time 0.9307 (1.0072) model_time 0.9305 (1.0056) loss 0.8709 (0.8228) grad_norm 7.3352 (8.3355/1.9886) mem 68106MB [2022-12-20 06:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][660/1519] eta 0:14:25 lr 0.000012 time 0.9311 (1.0071) model_time 0.9309 (1.0055) loss 0.7018 (0.8223) grad_norm 9.2977 (8.3278/1.9883) mem 68106MB [2022-12-20 06:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][670/1519] eta 0:14:14 lr 0.000012 time 0.9352 (1.0070) model_time 0.9351 (1.0055) loss 1.1484 (0.8222) grad_norm 10.1709 (8.3558/2.0159) mem 68106MB [2022-12-20 06:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][680/1519] eta 0:14:04 lr 0.000012 time 0.9283 (1.0070) model_time 0.9282 (1.0055) loss 1.0555 (0.8230) grad_norm 9.8083 (8.4039/2.0344) mem 68106MB [2022-12-20 06:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][690/1519] eta 0:13:54 lr 0.000012 time 0.9335 (1.0069) model_time 0.9334 (1.0054) loss 0.7476 (0.8236) grad_norm 8.4628 (8.3634/2.0384) mem 68106MB [2022-12-20 06:24:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][700/1519] eta 0:13:44 lr 0.000012 time 0.9341 (1.0067) model_time 0.9340 (1.0053) loss 0.6771 (0.8228) grad_norm 7.9580 (8.3520/2.0250) mem 68106MB [2022-12-20 06:25:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][710/1519] eta 0:13:34 lr 0.000012 time 0.9353 (1.0067) model_time 0.9352 (1.0053) loss 0.8260 (0.8233) grad_norm 12.9539 (8.3823/2.0515) mem 68106MB [2022-12-20 06:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][720/1519] eta 0:13:24 lr 0.000012 time 0.9290 (1.0067) model_time 0.9289 (1.0052) loss 0.7190 (0.8233) grad_norm 10.7927 (8.3942/2.0625) mem 68106MB [2022-12-20 06:25:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][730/1519] eta 0:13:14 lr 0.000012 time 0.9429 (1.0066) model_time 0.9428 (1.0052) loss 0.8010 (0.8232) grad_norm 7.9038 (8.4071/2.0709) mem 68106MB [2022-12-20 06:25:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][740/1519] eta 0:13:04 lr 0.000012 time 0.9381 (1.0065) model_time 0.9379 (1.0051) loss 1.0103 (0.8237) grad_norm 9.7158 (8.3998/2.0621) mem 68106MB [2022-12-20 06:25:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][750/1519] eta 0:12:53 lr 0.000012 time 0.9335 (1.0065) model_time 0.9333 (1.0051) loss 0.7553 (0.8234) grad_norm 10.4319 (8.4163/2.0646) mem 68106MB [2022-12-20 06:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][760/1519] eta 0:12:43 lr 0.000012 time 0.9346 (1.0065) model_time 0.9344 (1.0052) loss 0.8185 (0.8230) grad_norm 9.3856 (8.4272/2.0660) mem 68106MB [2022-12-20 06:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][770/1519] eta 0:12:33 lr 0.000012 time 0.9300 (1.0065) model_time 0.9298 (1.0051) loss 0.6755 (0.8230) grad_norm 6.6450 (8.4037/2.0274) mem 68106MB [2022-12-20 06:26:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][780/1519] eta 0:12:23 lr 0.000012 time 0.9321 (1.0066) model_time 0.9320 (1.0052) loss 0.6700 (0.8235) grad_norm 9.1421 (8.4096/2.0254) mem 68106MB [2022-12-20 06:26:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][790/1519] eta 0:12:13 lr 0.000012 time 0.9877 (1.0066) model_time 0.9875 (1.0053) loss 0.7015 (0.8233) grad_norm 8.9100 (8.4339/2.0520) mem 68106MB [2022-12-20 06:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][800/1519] eta 0:12:03 lr 0.000012 time 0.9328 (1.0066) model_time 0.9326 (1.0053) loss 0.8726 (0.8242) grad_norm 9.7542 (8.3942/1.9715) mem 68106MB [2022-12-20 06:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][810/1519] eta 0:11:53 lr 0.000012 time 0.9273 (1.0065) model_time 0.9272 (1.0052) loss 0.6827 (0.8234) grad_norm 8.0289 (8.3944/1.9715) mem 68106MB [2022-12-20 06:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][820/1519] eta 0:11:43 lr 0.000012 time 0.9313 (1.0067) model_time 0.9312 (1.0054) loss 0.6897 (0.8240) grad_norm 7.7021 (8.4223/1.9945) mem 68106MB [2022-12-20 06:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][830/1519] eta 0:11:33 lr 0.000012 time 0.9300 (1.0066) model_time 0.9299 (1.0053) loss 0.8101 (0.8235) grad_norm 6.0896 (8.4132/2.0102) mem 68106MB [2022-12-20 06:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][840/1519] eta 0:11:23 lr 0.000012 time 0.9381 (1.0066) model_time 0.9380 (1.0053) loss 0.9251 (0.8238) grad_norm 7.7746 (8.3834/1.9765) mem 68106MB [2022-12-20 06:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][850/1519] eta 0:11:13 lr 0.000012 time 0.9627 (1.0066) model_time 0.9626 (1.0053) loss 0.8858 (0.8243) grad_norm 6.9358 (8.3863/1.9792) mem 68106MB [2022-12-20 06:27:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][860/1519] eta 0:11:03 lr 0.000012 time 0.9243 (1.0065) model_time 0.9242 (1.0052) loss 0.8098 (0.8239) grad_norm 8.0243 (8.3831/1.9808) mem 68106MB [2022-12-20 06:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][870/1519] eta 0:10:53 lr 0.000012 time 0.9352 (1.0064) model_time 0.9351 (1.0052) loss 0.9320 (0.8244) grad_norm 6.5057 (8.4257/1.9892) mem 68106MB [2022-12-20 06:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][880/1519] eta 0:10:43 lr 0.000012 time 0.9289 (1.0064) model_time 0.9288 (1.0051) loss 0.7307 (0.8255) grad_norm 7.2157 (8.4297/1.9823) mem 68106MB [2022-12-20 06:28:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][890/1519] eta 0:10:32 lr 0.000012 time 0.9279 (1.0063) model_time 0.9278 (1.0051) loss 1.0430 (0.8255) grad_norm 9.6769 (8.4301/2.0201) mem 68106MB [2022-12-20 06:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][900/1519] eta 0:10:22 lr 0.000012 time 0.9166 (1.0063) model_time 0.9165 (1.0051) loss 0.7125 (0.8250) grad_norm 9.3338 (8.4548/2.0230) mem 68106MB [2022-12-20 06:28:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][910/1519] eta 0:10:12 lr 0.000012 time 0.9293 (1.0062) model_time 0.9292 (1.0050) loss 0.8242 (0.8253) grad_norm 6.5686 (8.4394/2.0353) mem 68106MB [2022-12-20 06:28:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][920/1519] eta 0:10:02 lr 0.000012 time 0.9294 (1.0064) model_time 0.9292 (1.0052) loss 0.8639 (0.8253) grad_norm 8.0400 (8.4640/2.0378) mem 68106MB [2022-12-20 06:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][930/1519] eta 0:09:52 lr 0.000012 time 0.9323 (1.0064) model_time 0.9321 (1.0052) loss 1.1001 (0.8262) grad_norm 10.1287 (8.4816/2.0297) mem 68106MB [2022-12-20 06:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][940/1519] eta 0:09:42 lr 0.000012 time 0.9309 (1.0063) model_time 0.9307 (1.0051) loss 0.7058 (0.8261) grad_norm 7.7465 (8.4911/2.0243) mem 68106MB [2022-12-20 06:29:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][950/1519] eta 0:09:32 lr 0.000012 time 0.9265 (1.0062) model_time 0.9263 (1.0051) loss 0.6998 (0.8252) grad_norm 5.2728 (8.5171/2.0523) mem 68106MB [2022-12-20 06:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][960/1519] eta 0:09:22 lr 0.000012 time 0.9147 (1.0066) model_time 0.9146 (1.0054) loss 0.6839 (0.8247) grad_norm 9.0894 (8.5141/2.0364) mem 68106MB [2022-12-20 06:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][970/1519] eta 0:09:12 lr 0.000012 time 1.0158 (1.0066) model_time 1.0157 (1.0054) loss 0.8034 (0.8248) grad_norm 10.2814 (8.5399/2.0392) mem 68106MB [2022-12-20 06:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][980/1519] eta 0:09:02 lr 0.000012 time 0.9334 (1.0065) model_time 0.9332 (1.0054) loss 0.6760 (0.8243) grad_norm 8.3478 (8.5710/2.0289) mem 68106MB [2022-12-20 06:29:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][990/1519] eta 0:08:52 lr 0.000012 time 0.9345 (1.0066) model_time 0.9343 (1.0055) loss 1.1843 (0.8250) grad_norm 9.8835 (8.5495/2.0016) mem 68106MB [2022-12-20 06:29:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1000/1519] eta 0:08:42 lr 0.000012 time 0.9320 (1.0065) model_time 0.9319 (1.0054) loss 0.8759 (0.8246) grad_norm 7.9283 (8.5401/2.0260) mem 68106MB [2022-12-20 06:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1010/1519] eta 0:08:32 lr 0.000012 time 0.9297 (1.0064) model_time 0.9296 (1.0053) loss 1.0509 (0.8252) grad_norm 9.1774 (8.5420/2.0299) mem 68106MB [2022-12-20 06:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1020/1519] eta 0:08:22 lr 0.000012 time 0.9331 (1.0064) model_time 0.9329 (1.0052) loss 0.7954 (0.8248) grad_norm 9.0667 (8.5802/2.0324) mem 68106MB [2022-12-20 06:30:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1030/1519] eta 0:08:12 lr 0.000012 time 0.9128 (1.0063) model_time 0.9127 (1.0052) loss 0.8479 (0.8247) grad_norm 14.6604 (8.6146/2.0612) mem 68106MB [2022-12-20 06:30:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1040/1519] eta 0:08:02 lr 0.000012 time 0.9505 (1.0063) model_time 0.9503 (1.0052) loss 0.7116 (0.8250) grad_norm 7.6113 (8.6262/2.0543) mem 68106MB [2022-12-20 06:30:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1050/1519] eta 0:07:51 lr 0.000012 time 0.9445 (1.0063) model_time 0.9444 (1.0052) loss 0.6858 (0.8252) grad_norm 12.9428 (8.6358/2.0761) mem 68106MB [2022-12-20 06:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1060/1519] eta 0:07:41 lr 0.000012 time 0.9278 (1.0062) model_time 0.9276 (1.0051) loss 0.8170 (0.8246) grad_norm 8.1098 (8.6269/2.0575) mem 68106MB [2022-12-20 06:31:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1070/1519] eta 0:07:31 lr 0.000012 time 0.9327 (1.0062) model_time 0.9326 (1.0051) loss 1.1355 (0.8253) grad_norm 6.4899 (8.5616/1.8154) mem 68106MB [2022-12-20 06:31:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1080/1519] eta 0:07:21 lr 0.000012 time 0.9295 (1.0061) model_time 0.9294 (1.0050) loss 0.6891 (0.8253) grad_norm 8.3081 (8.5394/1.7969) mem 68106MB [2022-12-20 06:31:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1090/1519] eta 0:07:11 lr 0.000012 time 0.9293 (1.0061) model_time 0.9292 (1.0050) loss 0.7711 (0.8258) grad_norm 10.4472 (8.5551/1.7961) mem 68106MB [2022-12-20 06:31:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1100/1519] eta 0:07:01 lr 0.000012 time 0.9242 (1.0061) model_time 0.9240 (1.0050) loss 0.8871 (0.8252) grad_norm 5.4924 (8.5422/1.8001) mem 68106MB [2022-12-20 06:31:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1110/1519] eta 0:06:51 lr 0.000012 time 0.9293 (1.0060) model_time 0.9292 (1.0049) loss 0.6973 (0.8248) grad_norm 7.7858 (8.5381/1.7997) mem 68106MB [2022-12-20 06:31:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1120/1519] eta 0:06:41 lr 0.000012 time 0.9320 (1.0059) model_time 0.9318 (1.0049) loss 0.6685 (0.8247) grad_norm 14.3401 (8.5882/1.8413) mem 68106MB [2022-12-20 06:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1130/1519] eta 0:06:31 lr 0.000012 time 0.9333 (1.0059) model_time 0.9331 (1.0049) loss 0.9059 (0.8256) grad_norm 6.6732 (8.5821/1.8510) mem 68106MB [2022-12-20 06:32:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1140/1519] eta 0:06:21 lr 0.000012 time 0.9306 (1.0060) model_time 0.9305 (1.0050) loss 1.1415 (0.8257) grad_norm 6.1841 (8.5532/1.8468) mem 68106MB [2022-12-20 06:32:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1150/1519] eta 0:06:11 lr 0.000012 time 0.9305 (1.0060) model_time 0.9303 (1.0050) loss 0.7092 (0.8258) grad_norm 7.2697 (8.5560/1.8373) mem 68106MB [2022-12-20 06:32:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1160/1519] eta 0:06:01 lr 0.000012 time 0.9332 (1.0060) model_time 0.9331 (1.0050) loss 0.6837 (0.8255) grad_norm 9.3221 (8.5812/1.8348) mem 68106MB [2022-12-20 06:32:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1170/1519] eta 0:05:51 lr 0.000012 time 0.9338 (1.0060) model_time 0.9337 (1.0049) loss 0.9879 (0.8259) grad_norm 7.5351 (8.6057/1.8274) mem 68106MB [2022-12-20 06:32:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1180/1519] eta 0:05:41 lr 0.000012 time 0.9563 (1.0060) model_time 0.9561 (1.0049) loss 0.8100 (0.8264) grad_norm 8.8823 (8.6260/1.8301) mem 68106MB [2022-12-20 06:33:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1190/1519] eta 0:05:30 lr 0.000012 time 0.9356 (1.0059) model_time 0.9354 (1.0049) loss 0.8843 (0.8260) grad_norm 9.3407 (8.6401/1.8260) mem 68106MB [2022-12-20 06:33:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1200/1519] eta 0:05:20 lr 0.000012 time 0.9357 (1.0059) model_time 0.9356 (1.0049) loss 0.8071 (0.8265) grad_norm 8.8933 (8.6404/1.8299) mem 68106MB [2022-12-20 06:33:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1210/1519] eta 0:05:10 lr 0.000012 time 0.9325 (1.0059) model_time 0.9323 (1.0049) loss 0.6710 (0.8262) grad_norm 7.3376 (8.6555/1.8294) mem 68106MB [2022-12-20 06:33:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1220/1519] eta 0:05:00 lr 0.000012 time 0.9286 (1.0058) model_time 0.9284 (1.0048) loss 0.6898 (0.8261) grad_norm 9.7035 (8.6820/1.8341) mem 68106MB [2022-12-20 06:33:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1230/1519] eta 0:04:50 lr 0.000012 time 0.9454 (1.0059) model_time 0.9452 (1.0049) loss 0.9557 (0.8267) grad_norm 10.0978 (8.6953/1.8306) mem 68106MB [2022-12-20 06:33:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1240/1519] eta 0:04:40 lr 0.000012 time 0.9296 (1.0058) model_time 0.9294 (1.0048) loss 1.3178 (0.8274) grad_norm 8.5565 (8.7215/1.8494) mem 68106MB [2022-12-20 06:34:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1250/1519] eta 0:04:30 lr 0.000012 time 0.9280 (1.0058) model_time 0.9279 (1.0048) loss 0.7103 (0.8272) grad_norm 15.8478 (8.7522/1.8889) mem 68106MB [2022-12-20 06:34:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1260/1519] eta 0:04:20 lr 0.000012 time 0.9338 (1.0058) model_time 0.9336 (1.0048) loss 0.8249 (0.8272) grad_norm 8.3060 (8.7816/1.8908) mem 68106MB [2022-12-20 06:34:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1270/1519] eta 0:04:10 lr 0.000012 time 0.9306 (1.0059) model_time 0.9304 (1.0049) loss 1.0007 (0.8278) grad_norm 7.6423 (8.7841/1.8768) mem 68106MB [2022-12-20 06:34:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1280/1519] eta 0:04:00 lr 0.000012 time 0.9382 (1.0059) model_time 0.9380 (1.0050) loss 0.8360 (0.8277) grad_norm 8.2746 (8.7797/1.8616) mem 68106MB [2022-12-20 06:34:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1290/1519] eta 0:03:50 lr 0.000012 time 0.9270 (1.0059) model_time 0.9269 (1.0049) loss 0.8586 (0.8275) grad_norm 6.3533 (8.8095/1.8716) mem 68106MB [2022-12-20 06:34:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1300/1519] eta 0:03:40 lr 0.000012 time 0.9898 (1.0059) model_time 0.9897 (1.0050) loss 0.7340 (0.8276) grad_norm 6.9929 (8.7945/1.8720) mem 68106MB [2022-12-20 06:35:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1310/1519] eta 0:03:30 lr 0.000012 time 0.9282 (1.0061) model_time 0.9281 (1.0051) loss 1.0769 (0.8274) grad_norm 9.4540 (8.8005/1.8797) mem 68106MB [2022-12-20 06:35:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1320/1519] eta 0:03:20 lr 0.000012 time 0.9369 (1.0060) model_time 0.9367 (1.0051) loss 0.8947 (0.8276) grad_norm 7.7515 (8.7627/1.8652) mem 68106MB [2022-12-20 06:35:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1330/1519] eta 0:03:10 lr 0.000012 time 0.9335 (1.0060) model_time 0.9333 (1.0051) loss 0.9919 (0.8278) grad_norm 8.6130 (8.7296/1.8680) mem 68106MB [2022-12-20 06:35:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1340/1519] eta 0:03:00 lr 0.000012 time 0.9292 (1.0060) model_time 0.9290 (1.0050) loss 0.8432 (0.8282) grad_norm 10.2002 (8.7496/1.8671) mem 68106MB [2022-12-20 06:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1350/1519] eta 0:02:50 lr 0.000012 time 0.9289 (1.0060) model_time 0.9287 (1.0050) loss 0.9043 (0.8277) grad_norm 6.1644 (8.7320/1.8681) mem 68106MB [2022-12-20 06:35:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1360/1519] eta 0:02:39 lr 0.000012 time 0.9301 (1.0060) model_time 0.9299 (1.0051) loss 0.8436 (0.8283) grad_norm 8.7234 (8.7224/1.8627) mem 68106MB [2022-12-20 06:36:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1370/1519] eta 0:02:29 lr 0.000012 time 0.9383 (1.0060) model_time 0.9382 (1.0050) loss 0.9243 (0.8288) grad_norm 7.3352 (8.6969/1.8727) mem 68106MB [2022-12-20 06:36:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1380/1519] eta 0:02:19 lr 0.000012 time 1.0499 (1.0060) model_time 1.0498 (1.0051) loss 0.8368 (0.8289) grad_norm 8.1596 (8.7121/1.8909) mem 68106MB [2022-12-20 06:36:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1390/1519] eta 0:02:09 lr 0.000012 time 0.9283 (1.0060) model_time 0.9281 (1.0051) loss 0.7361 (0.8289) grad_norm 8.4719 (8.7235/1.8816) mem 68106MB [2022-12-20 06:36:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1400/1519] eta 0:01:59 lr 0.000012 time 0.9357 (1.0060) model_time 0.9355 (1.0051) loss 1.2205 (0.8297) grad_norm 9.7319 (8.7142/1.8828) mem 68106MB [2022-12-20 06:36:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1410/1519] eta 0:01:49 lr 0.000012 time 0.9339 (1.0060) model_time 0.9338 (1.0050) loss 0.9272 (0.8299) grad_norm 6.3014 (8.7377/1.9029) mem 68106MB [2022-12-20 06:36:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1420/1519] eta 0:01:39 lr 0.000012 time 0.9340 (1.0060) model_time 0.9337 (1.0050) loss 0.8737 (0.8304) grad_norm 8.9819 (8.7027/1.8868) mem 68106MB [2022-12-20 06:37:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1430/1519] eta 0:01:29 lr 0.000012 time 0.9273 (1.0059) model_time 0.9271 (1.0050) loss 0.7987 (0.8303) grad_norm 8.2208 (8.7145/1.8920) mem 68106MB [2022-12-20 06:37:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1440/1519] eta 0:01:19 lr 0.000012 time 0.9376 (1.0059) model_time 0.9375 (1.0050) loss 0.8306 (0.8302) grad_norm 6.2633 (8.7256/1.8902) mem 68106MB [2022-12-20 06:37:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1450/1519] eta 0:01:09 lr 0.000012 time 0.9349 (1.0060) model_time 0.9347 (1.0051) loss 0.7239 (0.8303) grad_norm 14.8423 (8.7482/1.9278) mem 68106MB [2022-12-20 06:37:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1460/1519] eta 0:00:59 lr 0.000012 time 0.9283 (1.0060) model_time 0.9281 (1.0050) loss 0.6852 (0.8298) grad_norm 7.6988 (8.7450/1.9288) mem 68106MB [2022-12-20 06:37:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1470/1519] eta 0:00:49 lr 0.000012 time 0.9341 (1.0059) model_time 0.9340 (1.0050) loss 0.8408 (0.8294) grad_norm 7.1598 (8.7106/1.9259) mem 68106MB [2022-12-20 06:37:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1480/1519] eta 0:00:39 lr 0.000012 time 0.9915 (1.0060) model_time 0.9914 (1.0051) loss 0.9080 (0.8295) grad_norm 8.6605 (8.7185/1.9275) mem 68106MB [2022-12-20 06:38:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1490/1519] eta 0:00:29 lr 0.000012 time 0.9223 (1.0059) model_time 0.9222 (1.0050) loss 0.7384 (0.8295) grad_norm 9.8566 (8.7419/1.8898) mem 68106MB [2022-12-20 06:38:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1500/1519] eta 0:00:19 lr 0.000012 time 0.9257 (1.0059) model_time 0.9255 (1.0050) loss 0.6842 (0.8295) grad_norm 8.9560 (8.7381/1.8960) mem 68106MB [2022-12-20 06:38:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [59/100][1510/1519] eta 0:00:09 lr 0.000012 time 0.9227 (1.0059) model_time 0.9226 (1.0050) loss 0.6993 (0.8297) grad_norm 8.2305 (8.7290/1.8862) mem 68106MB [2022-12-20 06:38:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 59 training takes 0:25:27 [2022-12-20 06:38:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_59.pth saving...... [2022-12-20 06:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_59.pth saved !!! [2022-12-20 06:38:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.687 (0.687) Loss 0.5173 (0.5173) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 06:39:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.305 (0.332) Loss 0.5134 (0.4981) Acc@1 93.056 (92.740) Acc@5 98.264 (98.485) Mem 68106MB [2022-12-20 06:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.316) Loss 0.4750 (0.4974) Acc@1 92.014 (92.659) Acc@5 98.958 (98.313) Mem 68106MB [2022-12-20 06:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.303 (0.310) Loss 0.6037 (0.5040) Acc@1 89.583 (92.328) Acc@5 98.264 (98.297) Mem 68106MB [2022-12-20 06:39:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.307) Loss 0.4563 (0.4949) Acc@1 93.403 (92.429) Acc@5 98.611 (98.408) Mem 68106MB [2022-12-20 06:39:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 0.4824 (0.4930) Acc@1 90.625 (92.463) Acc@5 99.653 (98.468) Mem 68106MB [2022-12-20 06:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.304) Loss 0.5124 (0.4929) Acc@1 91.319 (92.441) Acc@5 97.917 (98.446) Mem 68106MB [2022-12-20 06:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.296 (0.304) Loss 0.5246 (0.4934) Acc@1 93.056 (92.410) Acc@5 98.611 (98.460) Mem 68106MB [2022-12-20 06:39:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4272 (0.4914) Acc@1 92.361 (92.404) Acc@5 98.958 (98.504) Mem 68106MB [2022-12-20 06:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:59] * Acc@1 92.362 Acc@5 98.510 [2022-12-20 06:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 06:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 06:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 06:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.36% [2022-12-20 06:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][0/1519] eta 0:35:13 lr 0.000012 time 1.3913 (1.3913) model_time 0.9782 (0.9782) loss 1.1098 (1.1098) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 06:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][10/1519] eta 0:26:10 lr 0.000012 time 0.9840 (1.0405) model_time 0.9839 (1.0026) loss 0.9710 (0.8447) grad_norm 9.6201 (8.2769/0.8416) mem 68106MB [2022-12-20 06:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][20/1519] eta 0:25:36 lr 0.000012 time 0.9316 (1.0253) model_time 0.9314 (1.0052) loss 0.7235 (0.8370) grad_norm 6.2590 (8.4398/1.1300) mem 68106MB [2022-12-20 06:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][30/1519] eta 0:25:14 lr 0.000012 time 0.9286 (1.0168) model_time 0.9284 (1.0031) loss 1.0118 (0.8192) grad_norm 6.7078 (8.3493/1.4002) mem 68106MB [2022-12-20 06:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][40/1519] eta 0:25:00 lr 0.000012 time 1.0096 (1.0143) model_time 1.0093 (1.0038) loss 0.8784 (0.8207) grad_norm 11.9205 (8.3081/1.6312) mem 68106MB [2022-12-20 06:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][50/1519] eta 0:24:46 lr 0.000012 time 0.9227 (1.0120) model_time 0.9225 (1.0035) loss 0.7398 (0.8289) grad_norm 8.6211 (8.1996/1.4889) mem 68106MB [2022-12-20 06:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][60/1519] eta 0:24:34 lr 0.000012 time 0.9994 (1.0109) model_time 0.9992 (1.0037) loss 1.0970 (0.8225) grad_norm 8.8712 (8.0850/1.4785) mem 68106MB [2022-12-20 06:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][70/1519] eta 0:24:25 lr 0.000012 time 0.9380 (1.0115) model_time 0.9378 (1.0053) loss 0.6851 (0.8222) grad_norm 7.9946 (8.1334/1.4734) mem 68106MB [2022-12-20 06:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][80/1519] eta 0:24:15 lr 0.000012 time 0.9277 (1.0117) model_time 0.9275 (1.0062) loss 0.7356 (0.8328) grad_norm 9.1992 (8.1755/1.4042) mem 68106MB [2022-12-20 06:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][90/1519] eta 0:24:03 lr 0.000012 time 0.9321 (1.0102) model_time 0.9319 (1.0053) loss 0.8576 (0.8366) grad_norm 7.8849 (8.2064/1.4221) mem 68106MB [2022-12-20 06:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][100/1519] eta 0:23:51 lr 0.000012 time 0.9284 (1.0090) model_time 0.9283 (1.0045) loss 0.7900 (0.8367) grad_norm 6.1841 (8.0994/1.4069) mem 68106MB [2022-12-20 06:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][110/1519] eta 0:23:41 lr 0.000012 time 0.9298 (1.0092) model_time 0.9292 (1.0050) loss 0.7027 (0.8373) grad_norm 8.8719 (8.0751/1.3723) mem 68106MB [2022-12-20 06:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][120/1519] eta 0:23:31 lr 0.000012 time 0.9362 (1.0086) model_time 0.9361 (1.0048) loss 0.6734 (0.8327) grad_norm 6.9529 (8.1586/1.4426) mem 68106MB [2022-12-20 06:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][130/1519] eta 0:23:20 lr 0.000012 time 0.9306 (1.0082) model_time 0.9304 (1.0046) loss 0.8767 (0.8356) grad_norm 6.3279 (8.0710/1.4773) mem 68106MB [2022-12-20 06:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][140/1519] eta 0:23:11 lr 0.000012 time 0.9444 (1.0090) model_time 0.9441 (1.0057) loss 0.8530 (0.8345) grad_norm 7.7309 (8.0833/1.4369) mem 68106MB [2022-12-20 06:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][150/1519] eta 0:23:01 lr 0.000012 time 0.9954 (1.0090) model_time 0.9953 (1.0059) loss 0.9000 (0.8381) grad_norm 8.2839 (8.0485/1.4216) mem 68106MB [2022-12-20 06:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][160/1519] eta 0:22:51 lr 0.000012 time 0.9368 (1.0088) model_time 0.9367 (1.0058) loss 0.6916 (0.8379) grad_norm 6.9061 (8.0350/1.4140) mem 68106MB [2022-12-20 06:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][170/1519] eta 0:22:41 lr 0.000012 time 1.0145 (1.0089) model_time 1.0144 (1.0061) loss 1.0178 (0.8381) grad_norm 7.0097 (8.0129/1.3882) mem 68106MB [2022-12-20 06:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][180/1519] eta 0:22:30 lr 0.000012 time 0.9380 (1.0084) model_time 0.9378 (1.0057) loss 0.8964 (0.8384) grad_norm 6.8714 (8.0019/1.3679) mem 68106MB [2022-12-20 06:43:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][190/1519] eta 0:22:19 lr 0.000012 time 0.9289 (1.0081) model_time 0.9287 (1.0056) loss 0.9217 (0.8369) grad_norm 5.3863 (7.9845/1.3738) mem 68106MB [2022-12-20 06:43:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][200/1519] eta 0:22:09 lr 0.000012 time 0.9303 (1.0078) model_time 0.9301 (1.0053) loss 0.6976 (0.8329) grad_norm 6.9892 (8.0257/1.3861) mem 68106MB [2022-12-20 06:43:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][210/1519] eta 0:21:58 lr 0.000012 time 0.9906 (1.0076) model_time 0.9905 (1.0052) loss 1.1025 (0.8351) grad_norm 5.1693 (7.9996/1.3836) mem 68106MB [2022-12-20 06:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][220/1519] eta 0:21:48 lr 0.000012 time 0.9952 (1.0075) model_time 0.9947 (1.0052) loss 0.6981 (0.8404) grad_norm 10.0772 (8.0397/1.3886) mem 68106MB [2022-12-20 06:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][230/1519] eta 0:21:38 lr 0.000012 time 0.9309 (1.0070) model_time 0.9308 (1.0048) loss 0.6979 (0.8404) grad_norm 10.3448 (8.0828/1.3825) mem 68106MB [2022-12-20 06:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][240/1519] eta 0:21:28 lr 0.000012 time 1.1760 (1.0077) model_time 1.1758 (1.0055) loss 0.7069 (0.8384) grad_norm 9.9997 (8.1614/1.4562) mem 68106MB [2022-12-20 06:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][250/1519] eta 0:21:18 lr 0.000012 time 0.9223 (1.0075) model_time 0.9221 (1.0054) loss 0.8857 (0.8372) grad_norm 14.0972 (8.2287/1.6005) mem 68106MB [2022-12-20 06:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][260/1519] eta 0:21:08 lr 0.000012 time 0.9294 (1.0072) model_time 0.9292 (1.0051) loss 0.7817 (0.8368) grad_norm 7.2596 (8.2244/1.5922) mem 68106MB [2022-12-20 06:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][270/1519] eta 0:20:57 lr 0.000012 time 0.9334 (1.0068) model_time 0.9332 (1.0049) loss 0.6911 (0.8353) grad_norm 9.7301 (8.2372/1.5765) mem 68106MB [2022-12-20 06:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][280/1519] eta 0:20:47 lr 0.000012 time 0.9331 (1.0068) model_time 0.9327 (1.0049) loss 0.7667 (0.8330) grad_norm 7.5544 (8.2346/1.5646) mem 68106MB [2022-12-20 06:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][290/1519] eta 0:20:38 lr 0.000012 time 0.9321 (1.0074) model_time 0.9319 (1.0055) loss 0.7521 (0.8339) grad_norm 8.6478 (8.2718/1.5885) mem 68106MB [2022-12-20 06:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][300/1519] eta 0:20:27 lr 0.000012 time 0.9327 (1.0071) model_time 0.9325 (1.0053) loss 0.8807 (0.8343) grad_norm 8.2934 (8.2778/1.5776) mem 68106MB [2022-12-20 06:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][310/1519] eta 0:20:18 lr 0.000012 time 0.9319 (1.0077) model_time 0.9317 (1.0059) loss 0.8894 (0.8352) grad_norm 8.3859 (8.2740/1.5599) mem 68106MB [2022-12-20 06:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][320/1519] eta 0:20:07 lr 0.000012 time 0.9277 (1.0074) model_time 0.9275 (1.0057) loss 0.6625 (0.8373) grad_norm 16.8692 (8.3562/1.6892) mem 68106MB [2022-12-20 06:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][330/1519] eta 0:19:57 lr 0.000012 time 0.9959 (1.0075) model_time 0.9958 (1.0058) loss 0.8114 (0.8373) grad_norm 18.2522 (8.4264/1.8561) mem 68106MB [2022-12-20 06:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][340/1519] eta 0:19:47 lr 0.000012 time 0.9345 (1.0072) model_time 0.9343 (1.0055) loss 0.7845 (0.8359) grad_norm 7.7615 (8.4438/1.8398) mem 68106MB [2022-12-20 06:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][350/1519] eta 0:19:37 lr 0.000012 time 0.9340 (1.0070) model_time 0.9338 (1.0054) loss 0.6695 (0.8352) grad_norm 8.6970 (8.4441/1.8499) mem 68106MB [2022-12-20 06:45:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][360/1519] eta 0:19:27 lr 0.000012 time 0.9619 (1.0070) model_time 0.9617 (1.0054) loss 0.7544 (0.8370) grad_norm 7.2303 (8.4275/1.8278) mem 68106MB [2022-12-20 06:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][370/1519] eta 0:19:16 lr 0.000012 time 0.9305 (1.0069) model_time 0.9303 (1.0053) loss 0.7581 (0.8390) grad_norm 13.9194 (8.4550/1.8524) mem 68106MB [2022-12-20 06:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][380/1519] eta 0:19:06 lr 0.000012 time 0.9174 (1.0068) model_time 0.9172 (1.0053) loss 0.6795 (0.8381) grad_norm 6.1629 (8.4593/1.8502) mem 68106MB [2022-12-20 06:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][390/1519] eta 0:18:56 lr 0.000012 time 0.9708 (1.0070) model_time 0.9706 (1.0054) loss 0.9181 (0.8363) grad_norm 9.1838 (8.4860/1.8390) mem 68106MB [2022-12-20 06:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][400/1519] eta 0:18:46 lr 0.000012 time 1.0019 (1.0069) model_time 1.0017 (1.0054) loss 0.8812 (0.8362) grad_norm 9.7597 (8.5048/1.8506) mem 68106MB [2022-12-20 06:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][410/1519] eta 0:18:36 lr 0.000012 time 0.9257 (1.0068) model_time 0.9255 (1.0052) loss 0.7139 (0.8354) grad_norm 6.2765 (8.4950/1.8443) mem 68106MB [2022-12-20 06:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][420/1519] eta 0:18:26 lr 0.000012 time 0.9326 (1.0066) model_time 0.9324 (1.0051) loss 0.8603 (0.8347) grad_norm 7.4562 (8.5010/1.8384) mem 68106MB [2022-12-20 06:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][430/1519] eta 0:18:16 lr 0.000012 time 0.9262 (1.0064) model_time 0.9261 (1.0050) loss 0.6785 (0.8351) grad_norm 7.4837 (8.4904/1.8236) mem 68106MB [2022-12-20 06:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][440/1519] eta 0:18:05 lr 0.000012 time 0.9294 (1.0063) model_time 0.9292 (1.0048) loss 0.7236 (0.8336) grad_norm 7.9411 (8.4919/1.8121) mem 68106MB [2022-12-20 06:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][450/1519] eta 0:17:55 lr 0.000012 time 0.9198 (1.0064) model_time 0.9197 (1.0050) loss 0.9939 (0.8353) grad_norm 6.8392 (8.5057/1.8241) mem 68106MB [2022-12-20 06:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][460/1519] eta 0:17:45 lr 0.000012 time 0.9277 (1.0064) model_time 0.9274 (1.0050) loss 0.8086 (0.8366) grad_norm 10.0591 (8.5295/1.8151) mem 68106MB [2022-12-20 06:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][470/1519] eta 0:17:35 lr 0.000012 time 0.9028 (1.0064) model_time 0.9025 (1.0051) loss 0.6698 (0.8348) grad_norm 7.0774 (8.4952/1.8156) mem 68106MB [2022-12-20 06:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][480/1519] eta 0:17:25 lr 0.000012 time 0.9316 (1.0063) model_time 0.9314 (1.0049) loss 0.6694 (0.8338) grad_norm 12.8186 (8.5125/1.8190) mem 68106MB [2022-12-20 06:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][490/1519] eta 0:17:15 lr 0.000012 time 0.9395 (1.0063) model_time 0.9394 (1.0050) loss 0.9350 (0.8344) grad_norm 8.8267 (8.5077/1.8105) mem 68106MB [2022-12-20 06:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][500/1519] eta 0:17:05 lr 0.000012 time 0.9337 (1.0061) model_time 0.9336 (1.0048) loss 0.7024 (0.8331) grad_norm 11.2602 (8.5083/1.8091) mem 68106MB [2022-12-20 06:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][510/1519] eta 0:16:55 lr 0.000012 time 0.9316 (1.0060) model_time 0.9314 (1.0047) loss 0.9739 (0.8333) grad_norm 7.1064 (8.5057/1.7973) mem 68106MB [2022-12-20 06:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][520/1519] eta 0:16:44 lr 0.000012 time 0.9371 (1.0058) model_time 0.9370 (1.0045) loss 0.8391 (0.8332) grad_norm 6.4446 (8.4791/1.7929) mem 68106MB [2022-12-20 06:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][530/1519] eta 0:16:34 lr 0.000012 time 0.9290 (1.0057) model_time 0.9289 (1.0044) loss 0.7729 (0.8337) grad_norm 7.4178 (8.5023/1.7935) mem 68106MB [2022-12-20 06:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][540/1519] eta 0:16:24 lr 0.000012 time 0.9317 (1.0055) model_time 0.9316 (1.0043) loss 0.6696 (0.8324) grad_norm 8.9699 (8.4964/1.7945) mem 68106MB [2022-12-20 06:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][550/1519] eta 0:16:14 lr 0.000012 time 0.9259 (1.0054) model_time 0.9258 (1.0042) loss 0.7376 (0.8314) grad_norm 14.6803 (8.5085/1.8211) mem 68106MB [2022-12-20 06:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][560/1519] eta 0:16:04 lr 0.000012 time 0.9400 (1.0054) model_time 0.9398 (1.0042) loss 0.6829 (0.8299) grad_norm 6.1716 (8.4800/1.8201) mem 68106MB [2022-12-20 06:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][570/1519] eta 0:15:54 lr 0.000012 time 0.9322 (1.0056) model_time 0.9321 (1.0044) loss 1.1047 (0.8307) grad_norm 6.1012 (8.4766/1.8200) mem 68106MB [2022-12-20 06:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][580/1519] eta 0:15:44 lr 0.000012 time 0.9284 (1.0055) model_time 0.9282 (1.0043) loss 0.7371 (0.8307) grad_norm 7.5180 (8.4622/1.8098) mem 68106MB [2022-12-20 06:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][590/1519] eta 0:15:34 lr 0.000012 time 0.9231 (1.0054) model_time 0.9229 (1.0042) loss 0.6839 (0.8299) grad_norm 7.0631 (8.4471/1.7987) mem 68106MB [2022-12-20 06:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][600/1519] eta 0:15:23 lr 0.000012 time 0.9236 (1.0054) model_time 0.9234 (1.0043) loss 0.8715 (0.8306) grad_norm 9.0322 (8.4550/1.7908) mem 68106MB [2022-12-20 06:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][610/1519] eta 0:15:13 lr 0.000012 time 0.9297 (1.0053) model_time 0.9296 (1.0042) loss 0.8254 (0.8309) grad_norm 6.6964 (8.4536/1.8102) mem 68106MB [2022-12-20 06:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][620/1519] eta 0:15:03 lr 0.000012 time 0.9208 (1.0055) model_time 0.9206 (1.0044) loss 0.8218 (0.8330) grad_norm 10.3792 (8.4501/1.8112) mem 68106MB [2022-12-20 06:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][630/1519] eta 0:14:53 lr 0.000012 time 0.9186 (1.0053) model_time 0.9185 (1.0042) loss 1.0295 (0.8323) grad_norm 7.6890 (8.4477/1.8051) mem 68106MB [2022-12-20 06:50:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][640/1519] eta 0:14:43 lr 0.000012 time 0.9222 (1.0053) model_time 0.9221 (1.0042) loss 0.7335 (0.8312) grad_norm 10.3854 (8.4574/1.7957) mem 68106MB [2022-12-20 06:50:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][650/1519] eta 0:14:33 lr 0.000012 time 0.9271 (1.0056) model_time 0.9269 (1.0045) loss 0.7734 (0.8321) grad_norm 7.6105 (8.4643/1.7958) mem 68106MB [2022-12-20 06:50:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][660/1519] eta 0:14:23 lr 0.000012 time 0.9327 (1.0055) model_time 0.9325 (1.0044) loss 0.8167 (0.8312) grad_norm 7.3593 (8.4756/1.7906) mem 68106MB [2022-12-20 06:51:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][670/1519] eta 0:14:13 lr 0.000012 time 0.9364 (1.0055) model_time 0.9363 (1.0044) loss 0.8306 (0.8314) grad_norm 7.3735 (8.4851/1.8065) mem 68106MB [2022-12-20 06:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][680/1519] eta 0:14:03 lr 0.000012 time 0.9375 (1.0055) model_time 0.9374 (1.0044) loss 0.9192 (0.8304) grad_norm 7.2280 (8.4911/1.8232) mem 68106MB [2022-12-20 06:51:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][690/1519] eta 0:13:53 lr 0.000012 time 0.9616 (1.0055) model_time 0.9614 (1.0045) loss 0.8010 (0.8310) grad_norm 9.9676 (8.4876/1.8207) mem 68106MB [2022-12-20 06:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][700/1519] eta 0:13:43 lr 0.000012 time 0.9203 (1.0054) model_time 0.9202 (1.0044) loss 0.8710 (0.8321) grad_norm 8.1963 (8.4996/1.8231) mem 68106MB [2022-12-20 06:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][710/1519] eta 0:13:33 lr 0.000012 time 0.9352 (1.0055) model_time 0.9350 (1.0045) loss 0.7564 (0.8329) grad_norm 8.3133 (8.5051/1.8223) mem 68106MB [2022-12-20 06:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][720/1519] eta 0:13:23 lr 0.000012 time 0.9321 (1.0055) model_time 0.9319 (1.0045) loss 0.7111 (0.8330) grad_norm 11.4483 (8.5239/1.8281) mem 68106MB [2022-12-20 06:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][730/1519] eta 0:13:13 lr 0.000012 time 0.9223 (1.0055) model_time 0.9222 (1.0044) loss 0.6923 (0.8331) grad_norm 13.1011 (8.5851/1.8562) mem 68106MB [2022-12-20 06:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][740/1519] eta 0:13:03 lr 0.000012 time 0.9254 (1.0054) model_time 0.9252 (1.0044) loss 0.7084 (0.8324) grad_norm 8.3924 (8.5791/1.8582) mem 68106MB [2022-12-20 06:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][750/1519] eta 0:12:53 lr 0.000012 time 0.9203 (1.0053) model_time 0.9202 (1.0043) loss 0.6722 (0.8318) grad_norm 6.0162 (8.5895/1.8643) mem 68106MB [2022-12-20 06:52:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][760/1519] eta 0:12:42 lr 0.000012 time 0.9705 (1.0053) model_time 0.9703 (1.0043) loss 0.6619 (0.8320) grad_norm 7.8785 (8.5945/1.8592) mem 68106MB [2022-12-20 06:52:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][770/1519] eta 0:12:33 lr 0.000012 time 0.9195 (1.0054) model_time 0.9193 (1.0044) loss 0.6867 (0.8317) grad_norm 7.2335 (8.5998/1.8701) mem 68106MB [2022-12-20 06:52:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][780/1519] eta 0:12:22 lr 0.000012 time 0.9246 (1.0054) model_time 0.9244 (1.0044) loss 0.7536 (0.8319) grad_norm 9.6637 (8.6069/1.8736) mem 68106MB [2022-12-20 06:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][790/1519] eta 0:12:12 lr 0.000012 time 0.9303 (1.0053) model_time 0.9302 (1.0044) loss 1.0114 (0.8318) grad_norm 6.7306 (8.6033/1.8698) mem 68106MB [2022-12-20 06:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][800/1519] eta 0:12:03 lr 0.000012 time 0.9269 (1.0057) model_time 0.9268 (1.0048) loss 0.7053 (0.8318) grad_norm 9.5324 (8.6073/1.8688) mem 68106MB [2022-12-20 06:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][810/1519] eta 0:11:53 lr 0.000012 time 0.9296 (1.0057) model_time 0.9295 (1.0047) loss 0.9001 (0.8313) grad_norm 8.1128 (8.6291/1.8654) mem 68106MB [2022-12-20 06:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][820/1519] eta 0:11:42 lr 0.000012 time 0.9199 (1.0055) model_time 0.9198 (1.0046) loss 1.1089 (0.8309) grad_norm 9.6880 (8.6257/1.8616) mem 68106MB [2022-12-20 06:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][830/1519] eta 0:11:32 lr 0.000012 time 0.9258 (1.0058) model_time 0.9256 (1.0049) loss 0.7975 (0.8315) grad_norm 10.0461 (8.6502/1.8891) mem 68106MB [2022-12-20 06:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][840/1519] eta 0:11:22 lr 0.000012 time 0.9303 (1.0057) model_time 0.9302 (1.0048) loss 0.6993 (0.8309) grad_norm 8.8275 (8.6251/1.8695) mem 68106MB [2022-12-20 06:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][850/1519] eta 0:11:12 lr 0.000012 time 0.9228 (1.0057) model_time 0.9226 (1.0048) loss 0.6924 (0.8307) grad_norm 11.2556 (8.6019/1.8278) mem 68106MB [2022-12-20 06:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][860/1519] eta 0:11:02 lr 0.000012 time 0.9246 (1.0056) model_time 0.9244 (1.0047) loss 0.6850 (0.8313) grad_norm 11.6846 (8.6316/1.8322) mem 68106MB [2022-12-20 06:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][870/1519] eta 0:10:52 lr 0.000012 time 0.9993 (1.0058) model_time 0.9992 (1.0049) loss 1.0470 (0.8312) grad_norm 7.0349 (8.6332/1.8311) mem 68106MB [2022-12-20 06:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][880/1519] eta 0:10:42 lr 0.000012 time 0.9291 (1.0058) model_time 0.9290 (1.0049) loss 0.6922 (0.8307) grad_norm 7.5492 (8.6467/1.8348) mem 68106MB [2022-12-20 06:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][890/1519] eta 0:10:32 lr 0.000012 time 0.9408 (1.0057) model_time 0.9407 (1.0049) loss 0.7892 (0.8306) grad_norm 8.7566 (8.6240/1.8209) mem 68106MB [2022-12-20 06:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][900/1519] eta 0:10:22 lr 0.000012 time 0.9323 (1.0057) model_time 0.9321 (1.0048) loss 0.8626 (0.8297) grad_norm 6.8488 (8.6458/1.8738) mem 68106MB [2022-12-20 06:55:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][910/1519] eta 0:10:12 lr 0.000012 time 0.9247 (1.0059) model_time 0.9245 (1.0050) loss 0.6846 (0.8295) grad_norm 8.8347 (8.6518/1.8752) mem 68106MB [2022-12-20 06:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][920/1519] eta 0:10:02 lr 0.000012 time 0.9235 (1.0058) model_time 0.9234 (1.0049) loss 0.6769 (0.8289) grad_norm 8.1563 (8.6197/1.8140) mem 68106MB [2022-12-20 06:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][930/1519] eta 0:09:52 lr 0.000012 time 0.9692 (1.0058) model_time 0.9690 (1.0050) loss 0.6968 (0.8291) grad_norm 6.3860 (8.5687/1.7260) mem 68106MB [2022-12-20 06:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][940/1519] eta 0:09:42 lr 0.000012 time 0.9688 (1.0058) model_time 0.9686 (1.0049) loss 0.6690 (0.8287) grad_norm 6.8161 (8.5556/1.7349) mem 68106MB [2022-12-20 06:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][950/1519] eta 0:09:32 lr 0.000012 time 0.9392 (1.0058) model_time 0.9391 (1.0049) loss 1.1694 (0.8293) grad_norm 10.1717 (8.5637/1.7497) mem 68106MB [2022-12-20 06:55:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][960/1519] eta 0:09:22 lr 0.000012 time 0.9265 (1.0059) model_time 0.9264 (1.0050) loss 0.9671 (0.8297) grad_norm 6.1718 (8.5890/1.7621) mem 68106MB [2022-12-20 06:56:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][970/1519] eta 0:09:12 lr 0.000012 time 0.9279 (1.0058) model_time 0.9277 (1.0050) loss 0.7999 (0.8295) grad_norm 9.5479 (8.5764/1.7425) mem 68106MB [2022-12-20 06:56:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][980/1519] eta 0:09:02 lr 0.000012 time 0.9286 (1.0058) model_time 0.9284 (1.0050) loss 1.0056 (0.8294) grad_norm 10.1794 (8.5594/1.7427) mem 68106MB [2022-12-20 06:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][990/1519] eta 0:08:52 lr 0.000012 time 0.9258 (1.0058) model_time 0.9257 (1.0050) loss 0.8150 (0.8292) grad_norm 7.4569 (8.5450/1.7641) mem 68106MB [2022-12-20 06:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1000/1519] eta 0:08:41 lr 0.000012 time 0.9235 (1.0057) model_time 0.9233 (1.0049) loss 0.9535 (0.8291) grad_norm 8.7582 (8.5261/1.7421) mem 68106MB [2022-12-20 06:56:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1010/1519] eta 0:08:31 lr 0.000012 time 0.9286 (1.0057) model_time 0.9284 (1.0049) loss 0.8708 (0.8291) grad_norm 6.9982 (8.5221/1.7410) mem 68106MB [2022-12-20 06:56:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1020/1519] eta 0:08:21 lr 0.000012 time 0.9269 (1.0057) model_time 0.9267 (1.0048) loss 0.9345 (0.8290) grad_norm 7.9434 (8.5203/1.7545) mem 68106MB [2022-12-20 06:57:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1030/1519] eta 0:08:11 lr 0.000012 time 0.9354 (1.0057) model_time 0.9351 (1.0048) loss 0.9980 (0.8293) grad_norm 7.5878 (8.5345/1.7763) mem 68106MB [2022-12-20 06:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1040/1519] eta 0:08:01 lr 0.000012 time 0.9207 (1.0056) model_time 0.9205 (1.0048) loss 0.6695 (0.8287) grad_norm 11.3507 (8.5417/1.7803) mem 68106MB [2022-12-20 06:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1050/1519] eta 0:07:51 lr 0.000012 time 0.9330 (1.0056) model_time 0.9328 (1.0048) loss 0.9546 (0.8286) grad_norm 7.3805 (8.5442/1.7703) mem 68106MB [2022-12-20 06:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1060/1519] eta 0:07:41 lr 0.000012 time 0.9246 (1.0055) model_time 0.9244 (1.0047) loss 0.7937 (0.8285) grad_norm 7.0104 (8.5104/1.7708) mem 68106MB [2022-12-20 06:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1070/1519] eta 0:07:31 lr 0.000012 time 0.9252 (1.0054) model_time 0.9251 (1.0046) loss 0.7302 (0.8292) grad_norm 5.5374 (8.5272/1.7657) mem 68106MB [2022-12-20 06:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1080/1519] eta 0:07:21 lr 0.000012 time 0.9697 (1.0055) model_time 0.9695 (1.0047) loss 0.6788 (0.8289) grad_norm 8.8736 (8.5076/1.7581) mem 68106MB [2022-12-20 06:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1090/1519] eta 0:07:11 lr 0.000012 time 0.9382 (1.0055) model_time 0.9380 (1.0047) loss 0.6776 (0.8286) grad_norm 13.0833 (8.5041/1.7815) mem 68106MB [2022-12-20 06:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1100/1519] eta 0:07:01 lr 0.000012 time 0.9233 (1.0054) model_time 0.9230 (1.0046) loss 1.0358 (0.8289) grad_norm 7.6091 (8.4918/1.7756) mem 68106MB [2022-12-20 06:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1110/1519] eta 0:06:51 lr 0.000012 time 0.9787 (1.0056) model_time 0.9786 (1.0048) loss 0.7574 (0.8292) grad_norm 6.3027 (8.4669/1.7987) mem 68106MB [2022-12-20 06:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1120/1519] eta 0:06:41 lr 0.000012 time 0.9214 (1.0055) model_time 0.9213 (1.0048) loss 0.6717 (0.8291) grad_norm 8.4310 (8.4865/1.7910) mem 68106MB [2022-12-20 06:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1130/1519] eta 0:06:31 lr 0.000012 time 0.9189 (1.0055) model_time 0.9188 (1.0047) loss 0.6802 (0.8289) grad_norm 8.9407 (8.4686/1.7833) mem 68106MB [2022-12-20 06:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1140/1519] eta 0:06:21 lr 0.000012 time 0.9264 (1.0056) model_time 0.9263 (1.0048) loss 0.6924 (0.8284) grad_norm 7.9186 (8.4764/1.7701) mem 68106MB [2022-12-20 06:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1150/1519] eta 0:06:11 lr 0.000012 time 0.9268 (1.0055) model_time 0.9265 (1.0047) loss 0.7023 (0.8277) grad_norm 8.2548 (8.4483/1.7356) mem 68106MB [2022-12-20 06:59:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1160/1519] eta 0:06:01 lr 0.000012 time 0.9437 (1.0057) model_time 0.9436 (1.0049) loss 0.9200 (0.8285) grad_norm 9.4660 (8.4789/1.7316) mem 68106MB [2022-12-20 06:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1170/1519] eta 0:05:50 lr 0.000011 time 0.9296 (1.0056) model_time 0.9294 (1.0049) loss 1.0218 (0.8289) grad_norm 8.6815 (8.4865/1.7345) mem 68106MB [2022-12-20 06:59:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1180/1519] eta 0:05:40 lr 0.000011 time 0.9862 (1.0057) model_time 0.9860 (1.0049) loss 0.7642 (0.8290) grad_norm 9.0536 (8.5251/1.8367) mem 68106MB [2022-12-20 06:59:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1190/1519] eta 0:05:30 lr 0.000011 time 1.0465 (1.0058) model_time 1.0463 (1.0051) loss 0.9784 (0.8288) grad_norm 10.1356 (8.5309/1.8396) mem 68106MB [2022-12-20 06:59:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1200/1519] eta 0:05:20 lr 0.000011 time 0.9264 (1.0058) model_time 0.9263 (1.0050) loss 0.7880 (0.8288) grad_norm 7.5484 (8.5157/1.8432) mem 68106MB [2022-12-20 07:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1210/1519] eta 0:05:10 lr 0.000011 time 0.9250 (1.0057) model_time 0.9248 (1.0050) loss 0.6864 (0.8287) grad_norm 10.7255 (8.5232/1.8292) mem 68106MB [2022-12-20 07:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1220/1519] eta 0:05:00 lr 0.000011 time 0.9949 (1.0057) model_time 0.9948 (1.0050) loss 0.8203 (0.8286) grad_norm 7.2648 (8.5503/1.8492) mem 68106MB [2022-12-20 07:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1230/1519] eta 0:04:50 lr 0.000011 time 0.9238 (1.0056) model_time 0.9237 (1.0049) loss 0.7751 (0.8291) grad_norm 9.5144 (8.5532/1.8566) mem 68106MB [2022-12-20 07:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1240/1519] eta 0:04:40 lr 0.000011 time 0.9971 (1.0056) model_time 0.9970 (1.0049) loss 0.6806 (0.8290) grad_norm 10.5657 (8.5569/1.8558) mem 68106MB [2022-12-20 07:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1250/1519] eta 0:04:30 lr 0.000011 time 0.9208 (1.0056) model_time 0.9207 (1.0048) loss 0.8106 (0.8289) grad_norm 7.2057 (8.5490/1.8610) mem 68106MB [2022-12-20 07:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1260/1519] eta 0:04:20 lr 0.000011 time 0.9212 (1.0056) model_time 0.9211 (1.0049) loss 0.8266 (0.8291) grad_norm 8.6159 (8.5353/1.8661) mem 68106MB [2022-12-20 07:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1270/1519] eta 0:04:10 lr 0.000011 time 0.9253 (1.0056) model_time 0.9251 (1.0049) loss 1.1158 (0.8293) grad_norm 9.5465 (8.5597/1.8862) mem 68106MB [2022-12-20 07:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1280/1519] eta 0:04:00 lr 0.000011 time 0.9202 (1.0056) model_time 0.9200 (1.0049) loss 0.7054 (0.8289) grad_norm 8.5754 (8.5658/1.8711) mem 68106MB [2022-12-20 07:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1290/1519] eta 0:03:50 lr 0.000011 time 0.9173 (1.0056) model_time 0.9171 (1.0048) loss 1.1321 (0.8288) grad_norm 8.6046 (8.5598/1.8756) mem 68106MB [2022-12-20 07:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1300/1519] eta 0:03:40 lr 0.000011 time 0.9829 (1.0056) model_time 0.9827 (1.0048) loss 0.9239 (0.8284) grad_norm 8.8590 (8.5790/1.8701) mem 68106MB [2022-12-20 07:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1310/1519] eta 0:03:30 lr 0.000011 time 0.9218 (1.0055) model_time 0.9217 (1.0048) loss 0.7745 (0.8280) grad_norm 7.9887 (8.5904/1.8678) mem 68106MB [2022-12-20 07:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1320/1519] eta 0:03:20 lr 0.000011 time 0.9236 (1.0055) model_time 0.9235 (1.0047) loss 0.7704 (0.8280) grad_norm 7.4742 (8.5652/1.8567) mem 68106MB [2022-12-20 07:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1330/1519] eta 0:03:10 lr 0.000011 time 0.9501 (1.0055) model_time 0.9499 (1.0047) loss 0.8304 (0.8280) grad_norm 12.6678 (8.5423/1.8260) mem 68106MB [2022-12-20 07:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1340/1519] eta 0:02:59 lr 0.000011 time 0.9371 (1.0055) model_time 0.9370 (1.0048) loss 0.8517 (0.8287) grad_norm 9.4509 (8.5482/1.8258) mem 68106MB [2022-12-20 07:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1350/1519] eta 0:02:49 lr 0.000011 time 0.9274 (1.0054) model_time 0.9273 (1.0047) loss 1.1718 (0.8289) grad_norm 6.4963 (8.5583/1.8312) mem 68106MB [2022-12-20 07:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1360/1519] eta 0:02:39 lr 0.000011 time 0.9862 (1.0055) model_time 0.9861 (1.0048) loss 0.7562 (0.8286) grad_norm 8.8841 (8.5814/1.8330) mem 68106MB [2022-12-20 07:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1370/1519] eta 0:02:29 lr 0.000011 time 0.9253 (1.0054) model_time 0.9251 (1.0047) loss 0.9012 (0.8285) grad_norm 10.2408 (8.6035/1.8366) mem 68106MB [2022-12-20 07:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1380/1519] eta 0:02:19 lr 0.000011 time 0.9343 (1.0054) model_time 0.9341 (1.0047) loss 0.9848 (0.8290) grad_norm 9.3599 (8.6028/1.8305) mem 68106MB [2022-12-20 07:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1390/1519] eta 0:02:09 lr 0.000011 time 0.9244 (1.0054) model_time 0.9243 (1.0047) loss 0.8097 (0.8292) grad_norm 8.3039 (8.6018/1.8319) mem 68106MB [2022-12-20 07:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1400/1519] eta 0:01:59 lr 0.000011 time 1.2006 (1.0056) model_time 1.2004 (1.0049) loss 0.7358 (0.8292) grad_norm 10.5971 (8.5880/1.8457) mem 68106MB [2022-12-20 07:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1410/1519] eta 0:01:49 lr 0.000011 time 0.9226 (1.0056) model_time 0.9225 (1.0049) loss 0.7084 (0.8289) grad_norm 9.3811 (8.5876/1.8441) mem 68106MB [2022-12-20 07:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1420/1519] eta 0:01:39 lr 0.000011 time 1.0753 (1.0056) model_time 1.0752 (1.0049) loss 0.7196 (0.8285) grad_norm 6.1753 (8.5927/1.8540) mem 68106MB [2022-12-20 07:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1430/1519] eta 0:01:29 lr 0.000011 time 0.9329 (1.0056) model_time 0.9328 (1.0049) loss 0.7242 (0.8286) grad_norm 10.1677 (8.5585/1.8251) mem 68106MB [2022-12-20 07:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1440/1519] eta 0:01:19 lr 0.000011 time 0.9216 (1.0056) model_time 0.9215 (1.0049) loss 0.9069 (0.8286) grad_norm 8.1574 (8.5597/1.8361) mem 68106MB [2022-12-20 07:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1450/1519] eta 0:01:09 lr 0.000011 time 0.9214 (1.0057) model_time 0.9212 (1.0051) loss 1.0038 (0.8290) grad_norm 9.8676 (8.5828/1.8447) mem 68106MB [2022-12-20 07:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1460/1519] eta 0:00:59 lr 0.000011 time 0.9317 (1.0057) model_time 0.9315 (1.0050) loss 0.8296 (0.8292) grad_norm 6.8485 (8.5473/1.8388) mem 68106MB [2022-12-20 07:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1470/1519] eta 0:00:49 lr 0.000011 time 0.9371 (1.0056) model_time 0.9370 (1.0050) loss 0.7506 (0.8293) grad_norm 7.3138 (8.5487/1.8394) mem 68106MB [2022-12-20 07:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1480/1519] eta 0:00:39 lr 0.000011 time 0.9845 (1.0057) model_time 0.9844 (1.0050) loss 0.7555 (0.8295) grad_norm 11.4158 (8.5318/1.8465) mem 68106MB [2022-12-20 07:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1490/1519] eta 0:00:29 lr 0.000011 time 0.9325 (1.0057) model_time 0.9324 (1.0050) loss 0.8783 (0.8293) grad_norm 9.2591 (8.5399/1.8487) mem 68106MB [2022-12-20 07:04:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1500/1519] eta 0:00:19 lr 0.000011 time 0.9221 (1.0057) model_time 0.9220 (1.0050) loss 0.7848 (0.8296) grad_norm 10.6377 (8.5152/1.8018) mem 68106MB [2022-12-20 07:05:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [60/100][1510/1519] eta 0:00:09 lr 0.000011 time 0.9331 (1.0058) model_time 0.9330 (1.0051) loss 0.9022 (0.8296) grad_norm 10.2292 (8.5546/1.8770) mem 68106MB [2022-12-20 07:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 60 training takes 0:25:27 [2022-12-20 07:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_60.pth saving...... [2022-12-20 07:05:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_60.pth saved !!! [2022-12-20 07:05:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.637 (0.637) Loss 0.5204 (0.5204) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 07:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.328) Loss 0.5160 (0.4985) Acc@1 93.056 (92.803) Acc@5 97.917 (98.422) Mem 68106MB [2022-12-20 07:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.313) Loss 0.4769 (0.4983) Acc@1 91.319 (92.477) Acc@5 99.306 (98.330) Mem 68106MB [2022-12-20 07:05:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.6111 (0.5043) Acc@1 89.236 (92.305) Acc@5 97.569 (98.354) Mem 68106MB [2022-12-20 07:05:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.294 (0.306) Loss 0.4388 (0.4934) Acc@1 93.056 (92.471) Acc@5 99.306 (98.501) Mem 68106MB [2022-12-20 07:05:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.305) Loss 0.4889 (0.4915) Acc@1 91.319 (92.463) Acc@5 99.306 (98.550) Mem 68106MB [2022-12-20 07:05:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.304) Loss 0.5161 (0.4908) Acc@1 90.625 (92.407) Acc@5 97.569 (98.514) Mem 68106MB [2022-12-20 07:06:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5334 (0.4917) Acc@1 92.014 (92.366) Acc@5 98.264 (98.494) Mem 68106MB [2022-12-20 07:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.299 (0.302) Loss 0.4301 (0.4899) Acc@1 93.056 (92.387) Acc@5 98.264 (98.530) Mem 68106MB [2022-12-20 07:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:60] * Acc@1 92.346 Acc@5 98.527 [2022-12-20 07:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.3% [2022-12-20 07:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.36% [2022-12-20 07:06:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][0/1519] eta 0:46:21 lr 0.000011 time 1.8312 (1.8312) model_time 1.0745 (1.0745) loss 1.1660 (1.1660) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 07:06:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][10/1519] eta 0:27:02 lr 0.000011 time 0.9281 (1.0753) model_time 0.9280 (1.0062) loss 0.8457 (0.8545) grad_norm 7.1034 (9.0795/1.8065) mem 68106MB [2022-12-20 07:06:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][20/1519] eta 0:25:57 lr 0.000011 time 0.9240 (1.0393) model_time 0.9239 (1.0030) loss 0.8086 (0.8315) grad_norm 8.8986 (9.0034/1.5221) mem 68106MB [2022-12-20 07:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][30/1519] eta 0:25:27 lr 0.000011 time 0.9229 (1.0262) model_time 0.9227 (1.0015) loss 0.7772 (0.8170) grad_norm 9.0989 (8.7563/1.3372) mem 68106MB [2022-12-20 07:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][40/1519] eta 0:25:07 lr 0.000011 time 0.9311 (1.0195) model_time 0.9310 (1.0007) loss 0.7883 (0.8002) grad_norm 13.8261 (9.1477/1.8188) mem 68106MB [2022-12-20 07:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][50/1519] eta 0:24:54 lr 0.000011 time 0.9288 (1.0176) model_time 0.9287 (1.0024) loss 0.7001 (0.7983) grad_norm 11.1788 (9.1231/1.7331) mem 68106MB [2022-12-20 07:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][60/1519] eta 0:24:40 lr 0.000011 time 0.9271 (1.0145) model_time 0.9270 (1.0017) loss 0.7287 (0.8077) grad_norm 8.7434 (9.1552/1.8713) mem 68106MB [2022-12-20 07:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][70/1519] eta 0:24:29 lr 0.000011 time 0.9163 (1.0139) model_time 0.9162 (1.0029) loss 0.6818 (0.8100) grad_norm 11.0439 (9.1153/1.9276) mem 68106MB [2022-12-20 07:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][80/1519] eta 0:24:16 lr 0.000011 time 0.9348 (1.0120) model_time 0.9346 (1.0023) loss 0.7057 (0.8099) grad_norm 6.9617 (8.9828/1.8685) mem 68106MB [2022-12-20 07:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][90/1519] eta 0:24:05 lr 0.000011 time 0.9073 (1.0112) model_time 0.9072 (1.0025) loss 0.9536 (0.8087) grad_norm 8.3178 (8.9957/1.8804) mem 68106MB [2022-12-20 07:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][100/1519] eta 0:23:53 lr 0.000011 time 0.9203 (1.0100) model_time 0.9201 (1.0022) loss 1.0274 (0.8041) grad_norm 6.1391 (8.8290/1.8707) mem 68106MB [2022-12-20 07:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][110/1519] eta 0:23:41 lr 0.000011 time 0.9328 (1.0091) model_time 0.9327 (1.0019) loss 0.6791 (0.8050) grad_norm 10.7074 (8.9217/1.9042) mem 68106MB [2022-12-20 07:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][120/1519] eta 0:23:30 lr 0.000011 time 0.9221 (1.0082) model_time 0.9219 (1.0015) loss 0.8524 (0.8039) grad_norm 7.9023 (8.7806/1.8868) mem 68106MB [2022-12-20 07:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][130/1519] eta 0:23:20 lr 0.000011 time 0.9400 (1.0081) model_time 0.9398 (1.0019) loss 0.8705 (0.8086) grad_norm 7.1026 (8.7402/1.8501) mem 68106MB [2022-12-20 07:08:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][140/1519] eta 0:23:10 lr 0.000011 time 0.9288 (1.0080) model_time 0.9286 (1.0023) loss 0.8092 (0.8092) grad_norm 10.1577 (8.7345/1.8038) mem 68106MB [2022-12-20 07:08:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][150/1519] eta 0:22:59 lr 0.000011 time 0.9281 (1.0079) model_time 0.9280 (1.0025) loss 0.9750 (0.8113) grad_norm 11.5156 (8.7258/1.8315) mem 68106MB [2022-12-20 07:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][160/1519] eta 0:22:49 lr 0.000011 time 0.9265 (1.0079) model_time 0.9264 (1.0028) loss 1.0573 (0.8151) grad_norm 11.3184 (8.6977/1.8487) mem 68106MB [2022-12-20 07:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][170/1519] eta 0:22:39 lr 0.000011 time 0.9313 (1.0077) model_time 0.9312 (1.0029) loss 1.0678 (0.8229) grad_norm 9.2579 (8.7524/1.8833) mem 68106MB [2022-12-20 07:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][180/1519] eta 0:22:28 lr 0.000011 time 0.9205 (1.0072) model_time 0.9203 (1.0027) loss 0.7777 (0.8222) grad_norm 8.0119 (8.7194/1.9039) mem 68106MB [2022-12-20 07:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][190/1519] eta 0:22:17 lr 0.000011 time 0.9277 (1.0067) model_time 0.9274 (1.0024) loss 0.7388 (0.8232) grad_norm 9.6530 (8.6985/1.8822) mem 68106MB [2022-12-20 07:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][200/1519] eta 0:22:08 lr 0.000011 time 0.9229 (1.0069) model_time 0.9227 (1.0028) loss 1.1018 (0.8223) grad_norm 7.1421 (8.6213/1.9195) mem 68106MB [2022-12-20 07:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][210/1519] eta 0:21:57 lr 0.000011 time 0.9293 (1.0067) model_time 0.9291 (1.0028) loss 1.0586 (0.8195) grad_norm 9.0983 (8.5838/1.8907) mem 68106MB [2022-12-20 07:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][220/1519] eta 0:21:47 lr 0.000011 time 0.9242 (1.0068) model_time 0.9240 (1.0030) loss 0.7119 (0.8195) grad_norm 7.3662 (8.5273/1.8872) mem 68106MB [2022-12-20 07:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][230/1519] eta 0:21:37 lr 0.000011 time 0.9263 (1.0066) model_time 0.9262 (1.0030) loss 0.7561 (0.8189) grad_norm 10.9742 (8.5511/1.8851) mem 68106MB [2022-12-20 07:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][240/1519] eta 0:21:27 lr 0.000011 time 0.9224 (1.0064) model_time 0.9222 (1.0029) loss 0.9052 (0.8166) grad_norm 10.2322 (8.5618/1.8764) mem 68106MB [2022-12-20 07:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][250/1519] eta 0:21:17 lr 0.000011 time 0.9329 (1.0065) model_time 0.9328 (1.0031) loss 1.1950 (0.8156) grad_norm 8.3569 (8.5708/1.8595) mem 68106MB [2022-12-20 07:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][260/1519] eta 0:21:07 lr 0.000011 time 0.9273 (1.0064) model_time 0.9272 (1.0032) loss 0.7871 (0.8165) grad_norm 14.1542 (8.6855/2.0029) mem 68106MB [2022-12-20 07:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][270/1519] eta 0:20:57 lr 0.000011 time 0.9232 (1.0064) model_time 0.9231 (1.0033) loss 0.9124 (0.8177) grad_norm 7.7521 (8.6899/1.9710) mem 68106MB [2022-12-20 07:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][280/1519] eta 0:20:46 lr 0.000011 time 0.9325 (1.0064) model_time 0.9324 (1.0034) loss 0.8534 (0.8189) grad_norm 11.4261 (8.6884/1.9670) mem 68106MB [2022-12-20 07:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][290/1519] eta 0:20:37 lr 0.000011 time 0.9321 (1.0066) model_time 0.9320 (1.0036) loss 0.8438 (0.8192) grad_norm 7.8043 (8.6603/1.9427) mem 68106MB [2022-12-20 07:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][300/1519] eta 0:20:27 lr 0.000011 time 0.9221 (1.0066) model_time 0.9219 (1.0038) loss 0.9105 (0.8202) grad_norm 8.2512 (8.6472/1.9180) mem 68106MB [2022-12-20 07:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][310/1519] eta 0:20:16 lr 0.000011 time 0.9217 (1.0065) model_time 0.9216 (1.0037) loss 0.7959 (0.8198) grad_norm 8.9435 (8.6358/1.8993) mem 68106MB [2022-12-20 07:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][320/1519] eta 0:20:06 lr 0.000011 time 0.9345 (1.0064) model_time 0.9343 (1.0037) loss 0.7843 (0.8227) grad_norm 10.8224 (8.6595/1.9087) mem 68106MB [2022-12-20 07:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][330/1519] eta 0:19:56 lr 0.000011 time 0.9199 (1.0062) model_time 0.9198 (1.0036) loss 0.6753 (0.8220) grad_norm 8.6761 (8.6568/1.8890) mem 68106MB [2022-12-20 07:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][340/1519] eta 0:19:46 lr 0.000011 time 0.9239 (1.0060) model_time 0.9238 (1.0035) loss 0.9741 (0.8212) grad_norm 8.8097 (8.6843/1.8847) mem 68106MB [2022-12-20 07:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][350/1519] eta 0:19:35 lr 0.000011 time 0.9293 (1.0058) model_time 0.9292 (1.0033) loss 1.0469 (0.8222) grad_norm 8.2520 (8.6868/1.8692) mem 68106MB [2022-12-20 07:12:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][360/1519] eta 0:19:25 lr 0.000011 time 1.0180 (1.0059) model_time 1.0179 (1.0035) loss 0.7161 (0.8226) grad_norm 7.1408 (8.6604/1.8807) mem 68106MB [2022-12-20 07:12:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][370/1519] eta 0:19:15 lr 0.000011 time 0.9210 (1.0058) model_time 0.9209 (1.0034) loss 0.8922 (0.8241) grad_norm 9.1703 (8.6550/1.8575) mem 68106MB [2022-12-20 07:12:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][380/1519] eta 0:19:06 lr 0.000011 time 1.2100 (1.0068) model_time 1.2099 (1.0045) loss 1.3600 (0.8249) grad_norm 5.7087 (8.6386/1.8695) mem 68106MB [2022-12-20 07:12:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][390/1519] eta 0:18:56 lr 0.000011 time 0.9200 (1.0067) model_time 0.9198 (1.0045) loss 0.8046 (0.8247) grad_norm 8.9246 (8.6956/1.9014) mem 68106MB [2022-12-20 07:12:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][400/1519] eta 0:18:46 lr 0.000011 time 0.9238 (1.0067) model_time 0.9236 (1.0045) loss 0.6981 (0.8248) grad_norm 6.5132 (8.6446/1.9071) mem 68106MB [2022-12-20 07:13:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][410/1519] eta 0:18:36 lr 0.000011 time 0.9191 (1.0067) model_time 0.9190 (1.0045) loss 0.8521 (0.8243) grad_norm 7.1470 (8.6363/1.8964) mem 68106MB [2022-12-20 07:13:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][420/1519] eta 0:18:26 lr 0.000011 time 0.9197 (1.0065) model_time 0.9195 (1.0044) loss 0.6995 (0.8228) grad_norm 9.3415 (8.6562/1.9073) mem 68106MB [2022-12-20 07:13:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][430/1519] eta 0:18:15 lr 0.000011 time 0.9264 (1.0063) model_time 0.9262 (1.0042) loss 1.1117 (0.8239) grad_norm 6.7617 (8.6273/1.8979) mem 68106MB [2022-12-20 07:13:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][440/1519] eta 0:18:06 lr 0.000011 time 0.9244 (1.0067) model_time 0.9243 (1.0047) loss 0.7054 (0.8221) grad_norm 7.6845 (8.6051/1.8829) mem 68106MB [2022-12-20 07:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][450/1519] eta 0:17:56 lr 0.000011 time 0.9299 (1.0066) model_time 0.9297 (1.0046) loss 0.7483 (0.8214) grad_norm 7.4367 (8.6205/1.8700) mem 68106MB [2022-12-20 07:13:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][460/1519] eta 0:17:46 lr 0.000011 time 0.9912 (1.0068) model_time 0.9911 (1.0048) loss 0.6828 (0.8221) grad_norm 8.4247 (8.6213/1.8727) mem 68106MB [2022-12-20 07:14:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][470/1519] eta 0:17:36 lr 0.000011 time 0.9192 (1.0067) model_time 0.9190 (1.0048) loss 0.6724 (0.8229) grad_norm 9.3757 (8.6381/1.8629) mem 68106MB [2022-12-20 07:14:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][480/1519] eta 0:17:25 lr 0.000011 time 0.9294 (1.0066) model_time 0.9293 (1.0047) loss 0.9627 (0.8215) grad_norm 26.8599 (8.7173/2.2033) mem 68106MB [2022-12-20 07:14:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][490/1519] eta 0:17:15 lr 0.000011 time 0.9297 (1.0066) model_time 0.9296 (1.0047) loss 0.7416 (0.8224) grad_norm 7.4211 (8.7414/2.2177) mem 68106MB [2022-12-20 07:14:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][500/1519] eta 0:17:05 lr 0.000011 time 0.9228 (1.0064) model_time 0.9226 (1.0045) loss 0.8501 (0.8224) grad_norm 7.8971 (8.7475/2.2134) mem 68106MB [2022-12-20 07:14:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][510/1519] eta 0:16:55 lr 0.000011 time 0.9263 (1.0063) model_time 0.9262 (1.0045) loss 0.6952 (0.8218) grad_norm 12.2590 (8.7502/2.2132) mem 68106MB [2022-12-20 07:14:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][520/1519] eta 0:16:45 lr 0.000011 time 0.9333 (1.0062) model_time 0.9332 (1.0044) loss 0.9615 (0.8226) grad_norm 10.0112 (8.7479/2.1949) mem 68106MB [2022-12-20 07:15:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][530/1519] eta 0:16:35 lr 0.000011 time 0.9272 (1.0064) model_time 0.9271 (1.0046) loss 0.7401 (0.8232) grad_norm 6.5313 (8.7353/2.1801) mem 68106MB [2022-12-20 07:15:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][540/1519] eta 0:16:25 lr 0.000011 time 1.0141 (1.0065) model_time 1.0140 (1.0048) loss 0.6785 (0.8230) grad_norm 7.6132 (8.7352/2.1678) mem 68106MB [2022-12-20 07:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][550/1519] eta 0:16:15 lr 0.000011 time 0.9276 (1.0065) model_time 0.9274 (1.0048) loss 0.7363 (0.8223) grad_norm 11.9645 (8.7374/2.1708) mem 68106MB [2022-12-20 07:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][560/1519] eta 0:16:05 lr 0.000011 time 1.0026 (1.0066) model_time 1.0025 (1.0049) loss 1.0888 (0.8245) grad_norm 9.0429 (8.7294/2.1543) mem 68106MB [2022-12-20 07:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][570/1519] eta 0:15:55 lr 0.000011 time 0.8894 (1.0065) model_time 0.8892 (1.0049) loss 0.9695 (0.8248) grad_norm 6.7735 (8.7125/2.1434) mem 68106MB [2022-12-20 07:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][580/1519] eta 0:15:44 lr 0.000011 time 0.9252 (1.0064) model_time 0.9250 (1.0047) loss 0.6817 (0.8234) grad_norm 7.1443 (8.7009/2.1365) mem 68106MB [2022-12-20 07:16:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][590/1519] eta 0:15:34 lr 0.000011 time 0.9269 (1.0063) model_time 0.9267 (1.0047) loss 0.9993 (0.8240) grad_norm 7.9989 (8.7003/2.1220) mem 68106MB [2022-12-20 07:16:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][600/1519] eta 0:15:24 lr 0.000011 time 0.9292 (1.0064) model_time 0.9291 (1.0048) loss 0.7648 (0.8233) grad_norm 8.0999 (8.6854/2.1086) mem 68106MB [2022-12-20 07:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][610/1519] eta 0:15:14 lr 0.000011 time 0.9267 (1.0064) model_time 0.9266 (1.0048) loss 0.9331 (0.8228) grad_norm 6.9357 (8.6979/2.1661) mem 68106MB [2022-12-20 07:16:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][620/1519] eta 0:15:04 lr 0.000011 time 0.9238 (1.0063) model_time 0.9237 (1.0048) loss 0.6865 (0.8225) grad_norm 8.3020 (8.7184/2.1861) mem 68106MB [2022-12-20 07:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][630/1519] eta 0:14:54 lr 0.000011 time 0.9206 (1.0062) model_time 0.9205 (1.0047) loss 0.8230 (0.8221) grad_norm 8.8145 (8.7430/2.2101) mem 68106MB [2022-12-20 07:16:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][640/1519] eta 0:14:44 lr 0.000011 time 0.9963 (1.0062) model_time 0.9962 (1.0047) loss 0.8285 (0.8210) grad_norm 11.6169 (8.7422/2.2107) mem 68106MB [2022-12-20 07:17:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][650/1519] eta 0:14:34 lr 0.000011 time 0.9268 (1.0062) model_time 0.9267 (1.0047) loss 0.6706 (0.8214) grad_norm 9.7407 (8.7432/2.2068) mem 68106MB [2022-12-20 07:17:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][660/1519] eta 0:14:24 lr 0.000011 time 0.9256 (1.0061) model_time 0.9254 (1.0046) loss 1.0424 (0.8219) grad_norm 7.4198 (8.7232/2.1968) mem 68106MB [2022-12-20 07:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][670/1519] eta 0:14:14 lr 0.000011 time 0.9233 (1.0061) model_time 0.9232 (1.0046) loss 1.0439 (0.8234) grad_norm 7.3251 (8.6937/2.1927) mem 68106MB [2022-12-20 07:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][680/1519] eta 0:14:04 lr 0.000011 time 0.9227 (1.0060) model_time 0.9226 (1.0045) loss 1.0052 (0.8237) grad_norm 5.9128 (8.7068/2.2113) mem 68106MB [2022-12-20 07:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][690/1519] eta 0:13:53 lr 0.000011 time 0.9217 (1.0060) model_time 0.9216 (1.0046) loss 0.7554 (0.8232) grad_norm 6.5932 (8.7021/2.2109) mem 68106MB [2022-12-20 07:17:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][700/1519] eta 0:13:43 lr 0.000011 time 0.9218 (1.0059) model_time 0.9217 (1.0045) loss 0.8322 (0.8231) grad_norm 10.2438 (8.7119/2.2103) mem 68106MB [2022-12-20 07:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][710/1519] eta 0:13:33 lr 0.000011 time 0.9065 (1.0059) model_time 0.9063 (1.0045) loss 1.0058 (0.8241) grad_norm 9.2869 (8.6823/2.1982) mem 68106MB [2022-12-20 07:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][720/1519] eta 0:13:23 lr 0.000011 time 0.9253 (1.0062) model_time 0.9251 (1.0048) loss 0.8660 (0.8235) grad_norm 8.8463 (8.7152/2.2167) mem 68106MB [2022-12-20 07:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][730/1519] eta 0:13:13 lr 0.000011 time 0.9216 (1.0062) model_time 0.9212 (1.0048) loss 0.7177 (0.8227) grad_norm 17.4908 (8.7350/2.2732) mem 68106MB [2022-12-20 07:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][740/1519] eta 0:13:03 lr 0.000011 time 0.9189 (1.0061) model_time 0.9188 (1.0048) loss 0.6846 (0.8227) grad_norm 10.9088 (8.7418/2.2814) mem 68106MB [2022-12-20 07:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][750/1519] eta 0:12:53 lr 0.000011 time 1.0225 (1.0062) model_time 1.0223 (1.0049) loss 0.7714 (0.8234) grad_norm 8.3522 (8.7256/2.2694) mem 68106MB [2022-12-20 07:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][760/1519] eta 0:12:43 lr 0.000011 time 0.9269 (1.0062) model_time 0.9268 (1.0049) loss 0.7931 (0.8236) grad_norm 10.2471 (8.7554/2.2900) mem 68106MB [2022-12-20 07:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][770/1519] eta 0:12:33 lr 0.000011 time 0.9318 (1.0063) model_time 0.9316 (1.0050) loss 0.8514 (0.8232) grad_norm 8.1925 (8.7474/2.2722) mem 68106MB [2022-12-20 07:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][780/1519] eta 0:12:23 lr 0.000011 time 0.9229 (1.0062) model_time 0.9228 (1.0049) loss 0.6995 (0.8224) grad_norm 11.4188 (8.7882/2.2900) mem 68106MB [2022-12-20 07:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][790/1519] eta 0:12:13 lr 0.000011 time 0.9226 (1.0062) model_time 0.9225 (1.0049) loss 0.7509 (0.8223) grad_norm 8.3842 (8.7880/2.2943) mem 68106MB [2022-12-20 07:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][800/1519] eta 0:12:03 lr 0.000011 time 0.9310 (1.0062) model_time 0.9306 (1.0049) loss 0.8058 (0.8223) grad_norm 7.9772 (8.8114/2.2738) mem 68106MB [2022-12-20 07:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][810/1519] eta 0:11:53 lr 0.000011 time 0.9217 (1.0062) model_time 0.9216 (1.0049) loss 0.7351 (0.8228) grad_norm 10.7379 (8.8626/2.2939) mem 68106MB [2022-12-20 07:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][820/1519] eta 0:11:43 lr 0.000011 time 0.9213 (1.0061) model_time 0.9212 (1.0048) loss 0.6637 (0.8234) grad_norm 10.1303 (8.9039/2.3015) mem 68106MB [2022-12-20 07:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][830/1519] eta 0:11:33 lr 0.000011 time 0.9442 (1.0061) model_time 0.9441 (1.0048) loss 1.0184 (0.8241) grad_norm 7.5039 (8.8890/2.2974) mem 68106MB [2022-12-20 07:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][840/1519] eta 0:11:23 lr 0.000011 time 0.9255 (1.0060) model_time 0.9254 (1.0047) loss 0.9736 (0.8243) grad_norm 10.6581 (8.8855/2.2967) mem 68106MB [2022-12-20 07:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][850/1519] eta 0:11:12 lr 0.000011 time 0.9243 (1.0059) model_time 0.9242 (1.0047) loss 0.9432 (0.8245) grad_norm 9.0242 (8.8919/2.3154) mem 68106MB [2022-12-20 07:20:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][860/1519] eta 0:11:02 lr 0.000011 time 0.9309 (1.0060) model_time 0.9307 (1.0048) loss 0.7529 (0.8244) grad_norm 6.8523 (8.8338/2.2613) mem 68106MB [2022-12-20 07:20:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][870/1519] eta 0:10:52 lr 0.000011 time 0.9259 (1.0061) model_time 0.9257 (1.0049) loss 0.7851 (0.8235) grad_norm 9.7434 (8.8029/2.2800) mem 68106MB [2022-12-20 07:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][880/1519] eta 0:10:42 lr 0.000011 time 0.9241 (1.0061) model_time 0.9239 (1.0049) loss 0.9588 (0.8238) grad_norm 5.8174 (8.7994/2.2821) mem 68106MB [2022-12-20 07:21:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][890/1519] eta 0:10:32 lr 0.000011 time 0.9195 (1.0060) model_time 0.9194 (1.0048) loss 0.8096 (0.8238) grad_norm 9.5550 (8.8242/2.3112) mem 68106MB [2022-12-20 07:21:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][900/1519] eta 0:10:22 lr 0.000011 time 0.9301 (1.0060) model_time 0.9299 (1.0048) loss 0.9775 (0.8238) grad_norm 9.4857 (8.8335/2.3081) mem 68106MB [2022-12-20 07:21:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][910/1519] eta 0:10:12 lr 0.000011 time 0.9236 (1.0060) model_time 0.9235 (1.0049) loss 0.7544 (0.8242) grad_norm 10.7215 (8.8295/2.3125) mem 68106MB [2022-12-20 07:21:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][920/1519] eta 0:10:02 lr 0.000011 time 0.9363 (1.0060) model_time 0.9361 (1.0049) loss 0.8690 (0.8241) grad_norm 10.9101 (8.8147/2.3030) mem 68106MB [2022-12-20 07:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][930/1519] eta 0:09:52 lr 0.000011 time 0.9259 (1.0059) model_time 0.9258 (1.0048) loss 1.0424 (0.8253) grad_norm 7.8700 (8.7935/2.3078) mem 68106MB [2022-12-20 07:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][940/1519] eta 0:09:42 lr 0.000011 time 0.9234 (1.0059) model_time 0.9232 (1.0047) loss 0.7730 (0.8248) grad_norm 8.6155 (8.8030/2.3624) mem 68106MB [2022-12-20 07:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][950/1519] eta 0:09:32 lr 0.000011 time 0.9213 (1.0058) model_time 0.9212 (1.0047) loss 1.0134 (0.8257) grad_norm 11.1212 (8.8269/2.3843) mem 68106MB [2022-12-20 07:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][960/1519] eta 0:09:22 lr 0.000011 time 0.9234 (1.0058) model_time 0.9232 (1.0047) loss 0.6838 (0.8252) grad_norm 6.9218 (8.8445/2.3725) mem 68106MB [2022-12-20 07:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][970/1519] eta 0:09:12 lr 0.000011 time 0.9190 (1.0058) model_time 0.9189 (1.0047) loss 0.8422 (0.8257) grad_norm 6.6123 (8.8220/2.3818) mem 68106MB [2022-12-20 07:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][980/1519] eta 0:09:02 lr 0.000011 time 0.9287 (1.0057) model_time 0.9285 (1.0046) loss 0.9316 (0.8253) grad_norm 8.0228 (8.8207/2.3676) mem 68106MB [2022-12-20 07:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][990/1519] eta 0:08:51 lr 0.000011 time 0.9281 (1.0056) model_time 0.9279 (1.0045) loss 0.7833 (0.8253) grad_norm 8.9244 (8.7846/2.3442) mem 68106MB [2022-12-20 07:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1000/1519] eta 0:08:41 lr 0.000011 time 0.9048 (1.0056) model_time 0.9046 (1.0045) loss 1.1619 (0.8263) grad_norm 8.5435 (8.8439/2.3586) mem 68106MB [2022-12-20 07:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1010/1519] eta 0:08:31 lr 0.000011 time 0.9322 (1.0055) model_time 0.9320 (1.0045) loss 0.7113 (0.8262) grad_norm 7.2640 (8.8493/2.3534) mem 68106MB [2022-12-20 07:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1020/1519] eta 0:08:21 lr 0.000011 time 0.9061 (1.0055) model_time 0.9059 (1.0045) loss 0.8424 (0.8260) grad_norm 7.1448 (8.8352/2.3443) mem 68106MB [2022-12-20 07:23:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1030/1519] eta 0:08:11 lr 0.000011 time 0.9246 (1.0056) model_time 0.9244 (1.0045) loss 0.6856 (0.8258) grad_norm 9.1472 (8.8531/2.3535) mem 68106MB [2022-12-20 07:23:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1040/1519] eta 0:08:01 lr 0.000011 time 0.9849 (1.0055) model_time 0.9847 (1.0045) loss 0.8984 (0.8257) grad_norm 10.5092 (8.9063/2.3878) mem 68106MB [2022-12-20 07:23:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1050/1519] eta 0:07:51 lr 0.000011 time 0.9291 (1.0055) model_time 0.9285 (1.0045) loss 0.6643 (0.8252) grad_norm 10.5237 (8.8951/2.3881) mem 68106MB [2022-12-20 07:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1060/1519] eta 0:07:41 lr 0.000011 time 1.2066 (1.0057) model_time 1.2065 (1.0047) loss 0.7654 (0.8249) grad_norm 9.9994 (8.9108/2.3806) mem 68106MB [2022-12-20 07:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1070/1519] eta 0:07:31 lr 0.000011 time 1.1694 (1.0060) model_time 1.1693 (1.0050) loss 0.7648 (0.8250) grad_norm 7.2032 (8.9121/2.3998) mem 68106MB [2022-12-20 07:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1080/1519] eta 0:07:21 lr 0.000011 time 0.9157 (1.0061) model_time 0.9155 (1.0050) loss 0.7035 (0.8241) grad_norm 9.9538 (8.8569/2.1594) mem 68106MB [2022-12-20 07:24:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1090/1519] eta 0:07:11 lr 0.000011 time 0.9209 (1.0061) model_time 0.9207 (1.0051) loss 0.6849 (0.8234) grad_norm 9.5910 (8.8249/2.1383) mem 68106MB [2022-12-20 07:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1100/1519] eta 0:07:01 lr 0.000011 time 0.9244 (1.0061) model_time 0.9243 (1.0050) loss 0.7166 (0.8233) grad_norm 8.1933 (8.8242/2.1303) mem 68106MB [2022-12-20 07:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1110/1519] eta 0:06:51 lr 0.000011 time 0.9279 (1.0061) model_time 0.9277 (1.0051) loss 0.8808 (0.8237) grad_norm 7.2479 (8.8225/2.1308) mem 68106MB [2022-12-20 07:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1120/1519] eta 0:06:41 lr 0.000011 time 0.9221 (1.0061) model_time 0.9219 (1.0051) loss 0.8587 (0.8239) grad_norm 6.4370 (8.8121/2.1332) mem 68106MB [2022-12-20 07:25:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1130/1519] eta 0:06:31 lr 0.000011 time 0.9285 (1.0061) model_time 0.9284 (1.0051) loss 0.6984 (0.8238) grad_norm 9.9056 (8.8204/2.1347) mem 68106MB [2022-12-20 07:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1140/1519] eta 0:06:21 lr 0.000011 time 0.9301 (1.0060) model_time 0.9300 (1.0050) loss 0.6793 (0.8239) grad_norm 8.8135 (8.8115/2.1354) mem 68106MB [2022-12-20 07:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1150/1519] eta 0:06:11 lr 0.000011 time 0.9250 (1.0059) model_time 0.9248 (1.0049) loss 0.9479 (0.8242) grad_norm 6.3537 (8.8215/2.1460) mem 68106MB [2022-12-20 07:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1160/1519] eta 0:06:01 lr 0.000011 time 0.9317 (1.0058) model_time 0.9316 (1.0049) loss 0.7350 (0.8234) grad_norm 11.0963 (8.8437/2.1591) mem 68106MB [2022-12-20 07:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1170/1519] eta 0:05:51 lr 0.000011 time 0.9377 (1.0059) model_time 0.9376 (1.0050) loss 0.9050 (0.8229) grad_norm 6.1514 (8.8369/2.1669) mem 68106MB [2022-12-20 07:25:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1180/1519] eta 0:05:41 lr 0.000011 time 0.9559 (1.0059) model_time 0.9557 (1.0049) loss 0.7337 (0.8227) grad_norm 10.6740 (8.8326/2.1720) mem 68106MB [2022-12-20 07:26:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1190/1519] eta 0:05:30 lr 0.000011 time 0.9315 (1.0059) model_time 0.9313 (1.0049) loss 1.0151 (0.8224) grad_norm 8.6677 (8.8255/2.1774) mem 68106MB [2022-12-20 07:26:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1200/1519] eta 0:05:20 lr 0.000011 time 0.9444 (1.0058) model_time 0.9442 (1.0048) loss 0.6854 (0.8223) grad_norm 7.7236 (8.8375/2.1743) mem 68106MB [2022-12-20 07:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1210/1519] eta 0:05:10 lr 0.000011 time 0.9322 (1.0059) model_time 0.9321 (1.0050) loss 0.7089 (0.8227) grad_norm 9.9622 (8.8178/2.1095) mem 68106MB [2022-12-20 07:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1220/1519] eta 0:05:00 lr 0.000011 time 0.9324 (1.0059) model_time 0.9322 (1.0049) loss 0.6841 (0.8230) grad_norm 8.8302 (8.7813/2.0916) mem 68106MB [2022-12-20 07:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1230/1519] eta 0:04:50 lr 0.000011 time 0.9367 (1.0059) model_time 0.9366 (1.0049) loss 1.1470 (0.8234) grad_norm 6.7056 (8.7463/2.0725) mem 68106MB [2022-12-20 07:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1240/1519] eta 0:04:40 lr 0.000011 time 0.9796 (1.0059) model_time 0.9794 (1.0049) loss 0.9991 (0.8235) grad_norm 11.0277 (8.7084/2.0546) mem 68106MB [2022-12-20 07:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1250/1519] eta 0:04:30 lr 0.000011 time 0.9355 (1.0058) model_time 0.9353 (1.0049) loss 0.7717 (0.8238) grad_norm 8.3257 (8.6879/2.0604) mem 68106MB [2022-12-20 07:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1260/1519] eta 0:04:20 lr 0.000011 time 0.9324 (1.0058) model_time 0.9323 (1.0049) loss 0.6877 (0.8237) grad_norm 7.1594 (8.6926/2.0558) mem 68106MB [2022-12-20 07:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1270/1519] eta 0:04:10 lr 0.000011 time 0.9545 (1.0059) model_time 0.9543 (1.0049) loss 1.1085 (0.8240) grad_norm 11.9530 (8.7198/2.0523) mem 68106MB [2022-12-20 07:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1280/1519] eta 0:04:00 lr 0.000011 time 0.9278 (1.0058) model_time 0.9277 (1.0049) loss 0.6718 (0.8238) grad_norm 6.4949 (8.6976/2.0453) mem 68106MB [2022-12-20 07:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1290/1519] eta 0:03:50 lr 0.000011 time 0.9275 (1.0060) model_time 0.9270 (1.0051) loss 0.7161 (0.8237) grad_norm 8.1055 (8.6906/2.0406) mem 68106MB [2022-12-20 07:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1300/1519] eta 0:03:40 lr 0.000011 time 0.9228 (1.0060) model_time 0.9226 (1.0050) loss 0.6859 (0.8243) grad_norm 7.9820 (8.7032/2.0452) mem 68106MB [2022-12-20 07:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1310/1519] eta 0:03:30 lr 0.000011 time 0.9419 (1.0059) model_time 0.9416 (1.0050) loss 0.9952 (0.8241) grad_norm 17.2299 (8.7203/2.1103) mem 68106MB [2022-12-20 07:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1320/1519] eta 0:03:20 lr 0.000011 time 0.9254 (1.0059) model_time 0.9252 (1.0050) loss 0.7332 (0.8241) grad_norm 10.2069 (8.7022/2.0919) mem 68106MB [2022-12-20 07:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1330/1519] eta 0:03:10 lr 0.000011 time 0.9278 (1.0058) model_time 0.9277 (1.0049) loss 0.7605 (0.8239) grad_norm 10.5098 (8.7158/2.0538) mem 68106MB [2022-12-20 07:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1340/1519] eta 0:03:00 lr 0.000011 time 0.9321 (1.0058) model_time 0.9320 (1.0049) loss 0.9529 (0.8237) grad_norm 9.3236 (8.6820/2.0597) mem 68106MB [2022-12-20 07:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1350/1519] eta 0:02:49 lr 0.000011 time 0.9305 (1.0058) model_time 0.9303 (1.0049) loss 0.8223 (0.8239) grad_norm 10.5234 (8.6995/2.0652) mem 68106MB [2022-12-20 07:28:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1360/1519] eta 0:02:39 lr 0.000011 time 1.0888 (1.0059) model_time 1.0886 (1.0051) loss 0.6647 (0.8238) grad_norm 6.7102 (8.6716/2.0272) mem 68106MB [2022-12-20 07:29:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1370/1519] eta 0:02:29 lr 0.000011 time 0.9262 (1.0059) model_time 0.9260 (1.0050) loss 0.8921 (0.8240) grad_norm 8.0132 (8.6817/2.0610) mem 68106MB [2022-12-20 07:29:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1380/1519] eta 0:02:19 lr 0.000011 time 0.9293 (1.0059) model_time 0.9291 (1.0050) loss 0.6850 (0.8236) grad_norm 12.9038 (8.6520/2.0369) mem 68106MB [2022-12-20 07:29:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1390/1519] eta 0:02:09 lr 0.000011 time 0.9252 (1.0059) model_time 0.9250 (1.0050) loss 0.7947 (0.8238) grad_norm 8.3641 (8.6542/2.0273) mem 68106MB [2022-12-20 07:29:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1400/1519] eta 0:01:59 lr 0.000011 time 0.9335 (1.0059) model_time 0.9333 (1.0050) loss 0.7283 (0.8238) grad_norm 11.3328 (8.6662/2.0410) mem 68106MB [2022-12-20 07:29:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1410/1519] eta 0:01:49 lr 0.000011 time 0.9312 (1.0059) model_time 0.9311 (1.0051) loss 0.8715 (0.8240) grad_norm 8.5043 (8.6280/2.0161) mem 68106MB [2022-12-20 07:29:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1420/1519] eta 0:01:39 lr 0.000011 time 0.9279 (1.0059) model_time 0.9277 (1.0050) loss 0.7348 (0.8243) grad_norm 5.9836 (8.6007/2.0001) mem 68106MB [2022-12-20 07:30:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1430/1519] eta 0:01:29 lr 0.000011 time 0.9261 (1.0059) model_time 0.9259 (1.0050) loss 0.7455 (0.8242) grad_norm 8.4093 (8.6218/2.0155) mem 68106MB [2022-12-20 07:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1440/1519] eta 0:01:19 lr 0.000011 time 0.9336 (1.0058) model_time 0.9334 (1.0050) loss 0.6969 (0.8245) grad_norm 8.4785 (8.6379/2.0205) mem 68106MB [2022-12-20 07:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1450/1519] eta 0:01:09 lr 0.000011 time 0.9421 (1.0058) model_time 0.9419 (1.0049) loss 0.7471 (0.8242) grad_norm 6.8506 (8.6119/1.9969) mem 68106MB [2022-12-20 07:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1460/1519] eta 0:00:59 lr 0.000011 time 0.9381 (1.0058) model_time 0.9379 (1.0049) loss 0.7866 (0.8237) grad_norm 7.8474 (8.6032/2.0105) mem 68106MB [2022-12-20 07:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1470/1519] eta 0:00:49 lr 0.000011 time 0.9306 (1.0057) model_time 0.9304 (1.0048) loss 0.7877 (0.8237) grad_norm 10.7734 (8.6561/2.0030) mem 68106MB [2022-12-20 07:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1480/1519] eta 0:00:39 lr 0.000011 time 0.9302 (1.0057) model_time 0.9300 (1.0049) loss 1.0006 (0.8236) grad_norm 8.7933 (8.6756/2.0052) mem 68106MB [2022-12-20 07:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1490/1519] eta 0:00:29 lr 0.000011 time 0.9309 (1.0057) model_time 0.9305 (1.0048) loss 0.9507 (0.8236) grad_norm 8.0493 (8.6750/1.9777) mem 68106MB [2022-12-20 07:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1500/1519] eta 0:00:19 lr 0.000011 time 0.9309 (1.0057) model_time 0.9307 (1.0049) loss 0.6784 (0.8233) grad_norm 6.4831 (8.6540/1.9859) mem 68106MB [2022-12-20 07:31:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [61/100][1510/1519] eta 0:00:09 lr 0.000011 time 0.9211 (1.0057) model_time 0.9210 (1.0049) loss 0.9015 (0.8234) grad_norm 8.1417 (8.6355/1.9953) mem 68106MB [2022-12-20 07:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 61 training takes 0:25:27 [2022-12-20 07:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_61.pth saving...... [2022-12-20 07:31:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_61.pth saved !!! [2022-12-20 07:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.652 (0.652) Loss 0.5142 (0.5142) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 07:32:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.329) Loss 0.5221 (0.4982) Acc@1 92.708 (92.582) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 07:32:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4823 (0.4966) Acc@1 91.667 (92.626) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 07:32:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 0.6121 (0.5016) Acc@1 90.625 (92.428) Acc@5 97.917 (98.398) Mem 68106MB [2022-12-20 07:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.305 (0.307) Loss 0.4546 (0.4921) Acc@1 93.750 (92.480) Acc@5 99.306 (98.526) Mem 68106MB [2022-12-20 07:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 0.4729 (0.4894) Acc@1 90.972 (92.477) Acc@5 99.653 (98.591) Mem 68106MB [2022-12-20 07:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.304) Loss 0.5124 (0.4896) Acc@1 91.319 (92.424) Acc@5 98.264 (98.577) Mem 68106MB [2022-12-20 07:32:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.297 (0.303) Loss 0.5364 (0.4909) Acc@1 92.014 (92.395) Acc@5 97.917 (98.552) Mem 68106MB [2022-12-20 07:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.301 (0.302) Loss 0.4235 (0.4888) Acc@1 93.403 (92.417) Acc@5 98.264 (98.594) Mem 68106MB [2022-12-20 07:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:61] * Acc@1 92.371 Acc@5 98.592 [2022-12-20 07:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 07:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 07:32:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 07:32:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.37% [2022-12-20 07:32:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][0/1519] eta 0:36:25 lr 0.000011 time 1.4385 (1.4385) model_time 0.9925 (0.9925) loss 0.8534 (0.8534) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 07:33:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][10/1519] eta 0:26:05 lr 0.000011 time 0.9228 (1.0375) model_time 0.9227 (0.9966) loss 0.6879 (0.8776) grad_norm 6.9563 (7.5637/0.6447) mem 68106MB [2022-12-20 07:33:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][20/1519] eta 0:25:31 lr 0.000011 time 0.9080 (1.0217) model_time 0.9079 (1.0002) loss 0.8181 (0.8472) grad_norm 9.3234 (8.1418/0.9374) mem 68106MB [2022-12-20 07:33:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][30/1519] eta 0:25:12 lr 0.000011 time 0.9336 (1.0159) model_time 0.9335 (1.0012) loss 0.7655 (0.8454) grad_norm 8.9336 (8.3209/0.9947) mem 68106MB [2022-12-20 07:33:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][40/1519] eta 0:25:00 lr 0.000011 time 0.9229 (1.0142) model_time 0.9228 (1.0031) loss 0.8622 (0.8256) grad_norm 5.7190 (8.1763/1.2525) mem 68106MB [2022-12-20 07:33:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][50/1519] eta 0:24:47 lr 0.000011 time 0.9252 (1.0128) model_time 0.9251 (1.0037) loss 0.7275 (0.8142) grad_norm 7.9574 (8.3093/1.4826) mem 68106MB [2022-12-20 07:33:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][60/1519] eta 0:24:36 lr 0.000011 time 0.9340 (1.0120) model_time 0.9339 (1.0043) loss 1.1250 (0.8297) grad_norm 8.9883 (8.2541/1.4011) mem 68106MB [2022-12-20 07:34:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][70/1519] eta 0:24:25 lr 0.000011 time 1.0107 (1.0115) model_time 1.0105 (1.0049) loss 0.8203 (0.8326) grad_norm 8.3542 (8.4216/1.3777) mem 68106MB [2022-12-20 07:34:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][80/1519] eta 0:24:14 lr 0.000011 time 0.9783 (1.0110) model_time 0.9781 (1.0051) loss 0.7914 (0.8324) grad_norm 13.8678 (8.5203/1.5656) mem 68106MB [2022-12-20 07:34:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][90/1519] eta 0:24:03 lr 0.000011 time 0.9238 (1.0104) model_time 0.9237 (1.0052) loss 1.0610 (0.8401) grad_norm 6.5230 (8.4915/1.5610) mem 68106MB [2022-12-20 07:34:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][100/1519] eta 0:23:52 lr 0.000011 time 0.9292 (1.0092) model_time 0.9290 (1.0045) loss 0.8347 (0.8355) grad_norm 7.8531 (8.4458/1.5131) mem 68106MB [2022-12-20 07:34:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][110/1519] eta 0:23:44 lr 0.000011 time 0.9292 (1.0113) model_time 0.9290 (1.0070) loss 0.7343 (0.8382) grad_norm 8.6355 (8.5534/1.5399) mem 68106MB [2022-12-20 07:34:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][120/1519] eta 0:23:34 lr 0.000011 time 0.9307 (1.0108) model_time 0.9305 (1.0068) loss 0.8672 (0.8341) grad_norm 7.5628 (8.4692/1.5148) mem 68106MB [2022-12-20 07:35:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][130/1519] eta 0:23:23 lr 0.000011 time 0.9192 (1.0103) model_time 0.9190 (1.0066) loss 0.6779 (0.8356) grad_norm 9.9845 (8.5174/1.5371) mem 68106MB [2022-12-20 07:35:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][140/1519] eta 0:23:12 lr 0.000011 time 0.9284 (1.0098) model_time 0.9280 (1.0063) loss 0.6745 (0.8298) grad_norm 7.7750 (8.4829/1.4904) mem 68106MB [2022-12-20 07:35:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][150/1519] eta 0:23:01 lr 0.000011 time 0.9282 (1.0091) model_time 0.9280 (1.0058) loss 0.7781 (0.8295) grad_norm 8.8963 (8.5204/1.5404) mem 68106MB [2022-12-20 07:35:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][160/1519] eta 0:22:50 lr 0.000011 time 0.9187 (1.0085) model_time 0.9186 (1.0054) loss 0.9543 (0.8302) grad_norm 6.6651 (8.5429/1.5628) mem 68106MB [2022-12-20 07:35:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][170/1519] eta 0:22:39 lr 0.000011 time 0.9219 (1.0080) model_time 0.9217 (1.0050) loss 0.8990 (0.8288) grad_norm 8.8780 (8.5073/1.6288) mem 68106MB [2022-12-20 07:35:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][180/1519] eta 0:22:29 lr 0.000011 time 0.9804 (1.0078) model_time 0.9803 (1.0050) loss 0.7336 (0.8308) grad_norm 9.2257 (8.4853/1.5982) mem 68106MB [2022-12-20 07:36:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][190/1519] eta 0:22:19 lr 0.000011 time 0.9536 (1.0076) model_time 0.9535 (1.0049) loss 0.8465 (0.8320) grad_norm 7.2979 (8.3885/1.6129) mem 68106MB [2022-12-20 07:36:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][200/1519] eta 0:22:09 lr 0.000011 time 0.9253 (1.0078) model_time 0.9252 (1.0052) loss 0.6958 (0.8273) grad_norm 6.9067 (8.4266/1.6479) mem 68106MB [2022-12-20 07:36:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][210/1519] eta 0:21:58 lr 0.000011 time 0.9223 (1.0075) model_time 0.9221 (1.0050) loss 0.8223 (0.8243) grad_norm 13.4518 (8.4457/1.7017) mem 68106MB [2022-12-20 07:36:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][220/1519] eta 0:21:48 lr 0.000011 time 0.9287 (1.0072) model_time 0.9286 (1.0048) loss 0.7811 (0.8250) grad_norm 6.8440 (8.4331/1.6809) mem 68106MB [2022-12-20 07:36:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][230/1519] eta 0:21:37 lr 0.000011 time 0.9287 (1.0069) model_time 0.9286 (1.0047) loss 1.1991 (0.8285) grad_norm 8.0702 (8.4330/1.6808) mem 68106MB [2022-12-20 07:36:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][240/1519] eta 0:21:27 lr 0.000011 time 0.9250 (1.0067) model_time 0.9249 (1.0045) loss 0.7542 (0.8284) grad_norm 9.5237 (8.5536/2.1309) mem 68106MB [2022-12-20 07:37:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][250/1519] eta 0:21:17 lr 0.000011 time 0.9216 (1.0071) model_time 0.9214 (1.0050) loss 0.9409 (0.8272) grad_norm 7.7196 (8.5271/2.1046) mem 68106MB [2022-12-20 07:37:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][260/1519] eta 0:21:07 lr 0.000011 time 0.9212 (1.0068) model_time 0.9210 (1.0047) loss 0.7724 (0.8257) grad_norm 9.3075 (8.4781/2.0943) mem 68106MB [2022-12-20 07:37:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][270/1519] eta 0:20:57 lr 0.000011 time 0.9262 (1.0067) model_time 0.9260 (1.0047) loss 1.1150 (0.8276) grad_norm 8.9983 (8.4852/2.0754) mem 68106MB [2022-12-20 07:37:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][280/1519] eta 0:20:47 lr 0.000011 time 0.9201 (1.0065) model_time 0.9200 (1.0046) loss 0.8377 (0.8305) grad_norm 8.0223 (8.5230/2.0903) mem 68106MB [2022-12-20 07:37:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][290/1519] eta 0:20:36 lr 0.000011 time 0.9292 (1.0063) model_time 0.9291 (1.0044) loss 0.7832 (0.8309) grad_norm 7.3792 (8.4849/2.0719) mem 68106MB [2022-12-20 07:37:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][300/1519] eta 0:20:26 lr 0.000011 time 0.9175 (1.0061) model_time 0.9174 (1.0042) loss 0.7150 (0.8298) grad_norm 9.2198 (8.4945/2.0612) mem 68106MB [2022-12-20 07:38:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][310/1519] eta 0:20:16 lr 0.000011 time 0.9267 (1.0059) model_time 0.9266 (1.0041) loss 0.8919 (0.8304) grad_norm 9.4398 (8.4739/2.0376) mem 68106MB [2022-12-20 07:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][320/1519] eta 0:20:05 lr 0.000011 time 0.9255 (1.0056) model_time 0.9254 (1.0039) loss 0.7176 (0.8281) grad_norm 11.0447 (8.4760/2.0273) mem 68106MB [2022-12-20 07:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][330/1519] eta 0:19:55 lr 0.000011 time 0.9225 (1.0055) model_time 0.9224 (1.0038) loss 0.8108 (0.8273) grad_norm 12.1026 (8.4646/2.0288) mem 68106MB [2022-12-20 07:38:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][340/1519] eta 0:19:45 lr 0.000011 time 0.9188 (1.0053) model_time 0.9187 (1.0037) loss 0.9031 (0.8269) grad_norm 9.3358 (8.4506/2.0102) mem 68106MB [2022-12-20 07:38:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][350/1519] eta 0:19:35 lr 0.000011 time 0.9276 (1.0057) model_time 0.9274 (1.0041) loss 0.7888 (0.8264) grad_norm 9.3702 (8.4326/1.9911) mem 68106MB [2022-12-20 07:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][360/1519] eta 0:19:25 lr 0.000011 time 0.9260 (1.0056) model_time 0.9259 (1.0041) loss 0.7160 (0.8269) grad_norm 7.2642 (8.4271/1.9941) mem 68106MB [2022-12-20 07:39:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][370/1519] eta 0:19:15 lr 0.000011 time 0.9031 (1.0056) model_time 0.9029 (1.0041) loss 0.6996 (0.8262) grad_norm 10.8442 (8.4859/2.0384) mem 68106MB [2022-12-20 07:39:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][380/1519] eta 0:19:05 lr 0.000011 time 0.9244 (1.0058) model_time 0.9242 (1.0043) loss 1.1126 (0.8270) grad_norm 17.5884 (8.5553/2.1554) mem 68106MB [2022-12-20 07:39:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][390/1519] eta 0:18:55 lr 0.000011 time 0.9337 (1.0060) model_time 0.9336 (1.0046) loss 0.8690 (0.8265) grad_norm 8.2341 (8.6094/2.1916) mem 68106MB [2022-12-20 07:39:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][400/1519] eta 0:18:46 lr 0.000011 time 0.9138 (1.0069) model_time 0.9136 (1.0054) loss 1.0113 (0.8284) grad_norm 11.7872 (8.6312/2.1860) mem 68106MB [2022-12-20 07:39:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][410/1519] eta 0:18:36 lr 0.000011 time 0.9228 (1.0067) model_time 0.9227 (1.0053) loss 0.7630 (0.8289) grad_norm 7.9042 (8.6249/2.1707) mem 68106MB [2022-12-20 07:39:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][420/1519] eta 0:18:26 lr 0.000011 time 0.9110 (1.0068) model_time 0.9109 (1.0055) loss 0.7334 (0.8296) grad_norm 7.0127 (8.6039/2.1590) mem 68106MB [2022-12-20 07:40:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][430/1519] eta 0:18:16 lr 0.000011 time 0.9084 (1.0068) model_time 0.9083 (1.0054) loss 0.6953 (0.8287) grad_norm 10.3040 (8.5982/2.1396) mem 68106MB [2022-12-20 07:40:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][440/1519] eta 0:18:06 lr 0.000011 time 0.9134 (1.0068) model_time 0.9133 (1.0054) loss 0.7083 (0.8312) grad_norm 7.6445 (8.6029/2.1514) mem 68106MB [2022-12-20 07:40:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][450/1519] eta 0:17:56 lr 0.000011 time 0.9731 (1.0068) model_time 0.9730 (1.0055) loss 1.2121 (0.8322) grad_norm 6.4112 (8.5913/2.1355) mem 68106MB [2022-12-20 07:40:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][460/1519] eta 0:17:45 lr 0.000011 time 0.9250 (1.0066) model_time 0.9248 (1.0053) loss 0.7633 (0.8308) grad_norm 6.1565 (8.6012/2.1548) mem 68106MB [2022-12-20 07:40:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][470/1519] eta 0:17:35 lr 0.000011 time 0.9183 (1.0064) model_time 0.9181 (1.0051) loss 0.7102 (0.8304) grad_norm 9.4509 (8.5935/2.1398) mem 68106MB [2022-12-20 07:40:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][480/1519] eta 0:17:25 lr 0.000011 time 0.9319 (1.0062) model_time 0.9318 (1.0050) loss 0.7679 (0.8298) grad_norm 10.3300 (8.6053/2.1379) mem 68106MB [2022-12-20 07:41:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][490/1519] eta 0:17:15 lr 0.000011 time 0.9279 (1.0061) model_time 0.9278 (1.0049) loss 0.9533 (0.8300) grad_norm 7.8813 (8.6116/2.1302) mem 68106MB [2022-12-20 07:41:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][500/1519] eta 0:17:05 lr 0.000011 time 0.9258 (1.0062) model_time 0.9256 (1.0049) loss 0.7237 (0.8303) grad_norm 10.7282 (8.6240/2.1252) mem 68106MB [2022-12-20 07:41:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][510/1519] eta 0:16:55 lr 0.000011 time 0.9297 (1.0060) model_time 0.9296 (1.0048) loss 1.0548 (0.8318) grad_norm 9.0920 (8.6241/2.1064) mem 68106MB [2022-12-20 07:41:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][520/1519] eta 0:16:44 lr 0.000011 time 0.9223 (1.0058) model_time 0.9221 (1.0046) loss 0.7236 (0.8315) grad_norm 9.9109 (8.6222/2.0912) mem 68106MB [2022-12-20 07:41:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][530/1519] eta 0:16:34 lr 0.000011 time 0.9349 (1.0058) model_time 0.9347 (1.0046) loss 0.6988 (0.8303) grad_norm 12.6456 (8.6414/2.1203) mem 68106MB [2022-12-20 07:41:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][540/1519] eta 0:16:24 lr 0.000011 time 0.9206 (1.0058) model_time 0.9204 (1.0047) loss 0.7048 (0.8308) grad_norm 8.2817 (8.6628/2.1403) mem 68106MB [2022-12-20 07:42:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][550/1519] eta 0:16:14 lr 0.000011 time 0.9221 (1.0057) model_time 0.9220 (1.0045) loss 0.7476 (0.8296) grad_norm 11.5195 (8.6731/2.1309) mem 68106MB [2022-12-20 07:42:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][560/1519] eta 0:16:04 lr 0.000011 time 0.8870 (1.0059) model_time 0.8869 (1.0048) loss 0.7779 (0.8289) grad_norm 5.5529 (8.6545/2.1250) mem 68106MB [2022-12-20 07:42:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][570/1519] eta 0:15:54 lr 0.000011 time 0.9258 (1.0057) model_time 0.9257 (1.0046) loss 0.6910 (0.8294) grad_norm 10.7814 (8.6539/2.1169) mem 68106MB [2022-12-20 07:42:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][580/1519] eta 0:15:44 lr 0.000011 time 0.9587 (1.0057) model_time 0.9586 (1.0046) loss 1.0968 (0.8301) grad_norm 6.5686 (8.6227/2.1156) mem 68106MB [2022-12-20 07:42:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][590/1519] eta 0:15:34 lr 0.000011 time 0.9218 (1.0056) model_time 0.9216 (1.0045) loss 1.2643 (0.8313) grad_norm 9.9365 (8.6251/2.0999) mem 68106MB [2022-12-20 07:42:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][600/1519] eta 0:15:24 lr 0.000011 time 0.9272 (1.0055) model_time 0.9271 (1.0045) loss 0.7550 (0.8309) grad_norm 8.1661 (8.6329/2.0853) mem 68106MB [2022-12-20 07:43:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][610/1519] eta 0:15:13 lr 0.000011 time 0.9258 (1.0054) model_time 0.9257 (1.0043) loss 0.6875 (0.8303) grad_norm 7.4937 (8.6569/2.0918) mem 68106MB [2022-12-20 07:43:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][620/1519] eta 0:15:03 lr 0.000011 time 0.9316 (1.0054) model_time 0.9315 (1.0043) loss 0.9411 (0.8318) grad_norm 9.8643 (8.6486/2.0929) mem 68106MB [2022-12-20 07:43:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][630/1519] eta 0:14:53 lr 0.000011 time 0.9774 (1.0053) model_time 0.9772 (1.0043) loss 0.6749 (0.8315) grad_norm 12.0257 (8.6395/2.1081) mem 68106MB [2022-12-20 07:43:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][640/1519] eta 0:14:43 lr 0.000011 time 0.9283 (1.0053) model_time 0.9282 (1.0043) loss 0.6935 (0.8304) grad_norm 8.7223 (8.6620/2.1372) mem 68106MB [2022-12-20 07:43:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][650/1519] eta 0:14:33 lr 0.000011 time 0.9223 (1.0052) model_time 0.9221 (1.0042) loss 0.8079 (0.8296) grad_norm 9.2684 (8.6529/2.1229) mem 68106MB [2022-12-20 07:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][660/1519] eta 0:14:23 lr 0.000011 time 1.0226 (1.0053) model_time 1.0225 (1.0043) loss 0.7036 (0.8289) grad_norm 8.2394 (8.6358/2.1334) mem 68106MB [2022-12-20 07:44:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][670/1519] eta 0:14:13 lr 0.000011 time 0.9288 (1.0052) model_time 0.9286 (1.0042) loss 0.8871 (0.8302) grad_norm 8.8435 (8.6157/2.1318) mem 68106MB [2022-12-20 07:44:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][680/1519] eta 0:14:03 lr 0.000011 time 0.9231 (1.0051) model_time 0.9230 (1.0041) loss 0.9110 (0.8305) grad_norm 9.0349 (8.6086/2.1230) mem 68106MB [2022-12-20 07:44:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][690/1519] eta 0:13:53 lr 0.000011 time 0.9227 (1.0053) model_time 0.9226 (1.0043) loss 1.0332 (0.8302) grad_norm 10.1712 (8.6196/2.1213) mem 68106MB [2022-12-20 07:44:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][700/1519] eta 0:13:43 lr 0.000011 time 0.9201 (1.0053) model_time 0.9200 (1.0044) loss 0.7801 (0.8302) grad_norm 5.9523 (8.6119/2.1293) mem 68106MB [2022-12-20 07:44:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][710/1519] eta 0:13:33 lr 0.000011 time 0.9230 (1.0052) model_time 0.9228 (1.0043) loss 0.8594 (0.8304) grad_norm 5.7218 (8.5916/2.1616) mem 68106MB [2022-12-20 07:44:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][720/1519] eta 0:13:23 lr 0.000011 time 0.9148 (1.0052) model_time 0.9146 (1.0043) loss 0.7473 (0.8305) grad_norm 6.6128 (8.6466/2.2092) mem 68106MB [2022-12-20 07:45:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][730/1519] eta 0:13:13 lr 0.000011 time 0.9806 (1.0052) model_time 0.9804 (1.0043) loss 0.6680 (0.8305) grad_norm 6.4594 (8.6095/2.2132) mem 68106MB [2022-12-20 07:45:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][740/1519] eta 0:13:03 lr 0.000011 time 0.9406 (1.0053) model_time 0.9405 (1.0044) loss 0.8274 (0.8306) grad_norm 11.5655 (8.6386/2.2293) mem 68106MB [2022-12-20 07:45:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][750/1519] eta 0:12:53 lr 0.000011 time 0.9294 (1.0053) model_time 0.9292 (1.0044) loss 0.9341 (0.8304) grad_norm 8.9452 (8.6435/2.2165) mem 68106MB [2022-12-20 07:45:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][760/1519] eta 0:12:43 lr 0.000011 time 0.9341 (1.0053) model_time 0.9339 (1.0044) loss 0.7744 (0.8302) grad_norm 7.5431 (8.6424/2.2123) mem 68106MB [2022-12-20 07:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][770/1519] eta 0:12:32 lr 0.000011 time 0.9267 (1.0052) model_time 0.9266 (1.0043) loss 0.6863 (0.8293) grad_norm 6.6103 (8.6323/2.1965) mem 68106MB [2022-12-20 07:45:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][780/1519] eta 0:12:22 lr 0.000011 time 0.9278 (1.0052) model_time 0.9277 (1.0043) loss 0.8669 (0.8291) grad_norm 8.6301 (8.6432/2.2139) mem 68106MB [2022-12-20 07:46:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][790/1519] eta 0:12:12 lr 0.000011 time 0.9230 (1.0051) model_time 0.9228 (1.0042) loss 0.7243 (0.8287) grad_norm 6.6978 (8.6547/2.2079) mem 68106MB [2022-12-20 07:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][800/1519] eta 0:12:02 lr 0.000011 time 0.9158 (1.0050) model_time 0.9157 (1.0041) loss 0.7659 (0.8285) grad_norm 6.3611 (8.6348/2.2018) mem 68106MB [2022-12-20 07:46:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][810/1519] eta 0:11:52 lr 0.000011 time 0.9211 (1.0050) model_time 0.9209 (1.0041) loss 0.8381 (0.8285) grad_norm 10.6358 (8.6595/2.2295) mem 68106MB [2022-12-20 07:46:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][820/1519] eta 0:11:42 lr 0.000011 time 0.9189 (1.0049) model_time 0.9188 (1.0041) loss 0.7469 (0.8291) grad_norm 7.8667 (8.6514/2.2297) mem 68106MB [2022-12-20 07:46:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][830/1519] eta 0:11:32 lr 0.000011 time 0.9339 (1.0049) model_time 0.9337 (1.0041) loss 0.8152 (0.8284) grad_norm 7.0585 (8.6510/2.2289) mem 68106MB [2022-12-20 07:46:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][840/1519] eta 0:11:22 lr 0.000011 time 0.9037 (1.0049) model_time 0.9036 (1.0041) loss 0.7636 (0.8280) grad_norm 6.6381 (8.5887/2.0667) mem 68106MB [2022-12-20 07:47:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][850/1519] eta 0:11:12 lr 0.000011 time 0.9138 (1.0049) model_time 0.9136 (1.0041) loss 1.0458 (0.8285) grad_norm 9.2237 (8.5821/2.0750) mem 68106MB [2022-12-20 07:47:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][860/1519] eta 0:11:02 lr 0.000011 time 0.9250 (1.0049) model_time 0.9249 (1.0040) loss 0.8826 (0.8289) grad_norm 7.6822 (8.6019/2.0694) mem 68106MB [2022-12-20 07:47:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][870/1519] eta 0:10:52 lr 0.000011 time 0.9096 (1.0050) model_time 0.9095 (1.0042) loss 0.6969 (0.8286) grad_norm 8.6302 (8.6137/2.0739) mem 68106MB [2022-12-20 07:47:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][880/1519] eta 0:10:42 lr 0.000011 time 0.9243 (1.0051) model_time 0.9242 (1.0042) loss 0.8717 (0.8279) grad_norm 8.2734 (8.5951/2.0557) mem 68106MB [2022-12-20 07:47:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][890/1519] eta 0:10:32 lr 0.000011 time 0.9270 (1.0051) model_time 0.9269 (1.0042) loss 0.8723 (0.8282) grad_norm 6.6323 (8.6114/2.0507) mem 68106MB [2022-12-20 07:47:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][900/1519] eta 0:10:22 lr 0.000011 time 0.9221 (1.0051) model_time 0.9220 (1.0043) loss 1.1402 (0.8289) grad_norm 8.0381 (8.5867/2.0472) mem 68106MB [2022-12-20 07:48:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][910/1519] eta 0:10:12 lr 0.000011 time 1.0222 (1.0052) model_time 1.0221 (1.0043) loss 0.8563 (0.8291) grad_norm 8.0999 (8.6005/2.0442) mem 68106MB [2022-12-20 07:48:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][920/1519] eta 0:10:02 lr 0.000011 time 0.9302 (1.0051) model_time 0.9301 (1.0043) loss 1.1151 (0.8302) grad_norm 7.8983 (8.6211/2.0496) mem 68106MB [2022-12-20 07:48:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][930/1519] eta 0:09:51 lr 0.000011 time 0.9353 (1.0051) model_time 0.9351 (1.0042) loss 0.8771 (0.8297) grad_norm 8.0889 (8.6099/2.0392) mem 68106MB [2022-12-20 07:48:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][940/1519] eta 0:09:41 lr 0.000011 time 0.9286 (1.0050) model_time 0.9284 (1.0042) loss 0.9060 (0.8297) grad_norm 9.1964 (8.6171/2.0354) mem 68106MB [2022-12-20 07:48:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][950/1519] eta 0:09:31 lr 0.000011 time 0.9289 (1.0050) model_time 0.9288 (1.0042) loss 0.9809 (0.8296) grad_norm 7.8623 (8.6190/2.0315) mem 68106MB [2022-12-20 07:48:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][960/1519] eta 0:09:21 lr 0.000011 time 0.9271 (1.0050) model_time 0.9269 (1.0042) loss 0.6951 (0.8286) grad_norm 8.3535 (8.6311/2.0226) mem 68106MB [2022-12-20 07:49:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][970/1519] eta 0:09:11 lr 0.000011 time 0.9757 (1.0050) model_time 0.9756 (1.0042) loss 0.9901 (0.8279) grad_norm 7.5231 (8.6127/2.0176) mem 68106MB [2022-12-20 07:49:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][980/1519] eta 0:09:01 lr 0.000011 time 0.9252 (1.0050) model_time 0.9251 (1.0042) loss 0.7120 (0.8275) grad_norm 7.9319 (8.5581/1.9286) mem 68106MB [2022-12-20 07:49:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][990/1519] eta 0:08:51 lr 0.000011 time 0.9355 (1.0049) model_time 0.9353 (1.0041) loss 0.8691 (0.8273) grad_norm 9.5448 (8.5031/1.8929) mem 68106MB [2022-12-20 07:49:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1000/1519] eta 0:08:41 lr 0.000011 time 1.0444 (1.0050) model_time 1.0443 (1.0042) loss 0.7092 (0.8267) grad_norm 11.5700 (8.4852/1.8878) mem 68106MB [2022-12-20 07:49:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1010/1519] eta 0:08:31 lr 0.000011 time 0.9238 (1.0050) model_time 0.9236 (1.0043) loss 0.8447 (0.8262) grad_norm 5.8452 (8.4693/1.8895) mem 68106MB [2022-12-20 07:49:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1020/1519] eta 0:08:21 lr 0.000011 time 0.9315 (1.0050) model_time 0.9314 (1.0042) loss 0.7514 (0.8261) grad_norm 6.9606 (8.4705/1.8928) mem 68106MB [2022-12-20 07:50:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1030/1519] eta 0:08:11 lr 0.000011 time 0.9256 (1.0049) model_time 0.9254 (1.0042) loss 0.9342 (0.8263) grad_norm 5.4932 (8.4661/1.9071) mem 68106MB [2022-12-20 07:50:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1040/1519] eta 0:08:01 lr 0.000011 time 0.9291 (1.0049) model_time 0.9290 (1.0042) loss 0.6733 (0.8264) grad_norm 11.0872 (8.4678/1.8832) mem 68106MB [2022-12-20 07:50:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1050/1519] eta 0:07:51 lr 0.000011 time 0.9308 (1.0050) model_time 0.9306 (1.0043) loss 0.7046 (0.8258) grad_norm 8.0815 (8.4825/1.8808) mem 68106MB [2022-12-20 07:50:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1060/1519] eta 0:07:41 lr 0.000011 time 0.9339 (1.0053) model_time 0.9337 (1.0046) loss 0.9480 (0.8259) grad_norm 8.3811 (8.4807/1.8633) mem 68106MB [2022-12-20 07:50:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1070/1519] eta 0:07:31 lr 0.000011 time 0.9252 (1.0053) model_time 0.9250 (1.0046) loss 0.8531 (0.8259) grad_norm 9.2232 (8.5009/1.8784) mem 68106MB [2022-12-20 07:50:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1080/1519] eta 0:07:21 lr 0.000011 time 0.9256 (1.0053) model_time 0.9254 (1.0046) loss 0.7840 (0.8259) grad_norm 11.0259 (8.5038/1.8720) mem 68106MB [2022-12-20 07:51:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1090/1519] eta 0:07:11 lr 0.000011 time 0.9264 (1.0052) model_time 0.9263 (1.0045) loss 0.7671 (0.8260) grad_norm 8.7905 (8.5014/1.8655) mem 68106MB [2022-12-20 07:51:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1100/1519] eta 0:07:01 lr 0.000011 time 0.9291 (1.0052) model_time 0.9289 (1.0045) loss 0.9746 (0.8262) grad_norm 7.7037 (8.4771/1.8545) mem 68106MB [2022-12-20 07:51:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1110/1519] eta 0:06:51 lr 0.000011 time 0.9329 (1.0052) model_time 0.9328 (1.0045) loss 0.7653 (0.8258) grad_norm 9.5162 (8.4637/1.8610) mem 68106MB [2022-12-20 07:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1120/1519] eta 0:06:41 lr 0.000011 time 0.9320 (1.0052) model_time 0.9319 (1.0045) loss 0.7012 (0.8264) grad_norm 7.0352 (8.4699/1.8694) mem 68106MB [2022-12-20 07:51:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1130/1519] eta 0:06:31 lr 0.000011 time 0.9194 (1.0052) model_time 0.9192 (1.0045) loss 0.8314 (0.8265) grad_norm 8.6878 (8.4487/1.8202) mem 68106MB [2022-12-20 07:51:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1140/1519] eta 0:06:20 lr 0.000011 time 0.9225 (1.0052) model_time 0.9223 (1.0045) loss 0.6903 (0.8264) grad_norm 6.0615 (8.4311/1.7960) mem 68106MB [2022-12-20 07:52:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1150/1519] eta 0:06:10 lr 0.000011 time 0.9895 (1.0052) model_time 0.9893 (1.0045) loss 1.1114 (0.8263) grad_norm 6.4608 (8.4206/1.7926) mem 68106MB [2022-12-20 07:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1160/1519] eta 0:06:00 lr 0.000011 time 0.9305 (1.0052) model_time 0.9303 (1.0045) loss 0.8309 (0.8261) grad_norm 7.0031 (8.4477/1.7954) mem 68106MB [2022-12-20 07:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1170/1519] eta 0:05:50 lr 0.000011 time 0.9143 (1.0051) model_time 0.9141 (1.0044) loss 0.8653 (0.8264) grad_norm 8.6978 (8.4463/1.7942) mem 68106MB [2022-12-20 07:52:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1180/1519] eta 0:05:40 lr 0.000011 time 0.9254 (1.0051) model_time 0.9252 (1.0045) loss 0.8756 (0.8268) grad_norm 10.5433 (8.5185/1.8828) mem 68106MB [2022-12-20 07:52:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1190/1519] eta 0:05:30 lr 0.000011 time 0.9239 (1.0054) model_time 0.9237 (1.0047) loss 0.8149 (0.8271) grad_norm 6.5080 (8.5094/1.8842) mem 68106MB [2022-12-20 07:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1200/1519] eta 0:05:20 lr 0.000011 time 0.9270 (1.0056) model_time 0.9268 (1.0049) loss 0.6802 (0.8266) grad_norm 6.5568 (8.5052/1.8884) mem 68106MB [2022-12-20 07:53:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1210/1519] eta 0:05:10 lr 0.000011 time 0.9267 (1.0056) model_time 0.9266 (1.0049) loss 0.6753 (0.8268) grad_norm 10.5646 (8.5154/1.8910) mem 68106MB [2022-12-20 07:53:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1220/1519] eta 0:05:00 lr 0.000011 time 0.9379 (1.0055) model_time 0.9378 (1.0049) loss 1.3746 (0.8274) grad_norm 7.6923 (8.5104/1.8905) mem 68106MB [2022-12-20 07:53:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1230/1519] eta 0:04:50 lr 0.000011 time 0.9225 (1.0058) model_time 0.9223 (1.0051) loss 0.7778 (0.8272) grad_norm 13.2272 (8.5225/1.9024) mem 68106MB [2022-12-20 07:53:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1240/1519] eta 0:04:40 lr 0.000010 time 0.9351 (1.0058) model_time 0.9350 (1.0051) loss 0.6808 (0.8266) grad_norm 9.6349 (8.5155/1.8534) mem 68106MB [2022-12-20 07:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1250/1519] eta 0:04:30 lr 0.000010 time 0.9199 (1.0058) model_time 0.9197 (1.0051) loss 0.7028 (0.8264) grad_norm 7.8851 (8.5311/1.8601) mem 68106MB [2022-12-20 07:53:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1260/1519] eta 0:04:20 lr 0.000010 time 0.9268 (1.0057) model_time 0.9267 (1.0051) loss 0.7394 (0.8262) grad_norm 7.9280 (8.5580/1.8485) mem 68106MB [2022-12-20 07:54:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1270/1519] eta 0:04:10 lr 0.000010 time 0.9238 (1.0057) model_time 0.9237 (1.0050) loss 0.7976 (0.8258) grad_norm 7.5117 (8.5592/1.8557) mem 68106MB [2022-12-20 07:54:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1280/1519] eta 0:04:00 lr 0.000010 time 0.9231 (1.0056) model_time 0.9230 (1.0050) loss 0.7670 (0.8255) grad_norm 7.8255 (8.5597/1.8487) mem 68106MB [2022-12-20 07:54:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1290/1519] eta 0:03:50 lr 0.000010 time 0.9190 (1.0056) model_time 0.9188 (1.0049) loss 0.6922 (0.8251) grad_norm 10.9813 (8.5902/1.9646) mem 68106MB [2022-12-20 07:54:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1300/1519] eta 0:03:40 lr 0.000010 time 0.9311 (1.0056) model_time 0.9310 (1.0049) loss 0.6803 (0.8247) grad_norm 6.7555 (8.5944/1.9598) mem 68106MB [2022-12-20 07:54:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1310/1519] eta 0:03:30 lr 0.000010 time 0.9302 (1.0056) model_time 0.9301 (1.0049) loss 0.6842 (0.8252) grad_norm 8.0187 (8.5819/1.9175) mem 68106MB [2022-12-20 07:54:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1320/1519] eta 0:03:20 lr 0.000010 time 0.9271 (1.0056) model_time 0.9269 (1.0050) loss 0.7654 (0.8248) grad_norm 8.9345 (8.5641/1.8688) mem 68106MB [2022-12-20 07:55:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1330/1519] eta 0:03:10 lr 0.000010 time 0.9280 (1.0056) model_time 0.9278 (1.0049) loss 0.7383 (0.8243) grad_norm 9.9727 (8.5778/1.8622) mem 68106MB [2022-12-20 07:55:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1340/1519] eta 0:02:59 lr 0.000010 time 0.9299 (1.0056) model_time 0.9298 (1.0049) loss 0.7920 (0.8239) grad_norm 6.4698 (8.5459/1.8510) mem 68106MB [2022-12-20 07:55:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1350/1519] eta 0:02:49 lr 0.000010 time 0.9297 (1.0055) model_time 0.9296 (1.0049) loss 0.9759 (0.8241) grad_norm 10.3773 (8.5364/1.8628) mem 68106MB [2022-12-20 07:55:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1360/1519] eta 0:02:39 lr 0.000010 time 0.9222 (1.0056) model_time 0.9221 (1.0049) loss 0.8403 (0.8244) grad_norm 9.7044 (8.5376/1.8574) mem 68106MB [2022-12-20 07:55:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1370/1519] eta 0:02:29 lr 0.000010 time 0.9233 (1.0056) model_time 0.9232 (1.0049) loss 0.8884 (0.8245) grad_norm 8.3735 (8.5503/1.8550) mem 68106MB [2022-12-20 07:55:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1380/1519] eta 0:02:19 lr 0.000010 time 0.9366 (1.0055) model_time 0.9365 (1.0049) loss 0.8613 (0.8250) grad_norm 9.6492 (8.5680/1.8567) mem 68106MB [2022-12-20 07:56:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1390/1519] eta 0:02:09 lr 0.000010 time 0.9246 (1.0055) model_time 0.9245 (1.0049) loss 0.6683 (0.8249) grad_norm 7.4736 (8.6039/1.8804) mem 68106MB [2022-12-20 07:56:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1400/1519] eta 0:01:59 lr 0.000010 time 0.9215 (1.0055) model_time 0.9214 (1.0049) loss 0.6927 (0.8246) grad_norm 9.4889 (8.6326/1.8827) mem 68106MB [2022-12-20 07:56:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1410/1519] eta 0:01:49 lr 0.000010 time 0.9265 (1.0055) model_time 0.9263 (1.0049) loss 0.6838 (0.8248) grad_norm 11.3788 (8.6365/1.8477) mem 68106MB [2022-12-20 07:56:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1420/1519] eta 0:01:39 lr 0.000010 time 0.9504 (1.0055) model_time 0.9503 (1.0048) loss 0.8174 (0.8246) grad_norm 7.8736 (8.6670/1.8591) mem 68106MB [2022-12-20 07:56:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1430/1519] eta 0:01:29 lr 0.000010 time 0.9216 (1.0054) model_time 0.9215 (1.0048) loss 0.7786 (0.8247) grad_norm 8.7969 (8.6629/1.8643) mem 68106MB [2022-12-20 07:56:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1440/1519] eta 0:01:19 lr 0.000010 time 0.9305 (1.0054) model_time 0.9303 (1.0047) loss 0.6810 (0.8251) grad_norm 9.2028 (8.6699/1.8692) mem 68106MB [2022-12-20 07:57:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1450/1519] eta 0:01:09 lr 0.000010 time 0.9512 (1.0053) model_time 0.9511 (1.0047) loss 0.8214 (0.8254) grad_norm 7.8448 (8.7210/2.0616) mem 68106MB [2022-12-20 07:57:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1460/1519] eta 0:00:59 lr 0.000010 time 0.9242 (1.0053) model_time 0.9241 (1.0047) loss 0.8724 (0.8254) grad_norm 8.3159 (8.7144/2.0565) mem 68106MB [2022-12-20 07:57:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1470/1519] eta 0:00:49 lr 0.000010 time 0.9222 (1.0054) model_time 0.9220 (1.0048) loss 0.6947 (0.8258) grad_norm 8.3719 (8.6868/2.0516) mem 68106MB [2022-12-20 07:57:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1480/1519] eta 0:00:39 lr 0.000010 time 0.9225 (1.0054) model_time 0.9224 (1.0048) loss 0.6887 (0.8257) grad_norm 9.7611 (8.6722/2.0578) mem 68106MB [2022-12-20 07:57:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1490/1519] eta 0:00:29 lr 0.000010 time 0.9295 (1.0055) model_time 0.9293 (1.0049) loss 0.7092 (0.8256) grad_norm 8.3797 (8.6637/2.0575) mem 68106MB [2022-12-20 07:57:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1500/1519] eta 0:00:19 lr 0.000010 time 0.9250 (1.0057) model_time 0.9248 (1.0051) loss 0.6543 (0.8253) grad_norm 6.8087 (8.6903/2.0534) mem 68106MB [2022-12-20 07:58:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [62/100][1510/1519] eta 0:00:09 lr 0.000010 time 0.9739 (1.0058) model_time 0.9738 (1.0052) loss 0.8020 (0.8252) grad_norm 9.0180 (8.6987/2.1044) mem 68106MB [2022-12-20 07:58:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 62 training takes 0:25:27 [2022-12-20 07:58:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_62.pth saving...... [2022-12-20 07:58:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_62.pth saved !!! [2022-12-20 07:58:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.675 (0.675) Loss 0.5133 (0.5133) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 07:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.331) Loss 0.5283 (0.4978) Acc@1 92.014 (92.677) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 07:58:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.299 (0.316) Loss 0.4597 (0.4949) Acc@1 91.667 (92.493) Acc@5 98.958 (98.429) Mem 68106MB [2022-12-20 07:58:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 0.6241 (0.5001) Acc@1 89.931 (92.316) Acc@5 98.264 (98.365) Mem 68106MB [2022-12-20 07:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.305 (0.308) Loss 0.4492 (0.4898) Acc@1 94.444 (92.370) Acc@5 99.306 (98.476) Mem 68106MB [2022-12-20 07:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.307) Loss 0.4804 (0.4865) Acc@1 91.319 (92.450) Acc@5 99.653 (98.550) Mem 68106MB [2022-12-20 07:59:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.5106 (0.4865) Acc@1 90.972 (92.435) Acc@5 97.917 (98.560) Mem 68106MB [2022-12-20 07:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.297 (0.304) Loss 0.5259 (0.4870) Acc@1 93.403 (92.410) Acc@5 98.264 (98.552) Mem 68106MB [2022-12-20 07:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4312 (0.4853) Acc@1 93.056 (92.421) Acc@5 98.611 (98.581) Mem 68106MB [2022-12-20 07:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:62] * Acc@1 92.387 Acc@5 98.580 [2022-12-20 07:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 07:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 07:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 07:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.39% [2022-12-20 07:59:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][0/1519] eta 0:35:25 lr 0.000010 time 1.3996 (1.3996) model_time 1.0113 (1.0113) loss 0.9656 (0.9656) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 07:59:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][10/1519] eta 0:26:15 lr 0.000010 time 1.0051 (1.0441) model_time 1.0051 (1.0085) loss 0.9123 (0.8233) grad_norm 7.4499 (7.8901/0.8826) mem 68106MB [2022-12-20 07:59:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][20/1519] eta 0:25:38 lr 0.000010 time 0.9397 (1.0260) model_time 0.9396 (1.0073) loss 0.8845 (0.8323) grad_norm 8.1104 (8.0276/0.7589) mem 68106MB [2022-12-20 08:00:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][30/1519] eta 0:25:16 lr 0.000010 time 0.9223 (1.0184) model_time 0.9222 (1.0056) loss 0.6766 (0.8159) grad_norm 6.6884 (7.9388/1.0103) mem 68106MB [2022-12-20 08:00:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][40/1519] eta 0:25:01 lr 0.000010 time 0.9330 (1.0155) model_time 0.9329 (1.0057) loss 0.8151 (0.8182) grad_norm 6.0192 (7.7955/1.0368) mem 68106MB [2022-12-20 08:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][50/1519] eta 0:24:49 lr 0.000010 time 0.9053 (1.0136) model_time 0.9051 (1.0057) loss 1.0558 (0.8329) grad_norm 8.6856 (7.8912/0.9587) mem 68106MB [2022-12-20 08:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][60/1519] eta 0:24:35 lr 0.000010 time 0.9239 (1.0110) model_time 0.9237 (1.0044) loss 0.8108 (0.8320) grad_norm 7.0550 (7.9776/1.3117) mem 68106MB [2022-12-20 08:00:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][70/1519] eta 0:24:22 lr 0.000010 time 0.9375 (1.0096) model_time 0.9373 (1.0038) loss 0.6863 (0.8255) grad_norm 5.7741 (7.8503/1.2875) mem 68106MB [2022-12-20 08:00:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][80/1519] eta 0:24:10 lr 0.000010 time 0.9313 (1.0083) model_time 0.9312 (1.0032) loss 0.6724 (0.8291) grad_norm 12.5031 (8.1298/1.8337) mem 68106MB [2022-12-20 08:01:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][90/1519] eta 0:24:01 lr 0.000010 time 0.9246 (1.0086) model_time 0.9245 (1.0040) loss 0.6904 (0.8324) grad_norm 9.9355 (8.1420/1.8078) mem 68106MB [2022-12-20 08:01:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][100/1519] eta 0:23:52 lr 0.000010 time 0.9855 (1.0098) model_time 0.9853 (1.0056) loss 0.7687 (0.8294) grad_norm 7.5945 (8.1940/1.7583) mem 68106MB [2022-12-20 08:01:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][110/1519] eta 0:23:41 lr 0.000010 time 0.9216 (1.0087) model_time 0.9215 (1.0049) loss 0.6647 (0.8270) grad_norm 8.9893 (8.1980/1.7033) mem 68106MB [2022-12-20 08:01:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][120/1519] eta 0:23:30 lr 0.000010 time 0.9251 (1.0081) model_time 0.9250 (1.0046) loss 0.8929 (0.8223) grad_norm 8.8789 (8.2146/1.6429) mem 68106MB [2022-12-20 08:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][130/1519] eta 0:23:19 lr 0.000010 time 0.9242 (1.0077) model_time 0.9241 (1.0044) loss 0.7088 (0.8196) grad_norm 6.6325 (8.2791/1.6707) mem 68106MB [2022-12-20 08:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][140/1519] eta 0:23:08 lr 0.000010 time 0.9310 (1.0070) model_time 0.9307 (1.0040) loss 0.7763 (0.8185) grad_norm 8.6006 (8.4132/1.7420) mem 68106MB [2022-12-20 08:02:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][150/1519] eta 0:22:59 lr 0.000010 time 1.0360 (1.0075) model_time 1.0359 (1.0046) loss 0.6775 (0.8153) grad_norm 6.6859 (8.3797/1.7008) mem 68106MB [2022-12-20 08:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][160/1519] eta 0:22:48 lr 0.000010 time 0.9227 (1.0070) model_time 0.9226 (1.0042) loss 0.9340 (0.8131) grad_norm 7.9709 (8.4884/1.7596) mem 68106MB [2022-12-20 08:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][170/1519] eta 0:22:38 lr 0.000010 time 0.9794 (1.0067) model_time 0.9793 (1.0042) loss 0.7251 (0.8105) grad_norm 8.4072 (8.4351/1.7374) mem 68106MB [2022-12-20 08:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][180/1519] eta 0:22:27 lr 0.000010 time 0.9298 (1.0062) model_time 0.9296 (1.0038) loss 0.8774 (0.8096) grad_norm 8.1436 (8.4238/1.6949) mem 68106MB [2022-12-20 08:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][190/1519] eta 0:22:18 lr 0.000010 time 0.9194 (1.0071) model_time 0.9193 (1.0048) loss 0.7793 (0.8100) grad_norm 8.2347 (8.3716/1.6681) mem 68106MB [2022-12-20 08:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][200/1519] eta 0:22:08 lr 0.000010 time 0.9315 (1.0073) model_time 0.9314 (1.0050) loss 0.7776 (0.8098) grad_norm 8.8907 (8.3491/1.6536) mem 68106MB [2022-12-20 08:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][210/1519] eta 0:21:58 lr 0.000010 time 0.9309 (1.0069) model_time 0.9307 (1.0048) loss 0.8201 (0.8094) grad_norm 9.2672 (8.3549/1.6264) mem 68106MB [2022-12-20 08:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][220/1519] eta 0:21:47 lr 0.000010 time 0.9305 (1.0068) model_time 0.9304 (1.0047) loss 0.7082 (0.8078) grad_norm 8.4327 (8.3329/1.5977) mem 68106MB [2022-12-20 08:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][230/1519] eta 0:21:37 lr 0.000010 time 0.9207 (1.0064) model_time 0.9205 (1.0044) loss 0.6985 (0.8054) grad_norm 9.0035 (8.3357/1.6028) mem 68106MB [2022-12-20 08:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][240/1519] eta 0:21:27 lr 0.000010 time 0.9230 (1.0064) model_time 0.9229 (1.0044) loss 0.8315 (0.8058) grad_norm 9.0452 (8.3748/1.7019) mem 68106MB [2022-12-20 08:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][250/1519] eta 0:21:16 lr 0.000010 time 0.9239 (1.0063) model_time 0.9238 (1.0044) loss 0.8098 (0.8055) grad_norm 9.0504 (8.4485/1.8594) mem 68106MB [2022-12-20 08:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][260/1519] eta 0:21:06 lr 0.000010 time 0.9233 (1.0060) model_time 0.9231 (1.0042) loss 0.7925 (0.8044) grad_norm 5.6187 (8.3770/1.8690) mem 68106MB [2022-12-20 08:04:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][270/1519] eta 0:20:56 lr 0.000010 time 0.9514 (1.0058) model_time 0.9512 (1.0040) loss 0.8734 (0.8031) grad_norm 6.5633 (8.3750/1.8598) mem 68106MB [2022-12-20 08:04:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][280/1519] eta 0:20:46 lr 0.000010 time 0.9737 (1.0059) model_time 0.9736 (1.0042) loss 0.6682 (0.8027) grad_norm 8.7698 (8.3883/1.8403) mem 68106MB [2022-12-20 08:04:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][290/1519] eta 0:20:36 lr 0.000010 time 0.9224 (1.0061) model_time 0.9222 (1.0044) loss 0.6738 (0.8034) grad_norm 11.2622 (8.3848/1.8354) mem 68106MB [2022-12-20 08:04:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][300/1519] eta 0:20:27 lr 0.000010 time 1.1935 (1.0070) model_time 1.1934 (1.0054) loss 0.7545 (0.8035) grad_norm 6.9147 (8.4200/1.8334) mem 68106MB [2022-12-20 08:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][310/1519] eta 0:20:18 lr 0.000010 time 1.1762 (1.0077) model_time 1.1760 (1.0061) loss 0.7562 (0.8026) grad_norm 9.1309 (8.4495/1.8662) mem 68106MB [2022-12-20 08:04:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][320/1519] eta 0:20:08 lr 0.000010 time 0.9233 (1.0075) model_time 0.9231 (1.0060) loss 0.8727 (0.8045) grad_norm 9.2339 (8.4309/1.8457) mem 68106MB [2022-12-20 08:05:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][330/1519] eta 0:19:58 lr 0.000010 time 0.9316 (1.0081) model_time 0.9314 (1.0066) loss 0.7768 (0.8051) grad_norm 10.6037 (8.4779/1.9146) mem 68106MB [2022-12-20 08:05:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][340/1519] eta 0:19:48 lr 0.000010 time 0.9851 (1.0082) model_time 0.9850 (1.0067) loss 0.7284 (0.8084) grad_norm 8.9487 (8.4936/1.9141) mem 68106MB [2022-12-20 08:05:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][350/1519] eta 0:19:38 lr 0.000010 time 0.9969 (1.0082) model_time 0.9968 (1.0068) loss 0.7762 (0.8081) grad_norm 8.7189 (8.4648/1.9021) mem 68106MB [2022-12-20 08:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][360/1519] eta 0:19:28 lr 0.000010 time 0.9559 (1.0082) model_time 0.9557 (1.0068) loss 0.8870 (0.8061) grad_norm 9.3149 (8.4704/1.8921) mem 68106MB [2022-12-20 08:05:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][370/1519] eta 0:19:18 lr 0.000010 time 0.9265 (1.0083) model_time 0.9264 (1.0070) loss 0.8296 (0.8061) grad_norm 11.1912 (8.4856/1.8806) mem 68106MB [2022-12-20 08:05:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][380/1519] eta 0:19:08 lr 0.000010 time 0.9224 (1.0081) model_time 0.9223 (1.0068) loss 0.9553 (0.8080) grad_norm 12.6570 (8.5080/1.8937) mem 68106MB [2022-12-20 08:06:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][390/1519] eta 0:18:57 lr 0.000010 time 0.9263 (1.0079) model_time 0.9261 (1.0066) loss 0.7192 (0.8081) grad_norm 7.4505 (8.4865/1.8743) mem 68106MB [2022-12-20 08:06:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][400/1519] eta 0:18:47 lr 0.000010 time 0.9165 (1.0080) model_time 0.9164 (1.0067) loss 0.8599 (0.8095) grad_norm 8.3087 (8.4778/1.8636) mem 68106MB [2022-12-20 08:06:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][410/1519] eta 0:18:37 lr 0.000010 time 0.9225 (1.0079) model_time 0.9223 (1.0067) loss 0.7675 (0.8084) grad_norm 10.6078 (8.4661/1.8566) mem 68106MB [2022-12-20 08:06:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][420/1519] eta 0:18:27 lr 0.000010 time 0.9457 (1.0079) model_time 0.9456 (1.0066) loss 0.6788 (0.8093) grad_norm 7.3692 (8.4704/1.8477) mem 68106MB [2022-12-20 08:06:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][430/1519] eta 0:18:17 lr 0.000010 time 0.9321 (1.0077) model_time 0.9320 (1.0064) loss 0.8067 (0.8118) grad_norm 6.9757 (8.4628/1.8418) mem 68106MB [2022-12-20 08:06:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][440/1519] eta 0:18:07 lr 0.000010 time 0.9259 (1.0076) model_time 0.9258 (1.0064) loss 0.6984 (0.8110) grad_norm 7.7281 (8.4341/1.8329) mem 68106MB [2022-12-20 08:07:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][450/1519] eta 0:17:56 lr 0.000010 time 0.9324 (1.0074) model_time 0.9323 (1.0062) loss 0.8170 (0.8103) grad_norm 7.1097 (8.4206/1.8194) mem 68106MB [2022-12-20 08:07:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][460/1519] eta 0:17:46 lr 0.000010 time 1.0200 (1.0075) model_time 1.0198 (1.0064) loss 1.2991 (0.8119) grad_norm 7.7373 (8.4438/1.8372) mem 68106MB [2022-12-20 08:07:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][470/1519] eta 0:17:37 lr 0.000010 time 0.9268 (1.0077) model_time 0.9266 (1.0065) loss 0.8496 (0.8121) grad_norm 7.1867 (8.4667/1.8706) mem 68106MB [2022-12-20 08:07:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][480/1519] eta 0:17:26 lr 0.000010 time 1.0025 (1.0077) model_time 1.0024 (1.0066) loss 0.8584 (0.8126) grad_norm 6.7684 (8.4573/1.8613) mem 68106MB [2022-12-20 08:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][490/1519] eta 0:17:16 lr 0.000010 time 1.0007 (1.0077) model_time 1.0005 (1.0066) loss 0.6664 (0.8122) grad_norm 9.6366 (8.4362/1.8608) mem 68106MB [2022-12-20 08:07:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][500/1519] eta 0:17:06 lr 0.000010 time 0.9196 (1.0075) model_time 0.9194 (1.0064) loss 0.7446 (0.8120) grad_norm 11.1420 (8.4515/1.8534) mem 68106MB [2022-12-20 08:08:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][510/1519] eta 0:16:56 lr 0.000010 time 0.9354 (1.0074) model_time 0.9352 (1.0063) loss 0.9590 (0.8116) grad_norm 8.3129 (8.4998/1.9134) mem 68106MB [2022-12-20 08:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][520/1519] eta 0:16:46 lr 0.000010 time 1.1765 (1.0078) model_time 1.1763 (1.0068) loss 0.7269 (0.8125) grad_norm 15.4526 (8.5204/1.9505) mem 68106MB [2022-12-20 08:08:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][530/1519] eta 0:16:36 lr 0.000010 time 1.0193 (1.0079) model_time 1.0192 (1.0068) loss 0.7336 (0.8133) grad_norm 8.1479 (8.5275/1.9648) mem 68106MB [2022-12-20 08:08:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][540/1519] eta 0:16:26 lr 0.000010 time 0.9235 (1.0077) model_time 0.9234 (1.0067) loss 0.8421 (0.8129) grad_norm 7.2523 (8.5107/1.9515) mem 68106MB [2022-12-20 08:08:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][550/1519] eta 0:16:16 lr 0.000010 time 1.0015 (1.0077) model_time 1.0013 (1.0067) loss 0.6732 (0.8114) grad_norm 6.7202 (8.5021/1.9585) mem 68106MB [2022-12-20 08:08:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][560/1519] eta 0:16:06 lr 0.000010 time 0.9237 (1.0076) model_time 0.9236 (1.0066) loss 0.7475 (0.8119) grad_norm 5.9660 (8.4971/1.9684) mem 68106MB [2022-12-20 08:09:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][570/1519] eta 0:15:56 lr 0.000010 time 0.9295 (1.0074) model_time 0.9293 (1.0064) loss 0.6774 (0.8115) grad_norm 7.3409 (8.4684/1.9641) mem 68106MB [2022-12-20 08:09:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][580/1519] eta 0:15:45 lr 0.000010 time 0.9213 (1.0073) model_time 0.9212 (1.0063) loss 0.7671 (0.8121) grad_norm 7.4161 (8.4740/1.9535) mem 68106MB [2022-12-20 08:09:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][590/1519] eta 0:15:35 lr 0.000010 time 0.9348 (1.0073) model_time 0.9346 (1.0064) loss 0.8168 (0.8126) grad_norm 12.0515 (8.4925/1.9526) mem 68106MB [2022-12-20 08:09:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][600/1519] eta 0:15:26 lr 0.000010 time 0.9246 (1.0077) model_time 0.9244 (1.0068) loss 0.6776 (0.8144) grad_norm 6.4071 (8.4861/1.9593) mem 68106MB [2022-12-20 08:09:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][610/1519] eta 0:15:15 lr 0.000010 time 0.9902 (1.0077) model_time 0.9900 (1.0067) loss 0.9493 (0.8147) grad_norm 7.4959 (8.5075/1.9658) mem 68106MB [2022-12-20 08:09:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][620/1519] eta 0:15:05 lr 0.000010 time 0.9287 (1.0076) model_time 0.9285 (1.0066) loss 0.7942 (0.8140) grad_norm 7.1386 (8.4987/1.9690) mem 68106MB [2022-12-20 08:10:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][630/1519] eta 0:14:55 lr 0.000010 time 0.9252 (1.0075) model_time 0.9250 (1.0065) loss 1.0906 (0.8155) grad_norm 10.0844 (8.5474/1.9940) mem 68106MB [2022-12-20 08:10:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][640/1519] eta 0:14:45 lr 0.000010 time 0.9266 (1.0074) model_time 0.9264 (1.0065) loss 0.8519 (0.8148) grad_norm 5.7418 (8.5529/1.9947) mem 68106MB [2022-12-20 08:10:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][650/1519] eta 0:14:35 lr 0.000010 time 0.9230 (1.0073) model_time 0.9229 (1.0064) loss 0.8454 (0.8156) grad_norm 10.2998 (8.5839/2.0125) mem 68106MB [2022-12-20 08:10:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][660/1519] eta 0:14:25 lr 0.000010 time 0.9234 (1.0073) model_time 0.9233 (1.0063) loss 0.8852 (0.8158) grad_norm 7.0817 (8.5743/1.9993) mem 68106MB [2022-12-20 08:10:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][670/1519] eta 0:14:15 lr 0.000010 time 0.9446 (1.0071) model_time 0.9445 (1.0062) loss 0.6927 (0.8151) grad_norm 9.6383 (8.6013/1.9948) mem 68106MB [2022-12-20 08:10:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][680/1519] eta 0:14:04 lr 0.000010 time 0.9226 (1.0071) model_time 0.9224 (1.0062) loss 0.9847 (0.8147) grad_norm 9.4907 (8.5572/1.9535) mem 68106MB [2022-12-20 08:11:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][690/1519] eta 0:13:54 lr 0.000010 time 0.9227 (1.0071) model_time 0.9226 (1.0062) loss 0.6686 (0.8146) grad_norm 5.7059 (8.5549/1.9512) mem 68106MB [2022-12-20 08:11:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][700/1519] eta 0:13:44 lr 0.000010 time 0.9234 (1.0070) model_time 0.9232 (1.0061) loss 0.8054 (0.8143) grad_norm 11.0578 (8.5515/1.9588) mem 68106MB [2022-12-20 08:11:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][710/1519] eta 0:13:34 lr 0.000010 time 0.9292 (1.0069) model_time 0.9290 (1.0060) loss 0.6752 (0.8151) grad_norm 7.6343 (8.5858/2.0018) mem 68106MB [2022-12-20 08:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][720/1519] eta 0:13:24 lr 0.000010 time 0.8869 (1.0070) model_time 0.8867 (1.0062) loss 1.0286 (0.8169) grad_norm 7.2226 (8.5799/2.0068) mem 68106MB [2022-12-20 08:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][730/1519] eta 0:13:14 lr 0.000010 time 0.9295 (1.0070) model_time 0.9294 (1.0061) loss 0.6888 (0.8168) grad_norm 6.9547 (8.5545/2.0019) mem 68106MB [2022-12-20 08:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][740/1519] eta 0:13:04 lr 0.000010 time 0.9257 (1.0069) model_time 0.9256 (1.0060) loss 0.8182 (0.8172) grad_norm 9.8476 (8.5634/2.0526) mem 68106MB [2022-12-20 08:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][750/1519] eta 0:12:54 lr 0.000010 time 0.9192 (1.0068) model_time 0.9190 (1.0060) loss 0.7355 (0.8168) grad_norm 7.7489 (8.5474/2.0644) mem 68106MB [2022-12-20 08:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][760/1519] eta 0:12:44 lr 0.000010 time 0.9401 (1.0067) model_time 0.9400 (1.0059) loss 0.7038 (0.8167) grad_norm 7.5974 (8.5273/2.0496) mem 68106MB [2022-12-20 08:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][770/1519] eta 0:12:34 lr 0.000010 time 0.9295 (1.0067) model_time 0.9294 (1.0059) loss 1.0974 (0.8170) grad_norm 7.2780 (8.5281/2.0510) mem 68106MB [2022-12-20 08:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][780/1519] eta 0:12:24 lr 0.000010 time 0.9245 (1.0069) model_time 0.9244 (1.0061) loss 0.8457 (0.8170) grad_norm 8.6506 (8.5268/2.0591) mem 68106MB [2022-12-20 08:12:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][790/1519] eta 0:12:14 lr 0.000010 time 0.9884 (1.0069) model_time 0.9882 (1.0061) loss 0.7413 (0.8177) grad_norm 8.7726 (8.5294/2.0641) mem 68106MB [2022-12-20 08:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][800/1519] eta 0:12:04 lr 0.000010 time 0.9228 (1.0072) model_time 0.9227 (1.0064) loss 0.9601 (0.8181) grad_norm 9.6903 (8.5506/2.0671) mem 68106MB [2022-12-20 08:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][810/1519] eta 0:11:54 lr 0.000010 time 0.9268 (1.0071) model_time 0.9266 (1.0063) loss 1.1052 (0.8184) grad_norm 8.6420 (8.5605/2.0785) mem 68106MB [2022-12-20 08:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][820/1519] eta 0:11:43 lr 0.000010 time 0.9226 (1.0070) model_time 0.9224 (1.0062) loss 1.0053 (0.8183) grad_norm 6.7233 (8.5482/2.0854) mem 68106MB [2022-12-20 08:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][830/1519] eta 0:11:33 lr 0.000010 time 0.9236 (1.0070) model_time 0.9234 (1.0062) loss 1.1577 (0.8197) grad_norm 8.1090 (8.5465/2.0832) mem 68106MB [2022-12-20 08:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][840/1519] eta 0:11:23 lr 0.000010 time 0.9731 (1.0071) model_time 0.9730 (1.0063) loss 0.9890 (0.8192) grad_norm 12.3988 (8.5358/2.0577) mem 68106MB [2022-12-20 08:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][850/1519] eta 0:11:13 lr 0.000010 time 0.9313 (1.0070) model_time 0.9310 (1.0063) loss 0.6803 (0.8191) grad_norm 8.1944 (8.5042/2.0055) mem 68106MB [2022-12-20 08:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][860/1519] eta 0:11:03 lr 0.000010 time 0.9277 (1.0070) model_time 0.9275 (1.0062) loss 0.6632 (0.8186) grad_norm 6.6137 (8.5208/1.9906) mem 68106MB [2022-12-20 08:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][870/1519] eta 0:10:53 lr 0.000010 time 0.9223 (1.0070) model_time 0.9221 (1.0062) loss 0.6697 (0.8188) grad_norm 11.5573 (8.5264/1.9957) mem 68106MB [2022-12-20 08:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][880/1519] eta 0:10:43 lr 0.000010 time 0.9260 (1.0070) model_time 0.9259 (1.0062) loss 1.1208 (0.8193) grad_norm 7.3843 (8.5360/2.0042) mem 68106MB [2022-12-20 08:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][890/1519] eta 0:10:33 lr 0.000010 time 0.9257 (1.0069) model_time 0.9255 (1.0062) loss 0.7061 (0.8187) grad_norm 8.1165 (8.5332/1.9933) mem 68106MB [2022-12-20 08:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][900/1519] eta 0:10:23 lr 0.000010 time 0.9324 (1.0069) model_time 0.9323 (1.0061) loss 0.7277 (0.8185) grad_norm 7.7196 (8.5171/1.9822) mem 68106MB [2022-12-20 08:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][910/1519] eta 0:10:13 lr 0.000010 time 0.9262 (1.0068) model_time 0.9261 (1.0061) loss 0.7217 (0.8183) grad_norm 8.4881 (8.4976/1.9618) mem 68106MB [2022-12-20 08:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][920/1519] eta 0:10:03 lr 0.000010 time 0.9349 (1.0068) model_time 0.9346 (1.0061) loss 1.0649 (0.8193) grad_norm 7.6741 (8.5107/1.9816) mem 68106MB [2022-12-20 08:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][930/1519] eta 0:09:52 lr 0.000010 time 0.9330 (1.0067) model_time 0.9329 (1.0060) loss 0.8567 (0.8205) grad_norm 10.1038 (8.4762/1.9461) mem 68106MB [2022-12-20 08:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][940/1519] eta 0:09:42 lr 0.000010 time 0.9266 (1.0067) model_time 0.9265 (1.0059) loss 0.8710 (0.8204) grad_norm 9.7356 (8.4773/1.9384) mem 68106MB [2022-12-20 08:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][950/1519] eta 0:09:32 lr 0.000010 time 0.9200 (1.0068) model_time 0.9199 (1.0061) loss 1.0419 (0.8209) grad_norm 8.4039 (8.4951/1.9347) mem 68106MB [2022-12-20 08:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][960/1519] eta 0:09:22 lr 0.000010 time 0.9375 (1.0069) model_time 0.9374 (1.0061) loss 0.7207 (0.8214) grad_norm 17.1777 (8.5071/1.9992) mem 68106MB [2022-12-20 08:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][970/1519] eta 0:09:12 lr 0.000010 time 0.9239 (1.0069) model_time 0.9238 (1.0061) loss 0.7875 (0.8213) grad_norm 6.9368 (8.4995/2.0024) mem 68106MB [2022-12-20 08:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][980/1519] eta 0:09:02 lr 0.000010 time 0.9260 (1.0068) model_time 0.9259 (1.0061) loss 0.8827 (0.8212) grad_norm 7.0661 (8.4786/1.9832) mem 68106MB [2022-12-20 08:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][990/1519] eta 0:08:52 lr 0.000010 time 0.9679 (1.0068) model_time 0.9678 (1.0061) loss 0.6876 (0.8215) grad_norm 5.7871 (8.4894/2.0020) mem 68106MB [2022-12-20 08:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1000/1519] eta 0:08:42 lr 0.000010 time 0.9320 (1.0067) model_time 0.9319 (1.0060) loss 0.6692 (0.8219) grad_norm 5.9630 (8.4720/2.0038) mem 68106MB [2022-12-20 08:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1010/1519] eta 0:08:32 lr 0.000010 time 0.9259 (1.0066) model_time 0.9258 (1.0059) loss 0.8551 (0.8217) grad_norm 6.5758 (8.4575/2.0045) mem 68106MB [2022-12-20 08:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1020/1519] eta 0:08:22 lr 0.000010 time 0.9830 (1.0066) model_time 0.9829 (1.0059) loss 0.8475 (0.8216) grad_norm 6.9047 (8.4517/2.0061) mem 68106MB [2022-12-20 08:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1030/1519] eta 0:08:12 lr 0.000010 time 0.9289 (1.0067) model_time 0.9288 (1.0060) loss 0.7444 (0.8220) grad_norm 10.0328 (8.4582/2.0003) mem 68106MB [2022-12-20 08:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1040/1519] eta 0:08:02 lr 0.000010 time 0.9375 (1.0067) model_time 0.9374 (1.0060) loss 0.8661 (0.8215) grad_norm 5.0006 (8.4617/2.0072) mem 68106MB [2022-12-20 08:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1050/1519] eta 0:07:52 lr 0.000010 time 0.9304 (1.0068) model_time 0.9303 (1.0061) loss 0.7179 (0.8211) grad_norm 9.8594 (8.4684/2.0090) mem 68106MB [2022-12-20 08:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1060/1519] eta 0:07:42 lr 0.000010 time 0.9729 (1.0068) model_time 0.9728 (1.0061) loss 0.9932 (0.8213) grad_norm 8.5850 (8.4395/1.9992) mem 68106MB [2022-12-20 08:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1070/1519] eta 0:07:32 lr 0.000010 time 0.9216 (1.0067) model_time 0.9214 (1.0060) loss 0.6877 (0.8210) grad_norm 6.5190 (8.4248/1.9759) mem 68106MB [2022-12-20 08:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1080/1519] eta 0:07:21 lr 0.000010 time 0.9209 (1.0067) model_time 0.9208 (1.0060) loss 0.9184 (0.8210) grad_norm 8.3017 (8.4167/1.9770) mem 68106MB [2022-12-20 08:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1090/1519] eta 0:07:11 lr 0.000010 time 0.9298 (1.0067) model_time 0.9297 (1.0061) loss 0.9911 (0.8212) grad_norm 7.2092 (8.4391/1.9766) mem 68106MB [2022-12-20 08:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1100/1519] eta 0:07:01 lr 0.000010 time 0.9306 (1.0068) model_time 0.9305 (1.0062) loss 0.7924 (0.8211) grad_norm 7.1961 (8.4264/1.9713) mem 68106MB [2022-12-20 08:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1110/1519] eta 0:06:51 lr 0.000010 time 0.9251 (1.0069) model_time 0.9250 (1.0062) loss 0.9456 (0.8215) grad_norm 5.9637 (8.4162/1.9535) mem 68106MB [2022-12-20 08:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1120/1519] eta 0:06:41 lr 0.000010 time 0.9256 (1.0068) model_time 0.9254 (1.0062) loss 0.7969 (0.8224) grad_norm 9.5223 (8.3885/1.9105) mem 68106MB [2022-12-20 08:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1130/1519] eta 0:06:31 lr 0.000010 time 0.9235 (1.0068) model_time 0.9233 (1.0062) loss 0.7915 (0.8225) grad_norm 10.3225 (8.3927/1.9026) mem 68106MB [2022-12-20 08:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1140/1519] eta 0:06:21 lr 0.000010 time 0.9224 (1.0068) model_time 0.9223 (1.0062) loss 0.6630 (0.8224) grad_norm 7.8081 (8.4235/1.9206) mem 68106MB [2022-12-20 08:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1150/1519] eta 0:06:11 lr 0.000010 time 0.9394 (1.0069) model_time 0.9393 (1.0063) loss 0.8383 (0.8229) grad_norm 8.3920 (8.4563/1.9232) mem 68106MB [2022-12-20 08:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1160/1519] eta 0:06:01 lr 0.000010 time 0.9261 (1.0069) model_time 0.9260 (1.0063) loss 0.8715 (0.8227) grad_norm 7.9412 (8.4499/1.9023) mem 68106MB [2022-12-20 08:19:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1170/1519] eta 0:05:51 lr 0.000010 time 0.9303 (1.0069) model_time 0.9301 (1.0062) loss 0.6688 (0.8222) grad_norm 8.2222 (8.4927/1.9023) mem 68106MB [2022-12-20 08:19:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1180/1519] eta 0:05:41 lr 0.000010 time 0.9250 (1.0068) model_time 0.9248 (1.0062) loss 0.8926 (0.8220) grad_norm 7.0320 (8.4976/1.9058) mem 68106MB [2022-12-20 08:19:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1190/1519] eta 0:05:31 lr 0.000010 time 0.9279 (1.0068) model_time 0.9277 (1.0061) loss 0.8016 (0.8219) grad_norm 8.1174 (8.4609/1.8978) mem 68106MB [2022-12-20 08:19:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1200/1519] eta 0:05:21 lr 0.000010 time 0.9232 (1.0067) model_time 0.9231 (1.0061) loss 0.7610 (0.8219) grad_norm 8.2941 (8.4735/1.8918) mem 68106MB [2022-12-20 08:19:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1210/1519] eta 0:05:11 lr 0.000010 time 0.9690 (1.0067) model_time 0.9688 (1.0061) loss 0.6888 (0.8212) grad_norm 7.8718 (8.4372/1.8911) mem 68106MB [2022-12-20 08:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1220/1519] eta 0:05:01 lr 0.000010 time 0.9271 (1.0068) model_time 0.9270 (1.0061) loss 1.0408 (0.8213) grad_norm 7.7820 (8.4724/1.9187) mem 68106MB [2022-12-20 08:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1230/1519] eta 0:04:50 lr 0.000010 time 0.9235 (1.0067) model_time 0.9234 (1.0061) loss 0.6801 (0.8209) grad_norm 10.8775 (8.4462/1.8867) mem 68106MB [2022-12-20 08:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1240/1519] eta 0:04:40 lr 0.000010 time 0.9728 (1.0067) model_time 0.9727 (1.0061) loss 0.6676 (0.8212) grad_norm 8.5437 (8.4570/1.8788) mem 68106MB [2022-12-20 08:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1250/1519] eta 0:04:30 lr 0.000010 time 0.9191 (1.0067) model_time 0.9190 (1.0061) loss 0.8112 (0.8212) grad_norm 8.6828 (8.4501/1.8854) mem 68106MB [2022-12-20 08:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1260/1519] eta 0:04:20 lr 0.000010 time 0.9257 (1.0066) model_time 0.9255 (1.0060) loss 0.8572 (0.8209) grad_norm 7.7979 (8.4633/1.8893) mem 68106MB [2022-12-20 08:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1270/1519] eta 0:04:10 lr 0.000010 time 0.9221 (1.0067) model_time 0.9219 (1.0061) loss 0.9332 (0.8206) grad_norm 8.9037 (8.4826/1.8948) mem 68106MB [2022-12-20 08:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1280/1519] eta 0:04:00 lr 0.000010 time 0.9225 (1.0066) model_time 0.9224 (1.0060) loss 0.7985 (0.8208) grad_norm 9.3333 (8.4892/1.8879) mem 68106MB [2022-12-20 08:21:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1290/1519] eta 0:03:50 lr 0.000010 time 0.9288 (1.0066) model_time 0.9286 (1.0060) loss 0.7756 (0.8212) grad_norm 6.7186 (8.4880/1.8949) mem 68106MB [2022-12-20 08:21:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1300/1519] eta 0:03:40 lr 0.000010 time 0.9504 (1.0066) model_time 0.9503 (1.0060) loss 0.8277 (0.8213) grad_norm 5.8752 (8.4919/1.9057) mem 68106MB [2022-12-20 08:21:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1310/1519] eta 0:03:30 lr 0.000010 time 0.9195 (1.0066) model_time 0.9193 (1.0059) loss 0.6970 (0.8214) grad_norm 7.0518 (8.4564/1.8680) mem 68106MB [2022-12-20 08:21:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1320/1519] eta 0:03:20 lr 0.000010 time 0.9199 (1.0065) model_time 0.9198 (1.0059) loss 0.9647 (0.8214) grad_norm 8.3139 (8.4452/1.8687) mem 68106MB [2022-12-20 08:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1330/1519] eta 0:03:10 lr 0.000010 time 0.9245 (1.0065) model_time 0.9244 (1.0059) loss 0.9987 (0.8213) grad_norm 11.0019 (8.4838/1.8749) mem 68106MB [2022-12-20 08:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1340/1519] eta 0:03:00 lr 0.000010 time 0.9252 (1.0065) model_time 0.9250 (1.0058) loss 0.8298 (0.8210) grad_norm 9.0200 (8.4657/1.8059) mem 68106MB [2022-12-20 08:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1350/1519] eta 0:02:50 lr 0.000010 time 0.9274 (1.0064) model_time 0.9273 (1.0058) loss 1.0760 (0.8209) grad_norm 7.7269 (8.4946/1.7922) mem 68106MB [2022-12-20 08:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1360/1519] eta 0:02:40 lr 0.000010 time 0.9295 (1.0064) model_time 0.9293 (1.0058) loss 0.6716 (0.8207) grad_norm 9.3811 (8.4793/1.7874) mem 68106MB [2022-12-20 08:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1370/1519] eta 0:02:29 lr 0.000010 time 0.9332 (1.0064) model_time 0.9330 (1.0058) loss 0.7680 (0.8206) grad_norm 10.2089 (8.4921/1.7883) mem 68106MB [2022-12-20 08:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1380/1519] eta 0:02:19 lr 0.000010 time 0.9328 (1.0063) model_time 0.9326 (1.0057) loss 0.7989 (0.8205) grad_norm 6.8692 (8.4930/1.7819) mem 68106MB [2022-12-20 08:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1390/1519] eta 0:02:09 lr 0.000010 time 0.9394 (1.0064) model_time 0.9392 (1.0058) loss 1.1958 (0.8203) grad_norm 7.8669 (8.4784/1.7875) mem 68106MB [2022-12-20 08:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1400/1519] eta 0:01:59 lr 0.000010 time 0.9203 (1.0063) model_time 0.9202 (1.0058) loss 0.9421 (0.8200) grad_norm 10.0104 (8.4878/1.8523) mem 68106MB [2022-12-20 08:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1410/1519] eta 0:01:49 lr 0.000010 time 0.9244 (1.0064) model_time 0.9243 (1.0058) loss 1.2120 (0.8200) grad_norm 10.4361 (8.4679/1.8501) mem 68106MB [2022-12-20 08:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1420/1519] eta 0:01:39 lr 0.000010 time 0.9203 (1.0064) model_time 0.9201 (1.0058) loss 0.7281 (0.8199) grad_norm 9.3259 (8.4801/1.8460) mem 68106MB [2022-12-20 08:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1430/1519] eta 0:01:29 lr 0.000010 time 0.9245 (1.0064) model_time 0.9244 (1.0058) loss 0.8780 (0.8200) grad_norm 10.0782 (8.5002/1.8472) mem 68106MB [2022-12-20 08:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1440/1519] eta 0:01:19 lr 0.000010 time 0.9215 (1.0062) model_time 0.9214 (1.0056) loss 1.1338 (0.8207) grad_norm 6.5199 (8.4895/1.8543) mem 68106MB [2022-12-20 08:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1450/1519] eta 0:01:09 lr 0.000010 time 0.9223 (1.0062) model_time 0.9221 (1.0056) loss 0.8593 (0.8206) grad_norm 9.3476 (8.4821/1.8307) mem 68106MB [2022-12-20 08:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1460/1519] eta 0:00:59 lr 0.000010 time 0.9186 (1.0062) model_time 0.9185 (1.0056) loss 0.7604 (0.8206) grad_norm 8.0074 (8.4888/1.8334) mem 68106MB [2022-12-20 08:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1470/1519] eta 0:00:49 lr 0.000010 time 0.9389 (1.0062) model_time 0.9388 (1.0056) loss 0.8708 (0.8205) grad_norm 7.6745 (8.4951/1.8252) mem 68106MB [2022-12-20 08:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1480/1519] eta 0:00:39 lr 0.000010 time 0.9654 (1.0062) model_time 0.9652 (1.0056) loss 0.8238 (0.8209) grad_norm 8.2110 (8.4501/1.8109) mem 68106MB [2022-12-20 08:24:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1490/1519] eta 0:00:29 lr 0.000010 time 0.9329 (1.0062) model_time 0.9327 (1.0056) loss 0.6774 (0.8206) grad_norm 9.0219 (8.4455/1.8194) mem 68106MB [2022-12-20 08:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1500/1519] eta 0:00:19 lr 0.000010 time 0.9242 (1.0062) model_time 0.9240 (1.0056) loss 0.9062 (0.8207) grad_norm 7.0005 (8.4312/1.8250) mem 68106MB [2022-12-20 08:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [63/100][1510/1519] eta 0:00:09 lr 0.000010 time 0.9189 (1.0061) model_time 0.9188 (1.0055) loss 0.9807 (0.8208) grad_norm 7.7421 (8.4620/1.8659) mem 68106MB [2022-12-20 08:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 63 training takes 0:25:28 [2022-12-20 08:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_63.pth saving...... [2022-12-20 08:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_63.pth saved !!! [2022-12-20 08:25:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.685 (0.685) Loss 0.5293 (0.5293) Acc@1 90.972 (90.972) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 08:25:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.299 (0.335) Loss 0.5234 (0.5024) Acc@1 93.403 (92.330) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-20 08:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.317) Loss 0.4771 (0.4968) Acc@1 91.667 (92.411) Acc@5 99.306 (98.396) Mem 68106MB [2022-12-20 08:25:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.305 (0.312) Loss 0.6202 (0.5017) Acc@1 89.583 (92.182) Acc@5 98.264 (98.410) Mem 68106MB [2022-12-20 08:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.309) Loss 0.4561 (0.4923) Acc@1 93.750 (92.353) Acc@5 98.958 (98.484) Mem 68106MB [2022-12-20 08:25:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.307) Loss 0.4852 (0.4900) Acc@1 90.625 (92.354) Acc@5 99.653 (98.557) Mem 68106MB [2022-12-20 08:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.303 (0.306) Loss 0.5037 (0.4896) Acc@1 91.319 (92.333) Acc@5 97.917 (98.537) Mem 68106MB [2022-12-20 08:25:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.296 (0.305) Loss 0.5498 (0.4907) Acc@1 91.667 (92.263) Acc@5 98.611 (98.538) Mem 68106MB [2022-12-20 08:25:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.304) Loss 0.4233 (0.4887) Acc@1 93.403 (92.280) Acc@5 98.611 (98.568) Mem 68106MB [2022-12-20 08:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:63] * Acc@1 92.256 Acc@5 98.572 [2022-12-20 08:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.3% [2022-12-20 08:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.39% [2022-12-20 08:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][0/1519] eta 0:45:55 lr 0.000010 time 1.8137 (1.8137) model_time 1.0872 (1.0872) loss 0.7044 (0.7044) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 08:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][10/1519] eta 0:26:56 lr 0.000010 time 0.9285 (1.0712) model_time 0.9284 (1.0048) loss 0.8647 (0.7986) grad_norm 8.0952 (7.7719/0.3618) mem 68106MB [2022-12-20 08:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][20/1519] eta 0:25:57 lr 0.000010 time 0.9305 (1.0391) model_time 0.9304 (1.0041) loss 0.8153 (0.7999) grad_norm 7.4092 (7.5826/0.8166) mem 68106MB [2022-12-20 08:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][30/1519] eta 0:25:29 lr 0.000010 time 0.9294 (1.0272) model_time 0.9293 (1.0035) loss 0.9221 (0.8377) grad_norm 8.7006 (8.8286/2.8693) mem 68106MB [2022-12-20 08:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][40/1519] eta 0:25:09 lr 0.000010 time 0.9266 (1.0207) model_time 0.9265 (1.0027) loss 0.9344 (0.8264) grad_norm 6.8059 (8.4990/2.6249) mem 68106MB [2022-12-20 08:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][50/1519] eta 0:24:52 lr 0.000010 time 0.9214 (1.0162) model_time 0.9213 (1.0017) loss 0.6609 (0.8146) grad_norm 6.3712 (8.2973/2.4194) mem 68106MB [2022-12-20 08:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][60/1519] eta 0:24:39 lr 0.000010 time 0.9752 (1.0139) model_time 0.9750 (1.0016) loss 1.0955 (0.8331) grad_norm 8.5478 (8.4133/2.2436) mem 68106MB [2022-12-20 08:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][70/1519] eta 0:24:26 lr 0.000010 time 0.9219 (1.0123) model_time 0.9217 (1.0018) loss 0.9571 (0.8295) grad_norm 9.4578 (8.6906/2.3705) mem 68106MB [2022-12-20 08:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][80/1519] eta 0:24:14 lr 0.000010 time 0.9173 (1.0105) model_time 0.9172 (1.0012) loss 1.1467 (0.8245) grad_norm 8.9484 (8.8457/2.4501) mem 68106MB [2022-12-20 08:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][90/1519] eta 0:24:02 lr 0.000010 time 0.9225 (1.0091) model_time 0.9224 (1.0008) loss 0.8687 (0.8252) grad_norm 8.1583 (8.7283/2.3389) mem 68106MB [2022-12-20 08:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][100/1519] eta 0:23:50 lr 0.000010 time 0.9247 (1.0080) model_time 0.9246 (1.0005) loss 0.6925 (0.8229) grad_norm 9.2008 (8.6826/2.2854) mem 68106MB [2022-12-20 08:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][110/1519] eta 0:23:41 lr 0.000010 time 0.9844 (1.0092) model_time 0.9842 (1.0023) loss 0.9638 (0.8269) grad_norm 6.3624 (8.5825/2.2155) mem 68106MB [2022-12-20 08:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][120/1519] eta 0:23:31 lr 0.000010 time 0.9255 (1.0093) model_time 0.9254 (1.0030) loss 0.9375 (0.8279) grad_norm 6.7985 (8.5192/2.1489) mem 68106MB [2022-12-20 08:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][130/1519] eta 0:23:20 lr 0.000010 time 0.9231 (1.0086) model_time 0.9230 (1.0027) loss 0.9085 (0.8329) grad_norm 7.8940 (8.4896/2.0930) mem 68106MB [2022-12-20 08:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][140/1519] eta 0:23:10 lr 0.000010 time 0.9272 (1.0083) model_time 0.9269 (1.0029) loss 0.8342 (0.8290) grad_norm 10.9066 (8.4689/2.0888) mem 68106MB [2022-12-20 08:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][150/1519] eta 0:22:59 lr 0.000010 time 0.9177 (1.0080) model_time 0.9176 (1.0029) loss 0.6938 (0.8242) grad_norm 8.1985 (8.4417/2.0327) mem 68106MB [2022-12-20 08:28:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][160/1519] eta 0:22:51 lr 0.000010 time 0.9213 (1.0093) model_time 0.9212 (1.0045) loss 0.7668 (0.8221) grad_norm 6.9021 (8.3550/2.0040) mem 68106MB [2022-12-20 08:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][170/1519] eta 0:22:40 lr 0.000010 time 0.9233 (1.0086) model_time 0.9232 (1.0040) loss 0.7444 (0.8194) grad_norm 7.8426 (8.3497/1.9483) mem 68106MB [2022-12-20 08:28:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][180/1519] eta 0:22:29 lr 0.000010 time 0.9259 (1.0080) model_time 0.9257 (1.0037) loss 0.8833 (0.8187) grad_norm 5.6479 (8.2872/1.9228) mem 68106MB [2022-12-20 08:29:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][190/1519] eta 0:22:19 lr 0.000010 time 0.9263 (1.0080) model_time 0.9262 (1.0039) loss 0.6719 (0.8236) grad_norm 6.1447 (8.2789/1.9223) mem 68106MB [2022-12-20 08:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][200/1519] eta 0:22:09 lr 0.000010 time 0.8913 (1.0078) model_time 0.8911 (1.0039) loss 0.7603 (0.8266) grad_norm 7.3299 (8.2927/1.9081) mem 68106MB [2022-12-20 08:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][210/1519] eta 0:21:59 lr 0.000010 time 0.9195 (1.0077) model_time 0.9194 (1.0039) loss 0.6679 (0.8237) grad_norm 9.2205 (8.3921/2.0663) mem 68106MB [2022-12-20 08:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][220/1519] eta 0:21:48 lr 0.000010 time 0.9249 (1.0072) model_time 0.9247 (1.0036) loss 0.8038 (0.8254) grad_norm 8.9039 (8.3863/2.0522) mem 68106MB [2022-12-20 08:29:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][230/1519] eta 0:21:38 lr 0.000010 time 0.9318 (1.0072) model_time 0.9316 (1.0038) loss 0.7202 (0.8280) grad_norm 10.7757 (8.3754/2.0482) mem 68106MB [2022-12-20 08:29:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][240/1519] eta 0:21:27 lr 0.000010 time 0.9251 (1.0070) model_time 0.9249 (1.0037) loss 0.8539 (0.8274) grad_norm 6.9855 (8.3902/2.0853) mem 68106MB [2022-12-20 08:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][250/1519] eta 0:21:17 lr 0.000010 time 0.9241 (1.0066) model_time 0.9240 (1.0034) loss 0.8461 (0.8257) grad_norm 8.2106 (8.3825/2.0587) mem 68106MB [2022-12-20 08:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][260/1519] eta 0:21:06 lr 0.000010 time 0.9213 (1.0063) model_time 0.9211 (1.0033) loss 0.6643 (0.8265) grad_norm 7.8085 (8.3930/2.0252) mem 68106MB [2022-12-20 08:30:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][270/1519] eta 0:20:56 lr 0.000010 time 0.9327 (1.0061) model_time 0.9325 (1.0031) loss 0.8146 (0.8244) grad_norm 14.2388 (8.4134/2.0609) mem 68106MB [2022-12-20 08:30:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][280/1519] eta 0:20:46 lr 0.000010 time 0.9267 (1.0059) model_time 0.9265 (1.0030) loss 0.7067 (0.8239) grad_norm 10.6301 (8.4053/2.0382) mem 68106MB [2022-12-20 08:30:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][290/1519] eta 0:20:35 lr 0.000010 time 0.9205 (1.0056) model_time 0.9203 (1.0028) loss 0.7212 (0.8241) grad_norm 9.6499 (8.4893/2.0623) mem 68106MB [2022-12-20 08:30:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][300/1519] eta 0:20:25 lr 0.000010 time 0.9289 (1.0054) model_time 0.9287 (1.0026) loss 0.8825 (0.8238) grad_norm 9.6941 (8.5184/2.0444) mem 68106MB [2022-12-20 08:31:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][310/1519] eta 0:20:15 lr 0.000010 time 1.0322 (1.0056) model_time 1.0321 (1.0030) loss 0.9050 (0.8250) grad_norm 11.3230 (8.5185/2.0426) mem 68106MB [2022-12-20 08:31:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][320/1519] eta 0:20:05 lr 0.000010 time 0.9286 (1.0054) model_time 0.9284 (1.0028) loss 0.7006 (0.8255) grad_norm 7.5235 (8.5109/2.0169) mem 68106MB [2022-12-20 08:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][330/1519] eta 0:19:56 lr 0.000010 time 0.9175 (1.0062) model_time 0.9174 (1.0037) loss 1.1865 (0.8262) grad_norm 6.8222 (8.4932/1.9921) mem 68106MB [2022-12-20 08:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][340/1519] eta 0:19:46 lr 0.000010 time 0.9204 (1.0061) model_time 0.9203 (1.0037) loss 0.8230 (0.8251) grad_norm 7.1829 (8.4675/1.9808) mem 68106MB [2022-12-20 08:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][350/1519] eta 0:19:35 lr 0.000010 time 0.9258 (1.0060) model_time 0.9256 (1.0036) loss 0.7869 (0.8226) grad_norm 8.3162 (8.4562/1.9586) mem 68106MB [2022-12-20 08:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][360/1519] eta 0:19:25 lr 0.000010 time 0.9317 (1.0058) model_time 0.9316 (1.0034) loss 0.8400 (0.8213) grad_norm 6.9455 (8.4297/1.9432) mem 68106MB [2022-12-20 08:32:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][370/1519] eta 0:19:15 lr 0.000010 time 0.9359 (1.0056) model_time 0.9358 (1.0033) loss 0.8713 (0.8211) grad_norm 8.1543 (8.4096/1.9250) mem 68106MB [2022-12-20 08:32:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][380/1519] eta 0:19:05 lr 0.000010 time 0.9320 (1.0057) model_time 0.9318 (1.0035) loss 0.9560 (0.8214) grad_norm 8.6433 (8.4109/1.9059) mem 68106MB [2022-12-20 08:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][390/1519] eta 0:18:55 lr 0.000010 time 0.9242 (1.0056) model_time 0.9241 (1.0034) loss 0.6841 (0.8201) grad_norm 12.7502 (8.4332/1.9104) mem 68106MB [2022-12-20 08:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][400/1519] eta 0:18:45 lr 0.000010 time 0.9251 (1.0055) model_time 0.9249 (1.0033) loss 0.9098 (0.8202) grad_norm 9.0612 (8.4551/1.8957) mem 68106MB [2022-12-20 08:32:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][410/1519] eta 0:18:34 lr 0.000010 time 0.9352 (1.0053) model_time 0.9350 (1.0032) loss 0.7013 (0.8215) grad_norm 9.4595 (8.4516/1.8787) mem 68106MB [2022-12-20 08:32:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][420/1519] eta 0:18:25 lr 0.000010 time 0.9297 (1.0055) model_time 0.9296 (1.0034) loss 0.7795 (0.8211) grad_norm 11.2294 (8.4656/1.8801) mem 68106MB [2022-12-20 08:33:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][430/1519] eta 0:18:15 lr 0.000010 time 0.9238 (1.0056) model_time 0.9237 (1.0037) loss 1.0051 (0.8211) grad_norm 7.0336 (8.4719/1.8735) mem 68106MB [2022-12-20 08:33:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][440/1519] eta 0:18:05 lr 0.000010 time 0.9349 (1.0057) model_time 0.9348 (1.0037) loss 0.6914 (0.8210) grad_norm 10.7869 (8.4934/1.8629) mem 68106MB [2022-12-20 08:33:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][450/1519] eta 0:17:55 lr 0.000010 time 0.9211 (1.0059) model_time 0.9210 (1.0039) loss 0.6904 (0.8200) grad_norm 18.6862 (8.5437/1.9647) mem 68106MB [2022-12-20 08:33:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][460/1519] eta 0:17:45 lr 0.000010 time 0.9250 (1.0057) model_time 0.9249 (1.0038) loss 0.6749 (0.8183) grad_norm 7.6253 (8.5282/1.9579) mem 68106MB [2022-12-20 08:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][470/1519] eta 0:17:35 lr 0.000010 time 0.9290 (1.0060) model_time 0.9288 (1.0042) loss 0.9385 (0.8180) grad_norm 10.2368 (8.5167/1.9494) mem 68106MB [2022-12-20 08:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][480/1519] eta 0:17:25 lr 0.000010 time 0.9229 (1.0059) model_time 0.9227 (1.0040) loss 0.7346 (0.8185) grad_norm 10.8093 (8.5079/1.9458) mem 68106MB [2022-12-20 08:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][490/1519] eta 0:17:15 lr 0.000010 time 1.0564 (1.0060) model_time 1.0563 (1.0042) loss 0.9570 (0.8199) grad_norm 6.2937 (8.4951/1.9575) mem 68106MB [2022-12-20 08:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][500/1519] eta 0:17:05 lr 0.000010 time 0.9268 (1.0060) model_time 0.9266 (1.0042) loss 0.8611 (0.8201) grad_norm 11.9592 (8.5411/1.9777) mem 68106MB [2022-12-20 08:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][510/1519] eta 0:16:55 lr 0.000010 time 0.9270 (1.0059) model_time 0.9269 (1.0042) loss 0.7676 (0.8203) grad_norm 6.2249 (8.5306/1.9681) mem 68106MB [2022-12-20 08:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][520/1519] eta 0:16:44 lr 0.000010 time 0.9276 (1.0059) model_time 0.9274 (1.0042) loss 0.7043 (0.8216) grad_norm 7.3366 (8.5203/1.9582) mem 68106MB [2022-12-20 08:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][530/1519] eta 0:16:34 lr 0.000010 time 0.9300 (1.0058) model_time 0.9299 (1.0041) loss 0.9551 (0.8226) grad_norm 6.5516 (8.5281/1.9584) mem 68106MB [2022-12-20 08:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][540/1519] eta 0:16:24 lr 0.000010 time 0.9303 (1.0058) model_time 0.9302 (1.0042) loss 0.6766 (0.8234) grad_norm 10.1817 (8.5194/1.9501) mem 68106MB [2022-12-20 08:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][550/1519] eta 0:16:14 lr 0.000010 time 0.9247 (1.0057) model_time 0.9246 (1.0040) loss 0.7210 (0.8243) grad_norm 8.8011 (8.4973/1.9424) mem 68106MB [2022-12-20 08:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][560/1519] eta 0:16:04 lr 0.000010 time 0.9210 (1.0058) model_time 0.9209 (1.0042) loss 0.6990 (0.8242) grad_norm 6.9587 (8.4722/1.9346) mem 68106MB [2022-12-20 08:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][570/1519] eta 0:15:54 lr 0.000010 time 0.9321 (1.0057) model_time 0.9320 (1.0041) loss 0.8674 (0.8234) grad_norm 10.7701 (8.4734/1.9279) mem 68106MB [2022-12-20 08:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][580/1519] eta 0:15:44 lr 0.000010 time 0.9344 (1.0057) model_time 0.9343 (1.0041) loss 1.0754 (0.8227) grad_norm 6.1080 (8.4866/1.9363) mem 68106MB [2022-12-20 08:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][590/1519] eta 0:15:34 lr 0.000010 time 0.9216 (1.0056) model_time 0.9215 (1.0040) loss 0.9014 (0.8223) grad_norm 7.5395 (8.4750/1.9240) mem 68106MB [2022-12-20 08:35:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][600/1519] eta 0:15:24 lr 0.000010 time 0.9319 (1.0055) model_time 0.9316 (1.0040) loss 0.9025 (0.8236) grad_norm 8.9088 (8.4816/1.9169) mem 68106MB [2022-12-20 08:36:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][610/1519] eta 0:15:13 lr 0.000010 time 0.9288 (1.0055) model_time 0.9287 (1.0040) loss 0.9585 (0.8236) grad_norm 7.9315 (8.5121/1.9307) mem 68106MB [2022-12-20 08:36:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][620/1519] eta 0:15:03 lr 0.000010 time 0.9045 (1.0055) model_time 0.9044 (1.0040) loss 0.9484 (0.8240) grad_norm 9.2485 (8.5452/1.9411) mem 68106MB [2022-12-20 08:36:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][630/1519] eta 0:14:53 lr 0.000010 time 0.9275 (1.0054) model_time 0.9273 (1.0039) loss 0.6894 (0.8225) grad_norm 7.3178 (8.5035/1.8612) mem 68106MB [2022-12-20 08:36:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][640/1519] eta 0:14:43 lr 0.000010 time 0.9218 (1.0053) model_time 0.9217 (1.0038) loss 0.6919 (0.8219) grad_norm 16.2101 (8.5379/1.9066) mem 68106MB [2022-12-20 08:36:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][650/1519] eta 0:14:33 lr 0.000010 time 0.9209 (1.0054) model_time 0.9208 (1.0040) loss 1.0252 (0.8223) grad_norm 6.7792 (8.5492/1.9128) mem 68106MB [2022-12-20 08:36:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][660/1519] eta 0:14:23 lr 0.000010 time 0.9263 (1.0053) model_time 0.9262 (1.0039) loss 0.7116 (0.8220) grad_norm 7.4127 (8.5369/1.9209) mem 68106MB [2022-12-20 08:37:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][670/1519] eta 0:14:13 lr 0.000010 time 0.9262 (1.0052) model_time 0.9261 (1.0038) loss 0.8568 (0.8223) grad_norm 11.5327 (8.5243/1.8990) mem 68106MB [2022-12-20 08:37:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][680/1519] eta 0:14:03 lr 0.000010 time 0.9234 (1.0052) model_time 0.9232 (1.0038) loss 0.9607 (0.8216) grad_norm 8.4444 (8.5051/1.8874) mem 68106MB [2022-12-20 08:37:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][690/1519] eta 0:13:53 lr 0.000010 time 0.9271 (1.0051) model_time 0.9269 (1.0037) loss 0.8589 (0.8220) grad_norm 8.6357 (8.5059/1.8919) mem 68106MB [2022-12-20 08:37:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][700/1519] eta 0:13:43 lr 0.000010 time 0.9210 (1.0050) model_time 0.9208 (1.0036) loss 0.6914 (0.8217) grad_norm 10.0720 (8.5215/1.8836) mem 68106MB [2022-12-20 08:37:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][710/1519] eta 0:13:32 lr 0.000010 time 0.9269 (1.0049) model_time 0.9268 (1.0035) loss 0.8153 (0.8211) grad_norm 7.3387 (8.5290/1.8896) mem 68106MB [2022-12-20 08:37:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][720/1519] eta 0:13:22 lr 0.000010 time 0.9165 (1.0048) model_time 0.9164 (1.0035) loss 0.6713 (0.8214) grad_norm 6.1599 (8.5361/1.9050) mem 68106MB [2022-12-20 08:38:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][730/1519] eta 0:13:12 lr 0.000010 time 0.9280 (1.0049) model_time 0.9278 (1.0036) loss 0.7957 (0.8212) grad_norm 7.0404 (8.5269/1.9037) mem 68106MB [2022-12-20 08:38:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][740/1519] eta 0:13:02 lr 0.000010 time 0.9233 (1.0050) model_time 0.9232 (1.0037) loss 0.7888 (0.8206) grad_norm 6.3762 (8.5262/1.9069) mem 68106MB [2022-12-20 08:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][750/1519] eta 0:12:52 lr 0.000010 time 0.9267 (1.0051) model_time 0.9266 (1.0038) loss 0.7763 (0.8206) grad_norm 8.7014 (8.5263/1.9095) mem 68106MB [2022-12-20 08:38:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][760/1519] eta 0:12:42 lr 0.000010 time 0.9280 (1.0051) model_time 0.9279 (1.0039) loss 0.6945 (0.8202) grad_norm 7.8158 (8.5368/1.9039) mem 68106MB [2022-12-20 08:38:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][770/1519] eta 0:12:32 lr 0.000010 time 0.9255 (1.0051) model_time 0.9254 (1.0039) loss 0.8303 (0.8206) grad_norm 8.5062 (8.5269/1.9099) mem 68106MB [2022-12-20 08:38:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][780/1519] eta 0:12:22 lr 0.000010 time 0.9790 (1.0051) model_time 0.9789 (1.0039) loss 0.7245 (0.8207) grad_norm 9.8358 (8.5504/1.9025) mem 68106MB [2022-12-20 08:39:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][790/1519] eta 0:12:12 lr 0.000010 time 0.9508 (1.0051) model_time 0.9505 (1.0038) loss 0.9085 (0.8215) grad_norm 7.2451 (8.5690/1.9154) mem 68106MB [2022-12-20 08:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][800/1519] eta 0:12:02 lr 0.000010 time 0.9398 (1.0050) model_time 0.9397 (1.0038) loss 0.6700 (0.8220) grad_norm 7.7365 (8.5517/1.9168) mem 68106MB [2022-12-20 08:39:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][810/1519] eta 0:11:52 lr 0.000010 time 0.9484 (1.0051) model_time 0.9482 (1.0039) loss 0.8325 (0.8230) grad_norm 8.1291 (8.5161/1.8582) mem 68106MB [2022-12-20 08:39:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][820/1519] eta 0:11:42 lr 0.000010 time 0.9245 (1.0053) model_time 0.9244 (1.0041) loss 0.8306 (0.8238) grad_norm 7.8337 (8.5086/1.8481) mem 68106MB [2022-12-20 08:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][830/1519] eta 0:11:32 lr 0.000010 time 0.9284 (1.0055) model_time 0.9283 (1.0043) loss 0.6895 (0.8229) grad_norm 9.9946 (8.5116/1.8453) mem 68106MB [2022-12-20 08:39:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][840/1519] eta 0:11:22 lr 0.000010 time 0.9009 (1.0054) model_time 0.9008 (1.0042) loss 0.6979 (0.8226) grad_norm 7.2325 (8.4929/1.8163) mem 68106MB [2022-12-20 08:40:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][850/1519] eta 0:11:12 lr 0.000010 time 0.9934 (1.0054) model_time 0.9932 (1.0042) loss 1.2629 (0.8239) grad_norm 6.0715 (8.4922/1.8200) mem 68106MB [2022-12-20 08:40:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][860/1519] eta 0:11:02 lr 0.000010 time 0.9277 (1.0053) model_time 0.9274 (1.0042) loss 1.1719 (0.8237) grad_norm 8.7939 (8.4837/1.8211) mem 68106MB [2022-12-20 08:40:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][870/1519] eta 0:10:52 lr 0.000010 time 0.9205 (1.0054) model_time 0.9204 (1.0042) loss 0.8143 (0.8234) grad_norm 6.1963 (8.4636/1.7949) mem 68106MB [2022-12-20 08:40:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][880/1519] eta 0:10:42 lr 0.000010 time 0.9247 (1.0053) model_time 0.9245 (1.0042) loss 0.9529 (0.8231) grad_norm 6.5378 (8.4973/1.8435) mem 68106MB [2022-12-20 08:40:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][890/1519] eta 0:10:32 lr 0.000010 time 0.9260 (1.0054) model_time 0.9259 (1.0042) loss 0.6821 (0.8232) grad_norm 7.7734 (8.4516/1.8223) mem 68106MB [2022-12-20 08:40:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][900/1519] eta 0:10:22 lr 0.000010 time 0.9224 (1.0053) model_time 0.9222 (1.0042) loss 0.9173 (0.8239) grad_norm 9.7023 (8.4202/1.8273) mem 68106MB [2022-12-20 08:41:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][910/1519] eta 0:10:12 lr 0.000010 time 0.9291 (1.0053) model_time 0.9289 (1.0041) loss 0.8011 (0.8252) grad_norm 6.3028 (8.4470/1.8620) mem 68106MB [2022-12-20 08:41:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][920/1519] eta 0:10:02 lr 0.000010 time 0.9233 (1.0053) model_time 0.9232 (1.0042) loss 0.6761 (0.8248) grad_norm 7.4626 (8.4312/1.8666) mem 68106MB [2022-12-20 08:41:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][930/1519] eta 0:09:52 lr 0.000010 time 0.9227 (1.0052) model_time 0.9225 (1.0041) loss 1.0460 (0.8244) grad_norm 7.3318 (8.4382/1.8699) mem 68106MB [2022-12-20 08:41:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][940/1519] eta 0:09:42 lr 0.000010 time 0.9369 (1.0053) model_time 0.9367 (1.0042) loss 0.9747 (0.8244) grad_norm 8.4637 (8.4502/1.8623) mem 68106MB [2022-12-20 08:41:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][950/1519] eta 0:09:31 lr 0.000010 time 0.9385 (1.0053) model_time 0.9384 (1.0042) loss 0.6919 (0.8247) grad_norm 6.4293 (8.4426/1.8716) mem 68106MB [2022-12-20 08:41:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][960/1519] eta 0:09:21 lr 0.000010 time 0.9746 (1.0053) model_time 0.9745 (1.0042) loss 0.9061 (0.8252) grad_norm 11.1229 (8.4632/1.8715) mem 68106MB [2022-12-20 08:42:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][970/1519] eta 0:09:11 lr 0.000010 time 0.9291 (1.0052) model_time 0.9289 (1.0042) loss 0.9251 (0.8251) grad_norm 7.0871 (8.4710/1.8679) mem 68106MB [2022-12-20 08:42:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][980/1519] eta 0:09:01 lr 0.000010 time 0.9231 (1.0052) model_time 0.9230 (1.0041) loss 0.7639 (0.8247) grad_norm 8.9534 (8.4788/1.8718) mem 68106MB [2022-12-20 08:42:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][990/1519] eta 0:08:51 lr 0.000010 time 0.9137 (1.0051) model_time 0.9135 (1.0040) loss 0.7624 (0.8254) grad_norm 6.3339 (8.4850/1.9036) mem 68106MB [2022-12-20 08:42:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1000/1519] eta 0:08:41 lr 0.000010 time 0.9227 (1.0050) model_time 0.9225 (1.0040) loss 0.7552 (0.8249) grad_norm 7.8119 (8.4786/1.9031) mem 68106MB [2022-12-20 08:42:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1010/1519] eta 0:08:31 lr 0.000010 time 0.9315 (1.0051) model_time 0.9314 (1.0040) loss 0.7681 (0.8250) grad_norm 13.5062 (8.4931/1.9234) mem 68106MB [2022-12-20 08:42:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1020/1519] eta 0:08:21 lr 0.000010 time 0.9230 (1.0051) model_time 0.9229 (1.0040) loss 0.7560 (0.8240) grad_norm 7.4598 (8.4819/1.9128) mem 68106MB [2022-12-20 08:43:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1030/1519] eta 0:08:11 lr 0.000010 time 0.9937 (1.0051) model_time 0.9936 (1.0040) loss 1.2123 (0.8247) grad_norm 6.7508 (8.4681/1.9110) mem 68106MB [2022-12-20 08:43:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1040/1519] eta 0:08:01 lr 0.000010 time 0.9692 (1.0050) model_time 0.9690 (1.0040) loss 1.0803 (0.8245) grad_norm 8.2470 (8.4352/1.9109) mem 68106MB [2022-12-20 08:43:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1050/1519] eta 0:07:51 lr 0.000010 time 1.0178 (1.0050) model_time 1.0177 (1.0040) loss 0.8024 (0.8245) grad_norm 9.9406 (8.3910/1.8220) mem 68106MB [2022-12-20 08:43:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1060/1519] eta 0:07:41 lr 0.000010 time 0.9259 (1.0050) model_time 0.9258 (1.0040) loss 0.7826 (0.8247) grad_norm 5.9974 (8.4057/1.8214) mem 68106MB [2022-12-20 08:43:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1070/1519] eta 0:07:31 lr 0.000010 time 0.9237 (1.0049) model_time 0.9235 (1.0039) loss 0.6584 (0.8239) grad_norm 11.3074 (8.4161/1.8268) mem 68106MB [2022-12-20 08:43:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1080/1519] eta 0:07:21 lr 0.000010 time 0.9250 (1.0049) model_time 0.9247 (1.0039) loss 0.6620 (0.8239) grad_norm 6.6375 (8.4291/1.8261) mem 68106MB [2022-12-20 08:44:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1090/1519] eta 0:07:11 lr 0.000010 time 0.9734 (1.0049) model_time 0.9733 (1.0039) loss 0.7184 (0.8241) grad_norm 8.1687 (8.4597/1.8448) mem 68106MB [2022-12-20 08:44:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1100/1519] eta 0:07:01 lr 0.000010 time 0.9229 (1.0048) model_time 0.9228 (1.0039) loss 0.9194 (0.8242) grad_norm 7.5791 (8.4586/1.8808) mem 68106MB [2022-12-20 08:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1110/1519] eta 0:06:50 lr 0.000010 time 0.9283 (1.0048) model_time 0.9281 (1.0038) loss 0.7929 (0.8240) grad_norm 10.8742 (8.4784/1.8818) mem 68106MB [2022-12-20 08:44:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1120/1519] eta 0:06:40 lr 0.000010 time 0.9275 (1.0048) model_time 0.9273 (1.0038) loss 0.9871 (0.8241) grad_norm 10.6872 (8.5210/1.9200) mem 68106MB [2022-12-20 08:44:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1130/1519] eta 0:06:30 lr 0.000010 time 0.9626 (1.0048) model_time 0.9625 (1.0038) loss 0.9047 (0.8238) grad_norm 7.9423 (8.5179/1.9107) mem 68106MB [2022-12-20 08:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1140/1519] eta 0:06:20 lr 0.000010 time 0.9279 (1.0047) model_time 0.9278 (1.0038) loss 0.7788 (0.8236) grad_norm 11.6575 (8.5359/1.9285) mem 68106MB [2022-12-20 08:45:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1150/1519] eta 0:06:10 lr 0.000010 time 0.9203 (1.0047) model_time 0.9201 (1.0037) loss 0.8864 (0.8232) grad_norm 5.4626 (8.5686/1.9375) mem 68106MB [2022-12-20 08:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1160/1519] eta 0:06:00 lr 0.000010 time 0.9248 (1.0047) model_time 0.9246 (1.0037) loss 0.8314 (0.8231) grad_norm 5.8166 (8.5964/1.9453) mem 68106MB [2022-12-20 08:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1170/1519] eta 0:05:50 lr 0.000010 time 0.9311 (1.0046) model_time 0.9310 (1.0037) loss 0.9466 (0.8240) grad_norm 8.6277 (8.6033/1.9505) mem 68106MB [2022-12-20 08:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1180/1519] eta 0:05:40 lr 0.000010 time 1.0127 (1.0047) model_time 1.0125 (1.0037) loss 0.6835 (0.8237) grad_norm 7.0524 (8.6112/1.9698) mem 68106MB [2022-12-20 08:45:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1190/1519] eta 0:05:30 lr 0.000010 time 0.9180 (1.0046) model_time 0.9178 (1.0037) loss 0.9078 (0.8235) grad_norm 8.1555 (8.6386/1.9723) mem 68106MB [2022-12-20 08:45:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1200/1519] eta 0:05:20 lr 0.000010 time 0.9638 (1.0046) model_time 0.9636 (1.0037) loss 0.8575 (0.8238) grad_norm 8.2710 (8.6479/1.9791) mem 68106MB [2022-12-20 08:46:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1210/1519] eta 0:05:10 lr 0.000010 time 0.9393 (1.0046) model_time 0.9392 (1.0037) loss 0.8657 (0.8239) grad_norm 8.9666 (8.6212/1.9671) mem 68106MB [2022-12-20 08:46:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1220/1519] eta 0:05:00 lr 0.000010 time 0.9707 (1.0047) model_time 0.9705 (1.0037) loss 0.7640 (0.8240) grad_norm 9.4140 (8.6059/1.9517) mem 68106MB [2022-12-20 08:46:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1230/1519] eta 0:04:50 lr 0.000010 time 1.0223 (1.0047) model_time 1.0221 (1.0038) loss 0.9423 (0.8246) grad_norm 10.1537 (8.6039/1.9387) mem 68106MB [2022-12-20 08:46:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1240/1519] eta 0:04:40 lr 0.000010 time 0.9460 (1.0047) model_time 0.9458 (1.0038) loss 0.7444 (0.8241) grad_norm 7.2276 (8.5831/1.8854) mem 68106MB [2022-12-20 08:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1250/1519] eta 0:04:30 lr 0.000010 time 0.9328 (1.0047) model_time 0.9327 (1.0038) loss 0.8367 (0.8240) grad_norm 6.5292 (8.5888/1.8801) mem 68106MB [2022-12-20 08:46:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1260/1519] eta 0:04:20 lr 0.000010 time 0.9296 (1.0048) model_time 0.9295 (1.0039) loss 0.6919 (0.8237) grad_norm 7.9537 (8.5882/1.8729) mem 68106MB [2022-12-20 08:47:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1270/1519] eta 0:04:10 lr 0.000010 time 0.9751 (1.0048) model_time 0.9750 (1.0039) loss 1.0315 (0.8237) grad_norm 6.6635 (8.5771/1.8837) mem 68106MB [2022-12-20 08:47:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1280/1519] eta 0:04:00 lr 0.000010 time 0.9205 (1.0049) model_time 0.9204 (1.0040) loss 0.7496 (0.8231) grad_norm 7.7635 (8.5924/1.8751) mem 68106MB [2022-12-20 08:47:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1290/1519] eta 0:03:50 lr 0.000010 time 0.9232 (1.0048) model_time 0.9230 (1.0039) loss 1.1472 (0.8239) grad_norm 10.8946 (8.6239/1.8748) mem 68106MB [2022-12-20 08:47:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1300/1519] eta 0:03:40 lr 0.000010 time 0.9246 (1.0048) model_time 0.9245 (1.0039) loss 0.7798 (0.8243) grad_norm 6.9189 (8.6028/1.8769) mem 68106MB [2022-12-20 08:47:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1310/1519] eta 0:03:29 lr 0.000010 time 0.9218 (1.0047) model_time 0.9216 (1.0039) loss 0.7002 (0.8242) grad_norm 7.0447 (8.5876/1.8749) mem 68106MB [2022-12-20 08:47:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1320/1519] eta 0:03:19 lr 0.000010 time 0.9246 (1.0048) model_time 0.9245 (1.0039) loss 0.6685 (0.8240) grad_norm 8.6631 (8.5975/1.8615) mem 68106MB [2022-12-20 08:48:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1330/1519] eta 0:03:09 lr 0.000010 time 0.9245 (1.0048) model_time 0.9244 (1.0039) loss 0.8830 (0.8238) grad_norm 8.3600 (8.6003/1.8716) mem 68106MB [2022-12-20 08:48:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1340/1519] eta 0:02:59 lr 0.000010 time 0.9250 (1.0047) model_time 0.9248 (1.0038) loss 0.6873 (0.8236) grad_norm 5.8866 (8.6015/1.8630) mem 68106MB [2022-12-20 08:48:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1350/1519] eta 0:02:49 lr 0.000010 time 0.9232 (1.0047) model_time 0.9231 (1.0038) loss 0.6684 (0.8232) grad_norm 8.0825 (8.6167/1.8728) mem 68106MB [2022-12-20 08:48:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1360/1519] eta 0:02:39 lr 0.000010 time 0.9220 (1.0047) model_time 0.9219 (1.0039) loss 0.8049 (0.8228) grad_norm 7.0924 (8.6358/1.8715) mem 68106MB [2022-12-20 08:48:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1370/1519] eta 0:02:29 lr 0.000010 time 0.9313 (1.0047) model_time 0.9312 (1.0038) loss 1.1561 (0.8227) grad_norm 8.4209 (8.6539/1.8682) mem 68106MB [2022-12-20 08:48:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1380/1519] eta 0:02:19 lr 0.000010 time 0.9176 (1.0046) model_time 0.9174 (1.0038) loss 0.8714 (0.8224) grad_norm 7.9479 (8.6503/1.8719) mem 68106MB [2022-12-20 08:49:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1390/1519] eta 0:02:09 lr 0.000009 time 0.9356 (1.0046) model_time 0.9354 (1.0038) loss 0.6735 (0.8223) grad_norm 6.3089 (8.6232/1.8498) mem 68106MB [2022-12-20 08:49:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1400/1519] eta 0:01:59 lr 0.000009 time 0.9192 (1.0046) model_time 0.9190 (1.0038) loss 0.7127 (0.8222) grad_norm 8.2855 (8.6280/1.8634) mem 68106MB [2022-12-20 08:49:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1410/1519] eta 0:01:49 lr 0.000009 time 0.9224 (1.0046) model_time 0.9223 (1.0037) loss 0.7446 (0.8220) grad_norm 8.1017 (8.6406/1.8780) mem 68106MB [2022-12-20 08:49:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1420/1519] eta 0:01:39 lr 0.000009 time 0.9251 (1.0046) model_time 0.9249 (1.0037) loss 0.8997 (0.8218) grad_norm 8.8406 (8.6545/1.8837) mem 68106MB [2022-12-20 08:49:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1430/1519] eta 0:01:29 lr 0.000009 time 0.9401 (1.0047) model_time 0.9400 (1.0038) loss 0.7694 (0.8215) grad_norm 7.8123 (8.6545/1.8713) mem 68106MB [2022-12-20 08:49:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1440/1519] eta 0:01:19 lr 0.000009 time 0.9203 (1.0046) model_time 0.9202 (1.0038) loss 0.8274 (0.8217) grad_norm 12.3602 (8.6817/1.8844) mem 68106MB [2022-12-20 08:50:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1450/1519] eta 0:01:09 lr 0.000009 time 0.9179 (1.0047) model_time 0.9177 (1.0039) loss 0.6688 (0.8214) grad_norm 7.5301 (8.6643/1.8848) mem 68106MB [2022-12-20 08:50:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1460/1519] eta 0:00:59 lr 0.000009 time 0.9282 (1.0047) model_time 0.9281 (1.0039) loss 0.8482 (0.8218) grad_norm 8.3355 (8.6634/1.8858) mem 68106MB [2022-12-20 08:50:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1470/1519] eta 0:00:49 lr 0.000009 time 0.9243 (1.0047) model_time 0.9241 (1.0039) loss 1.1006 (0.8219) grad_norm 8.6237 (8.6788/1.8886) mem 68106MB [2022-12-20 08:50:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1480/1519] eta 0:00:39 lr 0.000009 time 0.9231 (1.0046) model_time 0.9230 (1.0038) loss 0.8136 (0.8220) grad_norm 9.9632 (8.6403/1.8490) mem 68106MB [2022-12-20 08:50:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1490/1519] eta 0:00:29 lr 0.000009 time 0.9907 (1.0046) model_time 0.9906 (1.0038) loss 0.6827 (0.8218) grad_norm 8.8814 (8.6489/1.8391) mem 68106MB [2022-12-20 08:50:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1500/1519] eta 0:00:19 lr 0.000009 time 0.9214 (1.0046) model_time 0.9213 (1.0038) loss 0.9695 (0.8223) grad_norm 9.1540 (8.6522/1.8389) mem 68106MB [2022-12-20 08:51:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [64/100][1510/1519] eta 0:00:09 lr 0.000009 time 0.9235 (1.0046) model_time 0.9234 (1.0038) loss 0.9169 (0.8221) grad_norm 9.3719 (8.6135/1.7955) mem 68106MB [2022-12-20 08:51:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 64 training takes 0:25:25 [2022-12-20 08:51:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_64.pth saving...... [2022-12-20 08:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_64.pth saved !!! [2022-12-20 08:51:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.653 (0.653) Loss 0.5132 (0.5132) Acc@1 92.361 (92.361) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 08:51:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.331) Loss 0.5200 (0.4943) Acc@1 92.708 (92.708) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 08:51:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.316) Loss 0.4704 (0.4916) Acc@1 91.319 (92.692) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 08:51:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.311) Loss 0.6251 (0.4984) Acc@1 89.583 (92.328) Acc@5 97.917 (98.398) Mem 68106MB [2022-12-20 08:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.308) Loss 0.4583 (0.4896) Acc@1 94.097 (92.437) Acc@5 99.306 (98.493) Mem 68106MB [2022-12-20 08:51:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.306 (0.306) Loss 0.4851 (0.4869) Acc@1 90.972 (92.443) Acc@5 99.653 (98.557) Mem 68106MB [2022-12-20 08:52:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 0.5014 (0.4865) Acc@1 91.319 (92.384) Acc@5 97.917 (98.543) Mem 68106MB [2022-12-20 08:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5367 (0.4876) Acc@1 92.708 (92.356) Acc@5 98.264 (98.523) Mem 68106MB [2022-12-20 08:52:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.302 (0.303) Loss 0.4290 (0.4861) Acc@1 93.403 (92.370) Acc@5 98.264 (98.547) Mem 68106MB [2022-12-20 08:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:64] * Acc@1 92.313 Acc@5 98.547 [2022-12-20 08:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.3% [2022-12-20 08:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.39% [2022-12-20 08:52:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][0/1519] eta 0:47:57 lr 0.000009 time 1.8943 (1.8943) model_time 1.1494 (1.1494) loss 0.6930 (0.6930) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 08:52:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][10/1519] eta 0:27:10 lr 0.000009 time 0.9206 (1.0803) model_time 0.9205 (1.0123) loss 0.7138 (0.8228) grad_norm 10.5511 (7.6418/1.5033) mem 68106MB [2022-12-20 08:52:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][20/1519] eta 0:26:23 lr 0.000009 time 0.9040 (1.0563) model_time 0.9039 (1.0206) loss 0.7700 (0.8089) grad_norm 11.6082 (8.4544/1.6910) mem 68106MB [2022-12-20 08:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][30/1519] eta 0:25:55 lr 0.000009 time 1.0315 (1.0446) model_time 1.0314 (1.0202) loss 0.8606 (0.8128) grad_norm 9.7723 (8.4983/1.5633) mem 68106MB [2022-12-20 08:52:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][40/1519] eta 0:25:28 lr 0.000009 time 0.9295 (1.0332) model_time 0.9294 (1.0147) loss 1.0966 (0.8252) grad_norm 8.9858 (8.4621/1.3875) mem 68106MB [2022-12-20 08:52:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][50/1519] eta 0:25:08 lr 0.000009 time 0.9270 (1.0270) model_time 0.9269 (1.0121) loss 0.6711 (0.8188) grad_norm 11.2030 (8.7450/1.4621) mem 68106MB [2022-12-20 08:53:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][60/1519] eta 0:24:53 lr 0.000009 time 0.9296 (1.0238) model_time 0.9295 (1.0113) loss 0.7673 (0.8138) grad_norm 16.3158 (9.0422/1.9331) mem 68106MB [2022-12-20 08:53:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][70/1519] eta 0:24:39 lr 0.000009 time 0.9280 (1.0211) model_time 0.9278 (1.0103) loss 1.1051 (0.8204) grad_norm 6.4453 (8.9100/1.9863) mem 68106MB [2022-12-20 08:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][80/1519] eta 0:24:25 lr 0.000009 time 0.9223 (1.0182) model_time 0.9221 (1.0087) loss 0.8063 (0.8181) grad_norm 7.1352 (8.9870/2.1891) mem 68106MB [2022-12-20 08:53:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][90/1519] eta 0:24:12 lr 0.000009 time 0.9362 (1.0162) model_time 0.9361 (1.0077) loss 0.8620 (0.8195) grad_norm 7.9069 (8.8977/2.1275) mem 68106MB [2022-12-20 08:53:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][100/1519] eta 0:23:59 lr 0.000009 time 0.9231 (1.0143) model_time 0.9229 (1.0066) loss 0.7435 (0.8123) grad_norm 8.1349 (8.8034/2.0507) mem 68106MB [2022-12-20 08:54:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][110/1519] eta 0:23:48 lr 0.000009 time 0.9336 (1.0136) model_time 0.9334 (1.0065) loss 0.7105 (0.8095) grad_norm 7.4887 (8.7019/1.9836) mem 68106MB [2022-12-20 08:54:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][120/1519] eta 0:23:37 lr 0.000009 time 0.9156 (1.0134) model_time 0.9155 (1.0070) loss 0.8310 (0.8121) grad_norm 7.5555 (8.7177/1.9910) mem 68106MB [2022-12-20 08:54:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][130/1519] eta 0:23:26 lr 0.000009 time 0.9193 (1.0125) model_time 0.9191 (1.0065) loss 0.7607 (0.8161) grad_norm 7.5600 (8.6710/1.9274) mem 68106MB [2022-12-20 08:54:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][140/1519] eta 0:23:15 lr 0.000009 time 0.9225 (1.0122) model_time 0.9223 (1.0066) loss 0.8990 (0.8148) grad_norm 7.8767 (8.6156/1.8902) mem 68106MB [2022-12-20 08:54:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][150/1519] eta 0:23:04 lr 0.000009 time 0.9300 (1.0113) model_time 0.9299 (1.0061) loss 0.6815 (0.8187) grad_norm 12.8744 (8.6645/1.9089) mem 68106MB [2022-12-20 08:54:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][160/1519] eta 0:22:55 lr 0.000009 time 0.9309 (1.0124) model_time 0.9307 (1.0074) loss 0.8972 (0.8176) grad_norm 7.3222 (8.5599/1.8942) mem 68106MB [2022-12-20 08:55:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][170/1519] eta 0:22:44 lr 0.000009 time 0.9178 (1.0115) model_time 0.9177 (1.0068) loss 0.7003 (0.8144) grad_norm 6.9358 (8.5377/1.8574) mem 68106MB [2022-12-20 08:55:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][180/1519] eta 0:22:33 lr 0.000009 time 0.9192 (1.0107) model_time 0.9190 (1.0063) loss 0.7003 (0.8157) grad_norm 11.3312 (8.5176/1.8935) mem 68106MB [2022-12-20 08:55:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][190/1519] eta 0:22:22 lr 0.000009 time 0.9291 (1.0105) model_time 0.9289 (1.0063) loss 0.8098 (0.8127) grad_norm 6.7997 (8.4755/1.8655) mem 68106MB [2022-12-20 08:55:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][200/1519] eta 0:22:12 lr 0.000009 time 0.9180 (1.0100) model_time 0.9179 (1.0060) loss 0.6785 (0.8117) grad_norm 7.5304 (8.4537/1.8515) mem 68106MB [2022-12-20 08:55:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][210/1519] eta 0:22:01 lr 0.000009 time 0.9305 (1.0096) model_time 0.9304 (1.0057) loss 0.8003 (0.8133) grad_norm 8.0169 (8.4708/1.9117) mem 68106MB [2022-12-20 08:55:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][220/1519] eta 0:21:51 lr 0.000009 time 0.9263 (1.0096) model_time 0.9262 (1.0060) loss 0.6901 (0.8109) grad_norm 7.4722 (8.4688/1.9629) mem 68106MB [2022-12-20 08:56:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][230/1519] eta 0:21:41 lr 0.000009 time 0.9222 (1.0094) model_time 0.9221 (1.0059) loss 0.8794 (0.8114) grad_norm 7.4856 (8.4227/1.9394) mem 68106MB [2022-12-20 08:56:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][240/1519] eta 0:21:30 lr 0.000009 time 0.9984 (1.0094) model_time 0.9983 (1.0060) loss 0.6995 (0.8076) grad_norm 6.8927 (8.3891/1.9195) mem 68106MB [2022-12-20 08:56:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][250/1519] eta 0:21:20 lr 0.000009 time 0.9358 (1.0091) model_time 0.9356 (1.0059) loss 0.7768 (0.8074) grad_norm 11.2595 (8.4310/1.9213) mem 68106MB [2022-12-20 08:56:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][260/1519] eta 0:21:10 lr 0.000009 time 0.9265 (1.0087) model_time 0.9264 (1.0056) loss 0.7084 (0.8097) grad_norm 8.3440 (8.4328/1.9224) mem 68106MB [2022-12-20 08:56:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][270/1519] eta 0:20:59 lr 0.000009 time 0.9187 (1.0088) model_time 0.9186 (1.0057) loss 0.7399 (0.8074) grad_norm 8.9798 (8.4354/1.9289) mem 68106MB [2022-12-20 08:56:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][280/1519] eta 0:20:49 lr 0.000009 time 0.9324 (1.0084) model_time 0.9322 (1.0054) loss 0.9166 (0.8079) grad_norm 10.3860 (8.5084/2.0135) mem 68106MB [2022-12-20 08:57:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][290/1519] eta 0:20:38 lr 0.000009 time 0.9252 (1.0081) model_time 0.9250 (1.0052) loss 0.7797 (0.8072) grad_norm 8.0640 (8.4819/1.9941) mem 68106MB [2022-12-20 08:57:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][300/1519] eta 0:20:29 lr 0.000009 time 0.9121 (1.0082) model_time 0.9120 (1.0055) loss 0.8747 (0.8060) grad_norm 9.2942 (8.5021/1.9904) mem 68106MB [2022-12-20 08:57:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][310/1519] eta 0:20:18 lr 0.000009 time 0.9300 (1.0082) model_time 0.9298 (1.0055) loss 0.6977 (0.8061) grad_norm 7.4932 (8.4716/1.9688) mem 68106MB [2022-12-20 08:57:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][320/1519] eta 0:20:08 lr 0.000009 time 0.9236 (1.0081) model_time 0.9235 (1.0055) loss 0.7106 (0.8075) grad_norm 12.5104 (8.5096/1.9820) mem 68106MB [2022-12-20 08:57:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][330/1519] eta 0:19:58 lr 0.000009 time 0.9231 (1.0081) model_time 0.9229 (1.0055) loss 0.6809 (0.8058) grad_norm 7.3791 (8.5031/1.9653) mem 68106MB [2022-12-20 08:57:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][340/1519] eta 0:19:48 lr 0.000009 time 0.9295 (1.0082) model_time 0.9293 (1.0058) loss 0.9325 (0.8064) grad_norm 8.0689 (8.4932/1.9667) mem 68106MB [2022-12-20 08:58:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][350/1519] eta 0:19:38 lr 0.000009 time 0.9217 (1.0081) model_time 0.9216 (1.0057) loss 1.2078 (0.8098) grad_norm 5.9096 (8.4828/1.9522) mem 68106MB [2022-12-20 08:58:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][360/1519] eta 0:19:28 lr 0.000009 time 0.9254 (1.0078) model_time 0.9253 (1.0055) loss 0.7475 (0.8120) grad_norm 6.9883 (8.4793/1.9426) mem 68106MB [2022-12-20 08:58:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][370/1519] eta 0:19:17 lr 0.000009 time 0.9384 (1.0078) model_time 0.9382 (1.0055) loss 0.6906 (0.8119) grad_norm 7.9125 (8.4599/1.9299) mem 68106MB [2022-12-20 08:58:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][380/1519] eta 0:19:07 lr 0.000009 time 0.9280 (1.0077) model_time 0.9279 (1.0055) loss 0.7149 (0.8108) grad_norm 7.5680 (8.4843/1.9314) mem 68106MB [2022-12-20 08:58:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][390/1519] eta 0:18:57 lr 0.000009 time 0.9327 (1.0075) model_time 0.9326 (1.0053) loss 0.7820 (0.8103) grad_norm 14.7794 (8.5254/1.9736) mem 68106MB [2022-12-20 08:58:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][400/1519] eta 0:18:47 lr 0.000009 time 0.9234 (1.0074) model_time 0.9232 (1.0052) loss 1.0645 (0.8095) grad_norm 10.0287 (8.5067/1.9633) mem 68106MB [2022-12-20 08:59:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][410/1519] eta 0:18:36 lr 0.000009 time 0.9291 (1.0072) model_time 0.9290 (1.0051) loss 0.6928 (0.8079) grad_norm 8.4083 (8.5145/1.9630) mem 68106MB [2022-12-20 08:59:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][420/1519] eta 0:18:26 lr 0.000009 time 0.9260 (1.0070) model_time 0.9259 (1.0049) loss 0.7223 (0.8066) grad_norm 6.6500 (8.4721/1.9616) mem 68106MB [2022-12-20 08:59:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][430/1519] eta 0:18:16 lr 0.000009 time 0.9017 (1.0069) model_time 0.9016 (1.0049) loss 0.7354 (0.8064) grad_norm 7.1772 (8.4661/1.9491) mem 68106MB [2022-12-20 08:59:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][440/1519] eta 0:18:06 lr 0.000009 time 0.9252 (1.0067) model_time 0.9251 (1.0047) loss 0.6643 (0.8067) grad_norm 8.0252 (8.4775/1.9324) mem 68106MB [2022-12-20 08:59:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][450/1519] eta 0:17:56 lr 0.000009 time 0.9260 (1.0066) model_time 0.9258 (1.0046) loss 0.8951 (0.8061) grad_norm 8.1522 (8.4668/1.9190) mem 68106MB [2022-12-20 08:59:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][460/1519] eta 0:17:45 lr 0.000009 time 0.9273 (1.0064) model_time 0.9271 (1.0045) loss 0.6823 (0.8061) grad_norm 7.9722 (8.4553/1.9084) mem 68106MB [2022-12-20 09:00:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][470/1519] eta 0:17:35 lr 0.000009 time 0.9232 (1.0065) model_time 0.9231 (1.0046) loss 0.8075 (0.8062) grad_norm 10.7905 (8.4674/1.8978) mem 68106MB [2022-12-20 09:00:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][480/1519] eta 0:17:25 lr 0.000009 time 0.9232 (1.0064) model_time 0.9231 (1.0045) loss 0.8266 (0.8062) grad_norm 12.6546 (8.4862/1.8997) mem 68106MB [2022-12-20 09:00:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][490/1519] eta 0:17:15 lr 0.000009 time 0.9495 (1.0064) model_time 0.9494 (1.0045) loss 0.7375 (0.8072) grad_norm 6.8620 (8.4881/1.8886) mem 68106MB [2022-12-20 09:00:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][500/1519] eta 0:17:05 lr 0.000009 time 0.9232 (1.0062) model_time 0.9231 (1.0044) loss 0.9166 (0.8080) grad_norm 6.5861 (8.5019/1.8872) mem 68106MB [2022-12-20 09:00:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][510/1519] eta 0:16:55 lr 0.000009 time 0.9280 (1.0063) model_time 0.9279 (1.0045) loss 0.9716 (0.8084) grad_norm 8.5226 (8.5263/1.9053) mem 68106MB [2022-12-20 09:00:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][520/1519] eta 0:16:45 lr 0.000009 time 0.9197 (1.0063) model_time 0.9196 (1.0046) loss 1.1762 (0.8097) grad_norm 9.2790 (8.5268/1.8906) mem 68106MB [2022-12-20 09:01:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][530/1519] eta 0:16:35 lr 0.000009 time 0.9765 (1.0064) model_time 0.9764 (1.0047) loss 0.9760 (0.8112) grad_norm 5.9615 (8.5125/1.8860) mem 68106MB [2022-12-20 09:01:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][540/1519] eta 0:16:25 lr 0.000009 time 0.9303 (1.0064) model_time 0.9302 (1.0047) loss 0.7659 (0.8105) grad_norm 10.4467 (8.5031/1.8803) mem 68106MB [2022-12-20 09:01:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][550/1519] eta 0:16:15 lr 0.000009 time 0.9227 (1.0062) model_time 0.9225 (1.0046) loss 0.8183 (0.8113) grad_norm 8.6389 (8.4825/1.8739) mem 68106MB [2022-12-20 09:01:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][560/1519] eta 0:16:05 lr 0.000009 time 0.9026 (1.0065) model_time 0.9024 (1.0049) loss 0.6849 (0.8117) grad_norm 10.0854 (8.4779/1.8652) mem 68106MB [2022-12-20 09:01:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][570/1519] eta 0:15:55 lr 0.000009 time 0.9317 (1.0064) model_time 0.9316 (1.0048) loss 0.6716 (0.8110) grad_norm 17.1563 (8.5003/1.9217) mem 68106MB [2022-12-20 09:01:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][580/1519] eta 0:15:45 lr 0.000009 time 1.0046 (1.0065) model_time 1.0044 (1.0049) loss 0.6907 (0.8112) grad_norm 7.8748 (8.5121/1.9327) mem 68106MB [2022-12-20 09:02:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][590/1519] eta 0:15:34 lr 0.000009 time 0.9237 (1.0063) model_time 0.9234 (1.0047) loss 0.7365 (0.8116) grad_norm 7.8014 (8.5343/1.9342) mem 68106MB [2022-12-20 09:02:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][600/1519] eta 0:15:24 lr 0.000009 time 0.9368 (1.0062) model_time 0.9367 (1.0047) loss 0.6717 (0.8114) grad_norm 9.9144 (8.5403/1.9214) mem 68106MB [2022-12-20 09:02:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][610/1519] eta 0:15:14 lr 0.000009 time 0.9249 (1.0064) model_time 0.9248 (1.0048) loss 1.2528 (0.8118) grad_norm 8.3727 (8.5393/1.9161) mem 68106MB [2022-12-20 09:02:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][620/1519] eta 0:15:04 lr 0.000009 time 0.9214 (1.0063) model_time 0.9212 (1.0048) loss 0.9846 (0.8126) grad_norm 11.5010 (8.5375/1.9241) mem 68106MB [2022-12-20 09:02:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][630/1519] eta 0:14:54 lr 0.000009 time 0.9132 (1.0063) model_time 0.9130 (1.0049) loss 0.7236 (0.8125) grad_norm 9.4771 (8.5382/1.9235) mem 68106MB [2022-12-20 09:02:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][640/1519] eta 0:14:44 lr 0.000009 time 0.9701 (1.0063) model_time 0.9700 (1.0048) loss 0.6967 (0.8127) grad_norm 7.6886 (8.5343/1.9308) mem 68106MB [2022-12-20 09:03:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][650/1519] eta 0:14:34 lr 0.000009 time 0.9303 (1.0062) model_time 0.9302 (1.0048) loss 1.0290 (0.8122) grad_norm 9.4177 (8.5056/1.9258) mem 68106MB [2022-12-20 09:03:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][660/1519] eta 0:14:24 lr 0.000009 time 0.9239 (1.0061) model_time 0.9237 (1.0047) loss 0.7650 (0.8122) grad_norm 6.0281 (8.4914/1.9079) mem 68106MB [2022-12-20 09:03:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][670/1519] eta 0:14:14 lr 0.000009 time 0.9254 (1.0060) model_time 0.9253 (1.0046) loss 0.7572 (0.8111) grad_norm 5.1136 (8.4853/1.9023) mem 68106MB [2022-12-20 09:03:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][680/1519] eta 0:14:04 lr 0.000009 time 0.9279 (1.0060) model_time 0.9277 (1.0046) loss 0.6545 (0.8102) grad_norm 10.5385 (8.4673/1.8656) mem 68106MB [2022-12-20 09:03:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][690/1519] eta 0:13:53 lr 0.000009 time 0.9292 (1.0059) model_time 0.9291 (1.0045) loss 0.9181 (0.8114) grad_norm 7.1288 (8.4633/1.8607) mem 68106MB [2022-12-20 09:03:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][700/1519] eta 0:13:43 lr 0.000009 time 0.9274 (1.0058) model_time 0.9272 (1.0044) loss 0.6642 (0.8110) grad_norm 9.0983 (8.4713/1.8696) mem 68106MB [2022-12-20 09:04:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][710/1519] eta 0:13:33 lr 0.000009 time 0.9862 (1.0058) model_time 0.9860 (1.0044) loss 0.7887 (0.8115) grad_norm 9.1525 (8.4909/1.8740) mem 68106MB [2022-12-20 09:04:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][720/1519] eta 0:13:23 lr 0.000009 time 0.9201 (1.0057) model_time 0.9199 (1.0043) loss 0.7433 (0.8114) grad_norm 8.7722 (8.4783/1.8643) mem 68106MB [2022-12-20 09:04:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][730/1519] eta 0:13:13 lr 0.000009 time 0.9255 (1.0056) model_time 0.9253 (1.0043) loss 0.9296 (0.8117) grad_norm 10.5116 (8.4839/1.8817) mem 68106MB [2022-12-20 09:04:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][740/1519] eta 0:13:03 lr 0.000009 time 0.9259 (1.0058) model_time 0.9258 (1.0045) loss 0.7292 (0.8106) grad_norm 7.1124 (8.5054/1.8886) mem 68106MB [2022-12-20 09:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][750/1519] eta 0:12:53 lr 0.000009 time 0.9211 (1.0057) model_time 0.9210 (1.0044) loss 0.8327 (0.8100) grad_norm 9.6070 (8.5287/1.8992) mem 68106MB [2022-12-20 09:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][760/1519] eta 0:12:43 lr 0.000009 time 0.9196 (1.0056) model_time 0.9195 (1.0043) loss 0.8118 (0.8105) grad_norm 8.6571 (8.5465/1.8901) mem 68106MB [2022-12-20 09:05:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][770/1519] eta 0:12:33 lr 0.000009 time 0.9226 (1.0055) model_time 0.9225 (1.0042) loss 0.7419 (0.8106) grad_norm 7.2710 (8.5408/1.8911) mem 68106MB [2022-12-20 09:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][780/1519] eta 0:12:23 lr 0.000009 time 1.1325 (1.0057) model_time 1.1324 (1.0045) loss 0.8517 (0.8106) grad_norm 7.6681 (8.5425/1.8809) mem 68106MB [2022-12-20 09:05:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][790/1519] eta 0:12:13 lr 0.000009 time 0.9188 (1.0057) model_time 0.9187 (1.0044) loss 0.8029 (0.8101) grad_norm 11.9313 (8.5520/1.8936) mem 68106MB [2022-12-20 09:05:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][800/1519] eta 0:12:03 lr 0.000009 time 0.9237 (1.0057) model_time 0.9235 (1.0044) loss 0.8648 (0.8096) grad_norm 8.6774 (8.5695/1.9191) mem 68106MB [2022-12-20 09:05:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][810/1519] eta 0:11:52 lr 0.000009 time 0.9213 (1.0056) model_time 0.9212 (1.0044) loss 0.6653 (0.8097) grad_norm 7.5450 (8.5393/1.8978) mem 68106MB [2022-12-20 09:05:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][820/1519] eta 0:11:43 lr 0.000009 time 0.9026 (1.0058) model_time 0.9025 (1.0046) loss 0.6855 (0.8091) grad_norm 7.3877 (8.5328/1.8643) mem 68106MB [2022-12-20 09:06:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][830/1519] eta 0:11:33 lr 0.000009 time 0.9305 (1.0060) model_time 0.9304 (1.0048) loss 0.8342 (0.8087) grad_norm 7.0789 (8.5275/1.8720) mem 68106MB [2022-12-20 09:06:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][840/1519] eta 0:11:23 lr 0.000009 time 0.9806 (1.0060) model_time 0.9804 (1.0048) loss 0.6758 (0.8086) grad_norm 8.3147 (8.5422/1.8711) mem 68106MB [2022-12-20 09:06:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][850/1519] eta 0:11:13 lr 0.000009 time 1.1844 (1.0062) model_time 1.1842 (1.0050) loss 0.7135 (0.8087) grad_norm 13.7406 (8.5339/1.8915) mem 68106MB [2022-12-20 09:06:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][860/1519] eta 0:11:03 lr 0.000009 time 0.9278 (1.0061) model_time 0.9277 (1.0050) loss 0.8977 (0.8087) grad_norm 8.2879 (8.5349/1.8774) mem 68106MB [2022-12-20 09:06:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][870/1519] eta 0:10:52 lr 0.000009 time 0.9037 (1.0061) model_time 0.9036 (1.0049) loss 0.7254 (0.8080) grad_norm 7.7850 (8.5602/1.8842) mem 68106MB [2022-12-20 09:06:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][880/1519] eta 0:10:42 lr 0.000009 time 0.9233 (1.0061) model_time 0.9231 (1.0049) loss 0.7289 (0.8075) grad_norm 8.6400 (8.5217/1.8305) mem 68106MB [2022-12-20 09:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][890/1519] eta 0:10:32 lr 0.000009 time 0.9964 (1.0060) model_time 0.9963 (1.0049) loss 0.7554 (0.8078) grad_norm 6.4783 (8.5134/1.8349) mem 68106MB [2022-12-20 09:07:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][900/1519] eta 0:10:22 lr 0.000009 time 0.9233 (1.0060) model_time 0.9232 (1.0049) loss 0.8580 (0.8078) grad_norm 9.5169 (8.5080/1.8267) mem 68106MB [2022-12-20 09:07:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][910/1519] eta 0:10:12 lr 0.000009 time 0.9265 (1.0059) model_time 0.9262 (1.0048) loss 0.8339 (0.8082) grad_norm 6.5618 (8.5194/1.8380) mem 68106MB [2022-12-20 09:07:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][920/1519] eta 0:10:02 lr 0.000009 time 0.9624 (1.0059) model_time 0.9623 (1.0047) loss 0.8996 (0.8081) grad_norm 6.0293 (8.5011/1.8283) mem 68106MB [2022-12-20 09:07:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][930/1519] eta 0:09:52 lr 0.000009 time 0.9210 (1.0058) model_time 0.9208 (1.0047) loss 0.7679 (0.8077) grad_norm 7.7538 (8.5039/1.8373) mem 68106MB [2022-12-20 09:07:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][940/1519] eta 0:09:42 lr 0.000009 time 0.9248 (1.0058) model_time 0.9247 (1.0047) loss 1.0652 (0.8082) grad_norm 7.8179 (8.5055/1.8256) mem 68106MB [2022-12-20 09:08:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][950/1519] eta 0:09:32 lr 0.000009 time 0.9228 (1.0057) model_time 0.9226 (1.0046) loss 0.6797 (0.8077) grad_norm 8.5390 (8.5199/1.8247) mem 68106MB [2022-12-20 09:08:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][960/1519] eta 0:09:22 lr 0.000009 time 1.0459 (1.0059) model_time 1.0458 (1.0048) loss 0.8958 (0.8078) grad_norm 11.4371 (8.5267/1.8251) mem 68106MB [2022-12-20 09:08:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][970/1519] eta 0:09:12 lr 0.000009 time 0.9215 (1.0058) model_time 0.9214 (1.0047) loss 0.9329 (0.8080) grad_norm 6.9975 (8.5286/1.8295) mem 68106MB [2022-12-20 09:08:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][980/1519] eta 0:09:02 lr 0.000009 time 0.9194 (1.0057) model_time 0.9193 (1.0046) loss 1.1928 (0.8082) grad_norm 10.3816 (8.5190/1.8177) mem 68106MB [2022-12-20 09:08:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][990/1519] eta 0:08:52 lr 0.000009 time 0.9762 (1.0057) model_time 0.9760 (1.0047) loss 0.6725 (0.8084) grad_norm 7.4244 (8.4767/1.7909) mem 68106MB [2022-12-20 09:08:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1000/1519] eta 0:08:41 lr 0.000009 time 0.9320 (1.0057) model_time 0.9318 (1.0046) loss 0.8452 (0.8087) grad_norm 7.7482 (8.4817/1.7852) mem 68106MB [2022-12-20 09:09:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1010/1519] eta 0:08:31 lr 0.000009 time 0.9220 (1.0056) model_time 0.9219 (1.0045) loss 0.8371 (0.8088) grad_norm 10.9875 (8.4625/1.7843) mem 68106MB [2022-12-20 09:09:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1020/1519] eta 0:08:21 lr 0.000009 time 1.0226 (1.0056) model_time 1.0223 (1.0046) loss 0.7044 (0.8089) grad_norm 8.9014 (8.4995/1.7739) mem 68106MB [2022-12-20 09:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1030/1519] eta 0:08:11 lr 0.000009 time 0.9292 (1.0056) model_time 0.9290 (1.0046) loss 0.7406 (0.8084) grad_norm 15.5856 (8.5177/1.8196) mem 68106MB [2022-12-20 09:09:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1040/1519] eta 0:08:01 lr 0.000009 time 0.9177 (1.0055) model_time 0.9175 (1.0045) loss 1.0026 (0.8088) grad_norm 8.6480 (8.5055/1.8175) mem 68106MB [2022-12-20 09:09:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1050/1519] eta 0:07:51 lr 0.000009 time 0.9286 (1.0056) model_time 0.9284 (1.0045) loss 0.6775 (0.8088) grad_norm 7.9180 (8.5035/1.8140) mem 68106MB [2022-12-20 09:09:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1060/1519] eta 0:07:41 lr 0.000009 time 0.9240 (1.0055) model_time 0.9238 (1.0045) loss 0.6709 (0.8085) grad_norm 7.8009 (8.5339/1.8318) mem 68106MB [2022-12-20 09:10:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1070/1519] eta 0:07:31 lr 0.000009 time 0.9860 (1.0055) model_time 0.9858 (1.0045) loss 0.8938 (0.8083) grad_norm 7.8457 (8.5345/1.8289) mem 68106MB [2022-12-20 09:10:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1080/1519] eta 0:07:21 lr 0.000009 time 0.9265 (1.0055) model_time 0.9261 (1.0045) loss 0.6941 (0.8078) grad_norm 8.5953 (8.5072/1.8192) mem 68106MB [2022-12-20 09:10:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1090/1519] eta 0:07:11 lr 0.000009 time 0.9198 (1.0054) model_time 0.9196 (1.0044) loss 0.8301 (0.8085) grad_norm 7.4059 (8.4887/1.8265) mem 68106MB [2022-12-20 09:10:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1100/1519] eta 0:07:01 lr 0.000009 time 0.9317 (1.0054) model_time 0.9315 (1.0044) loss 0.8783 (0.8078) grad_norm 8.4869 (8.4701/1.8174) mem 68106MB [2022-12-20 09:10:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1110/1519] eta 0:06:51 lr 0.000009 time 0.8867 (1.0056) model_time 0.8866 (1.0046) loss 0.6888 (0.8077) grad_norm 8.1876 (8.4551/1.7884) mem 68106MB [2022-12-20 09:10:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1120/1519] eta 0:06:41 lr 0.000009 time 0.9407 (1.0056) model_time 0.9405 (1.0046) loss 0.6727 (0.8084) grad_norm 7.4002 (8.4496/1.8014) mem 68106MB [2022-12-20 09:11:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1130/1519] eta 0:06:31 lr 0.000009 time 0.9834 (1.0055) model_time 0.9833 (1.0046) loss 1.2238 (0.8094) grad_norm 5.1551 (8.4385/1.8031) mem 68106MB [2022-12-20 09:11:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1140/1519] eta 0:06:21 lr 0.000009 time 0.9316 (1.0056) model_time 0.9314 (1.0046) loss 0.7260 (0.8095) grad_norm 9.3145 (8.4646/1.8090) mem 68106MB [2022-12-20 09:11:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1150/1519] eta 0:06:11 lr 0.000009 time 0.9081 (1.0057) model_time 0.9075 (1.0047) loss 0.8273 (0.8104) grad_norm 17.4987 (8.5200/1.8886) mem 68106MB [2022-12-20 09:11:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1160/1519] eta 0:06:00 lr 0.000009 time 0.9267 (1.0056) model_time 0.9266 (1.0046) loss 0.9558 (0.8101) grad_norm 7.5393 (8.5302/1.8867) mem 68106MB [2022-12-20 09:11:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1170/1519] eta 0:05:50 lr 0.000009 time 0.9844 (1.0056) model_time 0.9842 (1.0047) loss 0.6808 (0.8106) grad_norm 9.5059 (8.5299/1.8339) mem 68106MB [2022-12-20 09:11:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1180/1519] eta 0:05:40 lr 0.000009 time 0.9316 (1.0056) model_time 0.9314 (1.0046) loss 0.7061 (0.8109) grad_norm 7.6382 (8.5123/1.8118) mem 68106MB [2022-12-20 09:12:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1190/1519] eta 0:05:30 lr 0.000009 time 0.9457 (1.0056) model_time 0.9456 (1.0046) loss 0.6714 (0.8111) grad_norm 5.8927 (8.5074/1.8421) mem 68106MB [2022-12-20 09:12:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1200/1519] eta 0:05:20 lr 0.000009 time 0.9302 (1.0055) model_time 0.9301 (1.0046) loss 0.8397 (0.8119) grad_norm 8.4360 (8.5062/1.8748) mem 68106MB [2022-12-20 09:12:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1210/1519] eta 0:05:10 lr 0.000009 time 0.9253 (1.0055) model_time 0.9252 (1.0046) loss 0.7769 (0.8113) grad_norm 7.1313 (8.5017/1.8818) mem 68106MB [2022-12-20 09:12:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1220/1519] eta 0:05:00 lr 0.000009 time 0.9229 (1.0055) model_time 0.9228 (1.0045) loss 0.8113 (0.8117) grad_norm 9.9767 (8.5018/1.8689) mem 68106MB [2022-12-20 09:12:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1230/1519] eta 0:04:50 lr 0.000009 time 0.9239 (1.0054) model_time 0.9237 (1.0045) loss 0.8249 (0.8115) grad_norm 8.4747 (8.4911/1.8676) mem 68106MB [2022-12-20 09:12:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1240/1519] eta 0:04:40 lr 0.000009 time 0.9276 (1.0055) model_time 0.9274 (1.0045) loss 0.6627 (0.8119) grad_norm 10.6216 (8.5085/1.8651) mem 68106MB [2022-12-20 09:13:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1250/1519] eta 0:04:30 lr 0.000009 time 0.9363 (1.0054) model_time 0.9361 (1.0045) loss 0.8546 (0.8120) grad_norm 13.5353 (8.5266/1.8822) mem 68106MB [2022-12-20 09:13:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1260/1519] eta 0:04:20 lr 0.000009 time 0.9331 (1.0054) model_time 0.9330 (1.0045) loss 1.0046 (0.8124) grad_norm 8.0148 (8.5018/1.8437) mem 68106MB [2022-12-20 09:13:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1270/1519] eta 0:04:10 lr 0.000009 time 0.9327 (1.0054) model_time 0.9325 (1.0045) loss 0.9157 (0.8122) grad_norm 6.8901 (8.5112/1.8325) mem 68106MB [2022-12-20 09:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1280/1519] eta 0:04:00 lr 0.000009 time 0.9153 (1.0054) model_time 0.9152 (1.0045) loss 0.7030 (0.8117) grad_norm 8.3402 (8.5314/1.8781) mem 68106MB [2022-12-20 09:13:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1290/1519] eta 0:03:50 lr 0.000009 time 0.9314 (1.0054) model_time 0.9312 (1.0045) loss 1.0070 (0.8121) grad_norm 7.4997 (8.5488/1.8802) mem 68106MB [2022-12-20 09:13:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1300/1519] eta 0:03:40 lr 0.000009 time 0.9191 (1.0054) model_time 0.9189 (1.0044) loss 0.7086 (0.8122) grad_norm 8.0661 (8.5624/1.9006) mem 68106MB [2022-12-20 09:14:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1310/1519] eta 0:03:30 lr 0.000009 time 1.0098 (1.0054) model_time 1.0097 (1.0045) loss 0.9594 (0.8124) grad_norm 10.8173 (8.5890/1.9126) mem 68106MB [2022-12-20 09:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1320/1519] eta 0:03:20 lr 0.000009 time 0.9287 (1.0053) model_time 0.9286 (1.0044) loss 0.7571 (0.8123) grad_norm 9.2561 (8.5938/1.9108) mem 68106MB [2022-12-20 09:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1330/1519] eta 0:03:09 lr 0.000009 time 0.9275 (1.0053) model_time 0.9273 (1.0044) loss 1.3984 (0.8131) grad_norm 10.4666 (8.5956/1.8975) mem 68106MB [2022-12-20 09:14:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1340/1519] eta 0:02:59 lr 0.000009 time 0.9273 (1.0054) model_time 0.9272 (1.0046) loss 0.8504 (0.8129) grad_norm 8.7932 (8.5746/1.8944) mem 68106MB [2022-12-20 09:14:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1350/1519] eta 0:02:49 lr 0.000009 time 0.9283 (1.0056) model_time 0.9282 (1.0047) loss 0.6634 (0.8130) grad_norm 7.2797 (8.5221/1.8683) mem 68106MB [2022-12-20 09:14:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1360/1519] eta 0:02:39 lr 0.000009 time 0.9227 (1.0056) model_time 0.9225 (1.0048) loss 0.9736 (0.8132) grad_norm 8.4984 (8.5316/1.8698) mem 68106MB [2022-12-20 09:15:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1370/1519] eta 0:02:29 lr 0.000009 time 0.9305 (1.0056) model_time 0.9303 (1.0048) loss 0.7679 (0.8130) grad_norm 6.8558 (8.5328/1.8708) mem 68106MB [2022-12-20 09:15:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1380/1519] eta 0:02:19 lr 0.000009 time 0.9315 (1.0056) model_time 0.9313 (1.0047) loss 0.6789 (0.8135) grad_norm 7.8832 (8.5271/1.8591) mem 68106MB [2022-12-20 09:15:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1390/1519] eta 0:02:09 lr 0.000009 time 0.9340 (1.0056) model_time 0.9339 (1.0048) loss 0.7549 (0.8139) grad_norm 11.5556 (8.5469/1.8520) mem 68106MB [2022-12-20 09:15:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1400/1519] eta 0:01:59 lr 0.000009 time 0.9340 (1.0056) model_time 0.9339 (1.0047) loss 0.6908 (0.8138) grad_norm 12.7856 (8.5337/1.8372) mem 68106MB [2022-12-20 09:15:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1410/1519] eta 0:01:49 lr 0.000009 time 0.9315 (1.0055) model_time 0.9313 (1.0047) loss 1.2765 (0.8143) grad_norm 8.5865 (8.5695/1.8484) mem 68106MB [2022-12-20 09:15:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1420/1519] eta 0:01:39 lr 0.000009 time 0.9277 (1.0056) model_time 0.9274 (1.0047) loss 0.9895 (0.8150) grad_norm 22.5498 (8.6106/2.0212) mem 68106MB [2022-12-20 09:16:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1430/1519] eta 0:01:29 lr 0.000009 time 0.9369 (1.0056) model_time 0.9368 (1.0047) loss 0.7673 (0.8149) grad_norm 6.7189 (8.6346/2.0191) mem 68106MB [2022-12-20 09:16:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1440/1519] eta 0:01:19 lr 0.000009 time 0.9214 (1.0056) model_time 0.9212 (1.0047) loss 0.9494 (0.8150) grad_norm 12.7143 (8.6496/2.0376) mem 68106MB [2022-12-20 09:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1450/1519] eta 0:01:09 lr 0.000009 time 0.9672 (1.0056) model_time 0.9670 (1.0048) loss 0.7874 (0.8151) grad_norm 10.5961 (8.6428/2.0101) mem 68106MB [2022-12-20 09:16:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1460/1519] eta 0:00:59 lr 0.000009 time 0.9310 (1.0057) model_time 0.9307 (1.0048) loss 0.8078 (0.8149) grad_norm 7.0689 (8.6312/2.0117) mem 68106MB [2022-12-20 09:16:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1470/1519] eta 0:00:49 lr 0.000009 time 0.9272 (1.0056) model_time 0.9270 (1.0048) loss 0.7213 (0.8145) grad_norm 7.9537 (8.5877/1.9980) mem 68106MB [2022-12-20 09:16:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1480/1519] eta 0:00:39 lr 0.000009 time 0.9317 (1.0056) model_time 0.9316 (1.0047) loss 0.7677 (0.8144) grad_norm 9.2806 (8.6105/2.0022) mem 68106MB [2022-12-20 09:17:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1490/1519] eta 0:00:29 lr 0.000009 time 0.9189 (1.0056) model_time 0.9187 (1.0048) loss 0.7645 (0.8146) grad_norm 8.8807 (8.6408/2.0363) mem 68106MB [2022-12-20 09:17:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1500/1519] eta 0:00:19 lr 0.000009 time 0.9274 (1.0056) model_time 0.9273 (1.0048) loss 0.9653 (0.8145) grad_norm 8.1806 (8.6602/2.0572) mem 68106MB [2022-12-20 09:17:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [65/100][1510/1519] eta 0:00:09 lr 0.000009 time 0.9293 (1.0056) model_time 0.9292 (1.0047) loss 0.7101 (0.8146) grad_norm 8.3398 (8.6817/2.0652) mem 68106MB [2022-12-20 09:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 65 training takes 0:25:27 [2022-12-20 09:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_65.pth saving...... [2022-12-20 09:17:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_65.pth saved !!! [2022-12-20 09:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.661 (0.661) Loss 0.5337 (0.5337) Acc@1 91.319 (91.319) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 09:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.308 (0.331) Loss 0.5264 (0.5051) Acc@1 92.708 (92.456) Acc@5 97.917 (98.453) Mem 68106MB [2022-12-20 09:18:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.317) Loss 0.4887 (0.5024) Acc@1 91.667 (92.543) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 09:18:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.311) Loss 0.6220 (0.5085) Acc@1 90.625 (92.339) Acc@5 97.917 (98.376) Mem 68106MB [2022-12-20 09:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.308) Loss 0.4623 (0.4992) Acc@1 93.403 (92.437) Acc@5 99.306 (98.493) Mem 68106MB [2022-12-20 09:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.307) Loss 0.4924 (0.4965) Acc@1 91.667 (92.470) Acc@5 99.306 (98.550) Mem 68106MB [2022-12-20 09:18:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.5129 (0.4964) Acc@1 91.319 (92.424) Acc@5 98.264 (98.537) Mem 68106MB [2022-12-20 09:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5489 (0.4977) Acc@1 92.361 (92.361) Acc@5 97.917 (98.523) Mem 68106MB [2022-12-20 09:18:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.304) Loss 0.4411 (0.4959) Acc@1 92.708 (92.374) Acc@5 98.264 (98.560) Mem 68106MB [2022-12-20 09:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:65] * Acc@1 92.342 Acc@5 98.559 [2022-12-20 09:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.3% [2022-12-20 09:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.39% [2022-12-20 09:18:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][0/1519] eta 0:46:15 lr 0.000009 time 1.8269 (1.8269) model_time 1.0422 (1.0422) loss 0.6831 (0.6831) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 09:18:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][10/1519] eta 0:27:01 lr 0.000009 time 0.9317 (1.0745) model_time 0.9315 (1.0028) loss 0.8103 (0.7714) grad_norm 11.5147 (8.8498/2.3334) mem 68106MB [2022-12-20 09:18:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][20/1519] eta 0:26:02 lr 0.000009 time 1.0017 (1.0422) model_time 1.0016 (1.0044) loss 1.0436 (0.8037) grad_norm 8.2020 (8.5169/1.7150) mem 68106MB [2022-12-20 09:18:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][30/1519] eta 0:25:31 lr 0.000009 time 0.9293 (1.0286) model_time 0.9292 (1.0029) loss 0.6743 (0.8340) grad_norm 7.9824 (8.3533/1.9305) mem 68106MB [2022-12-20 09:19:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][40/1519] eta 0:25:14 lr 0.000009 time 0.9331 (1.0243) model_time 0.9329 (1.0047) loss 0.9876 (0.8330) grad_norm 9.2370 (8.2959/1.7287) mem 68106MB [2022-12-20 09:19:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][50/1519] eta 0:24:57 lr 0.000009 time 0.9334 (1.0194) model_time 0.9333 (1.0036) loss 0.7820 (0.8307) grad_norm 12.9651 (8.2813/1.9006) mem 68106MB [2022-12-20 09:19:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][60/1519] eta 0:24:41 lr 0.000009 time 0.9322 (1.0156) model_time 0.9321 (1.0024) loss 0.9154 (0.8327) grad_norm 10.9912 (8.3205/1.8629) mem 68106MB [2022-12-20 09:19:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][70/1519] eta 0:24:29 lr 0.000009 time 0.9339 (1.0138) model_time 0.9337 (1.0024) loss 0.9181 (0.8334) grad_norm 10.6263 (8.4857/1.8026) mem 68106MB [2022-12-20 09:19:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][80/1519] eta 0:24:17 lr 0.000009 time 0.9302 (1.0128) model_time 0.9300 (1.0027) loss 1.5530 (0.8319) grad_norm 7.2382 (8.3364/1.7630) mem 68106MB [2022-12-20 09:19:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][90/1519] eta 0:24:06 lr 0.000009 time 0.9954 (1.0122) model_time 0.9952 (1.0032) loss 0.7629 (0.8255) grad_norm 8.0803 (8.3022/1.8148) mem 68106MB [2022-12-20 09:20:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][100/1519] eta 0:23:57 lr 0.000009 time 1.0076 (1.0128) model_time 1.0074 (1.0046) loss 0.7037 (0.8292) grad_norm 8.2721 (8.2573/1.7778) mem 68106MB [2022-12-20 09:20:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][110/1519] eta 0:23:46 lr 0.000009 time 0.9292 (1.0121) model_time 0.9290 (1.0046) loss 0.7201 (0.8269) grad_norm 6.4649 (8.2090/1.7218) mem 68106MB [2022-12-20 09:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][120/1519] eta 0:23:35 lr 0.000009 time 1.0225 (1.0119) model_time 1.0223 (1.0050) loss 0.7049 (0.8299) grad_norm 7.3151 (8.2024/1.6998) mem 68106MB [2022-12-20 09:20:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][130/1519] eta 0:23:24 lr 0.000009 time 0.9291 (1.0113) model_time 0.9289 (1.0049) loss 0.7448 (0.8290) grad_norm 8.4274 (8.1774/1.6554) mem 68106MB [2022-12-20 09:20:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][140/1519] eta 0:23:13 lr 0.000009 time 0.9293 (1.0107) model_time 0.9290 (1.0048) loss 0.7488 (0.8275) grad_norm 8.1587 (8.1849/1.7000) mem 68106MB [2022-12-20 09:20:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][150/1519] eta 0:23:02 lr 0.000009 time 0.9271 (1.0098) model_time 0.9269 (1.0042) loss 0.9589 (0.8284) grad_norm 6.2204 (8.1665/1.6690) mem 68106MB [2022-12-20 09:21:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][160/1519] eta 0:22:52 lr 0.000009 time 0.9362 (1.0101) model_time 0.9360 (1.0048) loss 0.8065 (0.8310) grad_norm 7.5765 (8.1804/1.6532) mem 68106MB [2022-12-20 09:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][170/1519] eta 0:22:44 lr 0.000009 time 0.9413 (1.0111) model_time 0.9411 (1.0061) loss 0.7568 (0.8311) grad_norm 10.0757 (8.2795/1.7787) mem 68106MB [2022-12-20 09:21:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][180/1519] eta 0:22:32 lr 0.000009 time 0.9303 (1.0103) model_time 0.9301 (1.0056) loss 0.6909 (0.8283) grad_norm 5.2640 (8.1649/1.8014) mem 68106MB [2022-12-20 09:21:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][190/1519] eta 0:22:22 lr 0.000009 time 0.9154 (1.0099) model_time 0.9153 (1.0053) loss 0.7647 (0.8279) grad_norm 8.7546 (8.1544/1.7705) mem 68106MB [2022-12-20 09:21:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][200/1519] eta 0:22:11 lr 0.000009 time 1.0278 (1.0098) model_time 1.0277 (1.0055) loss 1.0819 (0.8271) grad_norm 6.4122 (8.1565/1.7411) mem 68106MB [2022-12-20 09:21:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][210/1519] eta 0:22:01 lr 0.000009 time 0.9550 (1.0097) model_time 0.9549 (1.0055) loss 0.8094 (0.8264) grad_norm 8.2850 (8.1736/1.7127) mem 68106MB [2022-12-20 09:22:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][220/1519] eta 0:21:50 lr 0.000009 time 0.9292 (1.0092) model_time 0.9290 (1.0052) loss 0.8977 (0.8282) grad_norm 6.9827 (8.1423/1.6853) mem 68106MB [2022-12-20 09:22:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][230/1519] eta 0:21:40 lr 0.000009 time 0.9291 (1.0091) model_time 0.9290 (1.0053) loss 0.8756 (0.8261) grad_norm 7.1584 (8.2245/1.7409) mem 68106MB [2022-12-20 09:22:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][240/1519] eta 0:21:30 lr 0.000009 time 0.9304 (1.0087) model_time 0.9302 (1.0051) loss 0.6898 (0.8269) grad_norm 9.9738 (8.2402/1.7385) mem 68106MB [2022-12-20 09:22:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][250/1519] eta 0:21:19 lr 0.000009 time 0.9265 (1.0086) model_time 0.9263 (1.0051) loss 0.7957 (0.8245) grad_norm 13.5005 (8.2624/1.7754) mem 68106MB [2022-12-20 09:22:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][260/1519] eta 0:21:09 lr 0.000009 time 0.9286 (1.0082) model_time 0.9283 (1.0048) loss 0.8119 (0.8219) grad_norm 9.9488 (8.2911/1.7520) mem 68106MB [2022-12-20 09:22:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][270/1519] eta 0:20:59 lr 0.000009 time 0.9963 (1.0081) model_time 0.9961 (1.0048) loss 0.6723 (0.8204) grad_norm 9.5200 (8.3667/1.8339) mem 68106MB [2022-12-20 09:23:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][280/1519] eta 0:20:48 lr 0.000009 time 0.9306 (1.0080) model_time 0.9305 (1.0048) loss 0.7782 (0.8202) grad_norm 10.1436 (8.3583/1.8130) mem 68106MB [2022-12-20 09:23:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][290/1519] eta 0:20:39 lr 0.000009 time 0.9364 (1.0086) model_time 0.9362 (1.0055) loss 0.6576 (0.8233) grad_norm 10.1395 (8.3593/1.7920) mem 68106MB [2022-12-20 09:23:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][300/1519] eta 0:20:29 lr 0.000009 time 0.9280 (1.0083) model_time 0.9278 (1.0053) loss 0.7326 (0.8216) grad_norm 6.2708 (8.3399/1.7741) mem 68106MB [2022-12-20 09:23:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][310/1519] eta 0:20:19 lr 0.000009 time 0.9221 (1.0085) model_time 0.9220 (1.0056) loss 1.0164 (0.8212) grad_norm 8.8140 (8.3328/1.7510) mem 68106MB [2022-12-20 09:23:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][320/1519] eta 0:20:08 lr 0.000009 time 0.9205 (1.0082) model_time 0.9203 (1.0053) loss 0.7859 (0.8212) grad_norm 5.8903 (8.3234/1.7418) mem 68106MB [2022-12-20 09:23:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][330/1519] eta 0:19:58 lr 0.000009 time 0.9263 (1.0079) model_time 0.9257 (1.0051) loss 0.8252 (0.8220) grad_norm 8.6845 (8.3521/1.7252) mem 68106MB [2022-12-20 09:24:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][340/1519] eta 0:19:48 lr 0.000009 time 0.9279 (1.0079) model_time 0.9276 (1.0052) loss 0.8253 (0.8258) grad_norm 7.4448 (8.3170/1.7150) mem 68106MB [2022-12-20 09:24:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][350/1519] eta 0:19:38 lr 0.000009 time 0.9243 (1.0081) model_time 0.9242 (1.0054) loss 0.6881 (0.8235) grad_norm 8.3026 (8.2911/1.7127) mem 68106MB [2022-12-20 09:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][360/1519] eta 0:19:28 lr 0.000009 time 0.9388 (1.0078) model_time 0.9386 (1.0052) loss 0.8136 (0.8224) grad_norm 10.1844 (8.2580/1.7266) mem 68106MB [2022-12-20 09:24:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][370/1519] eta 0:19:17 lr 0.000009 time 0.9329 (1.0075) model_time 0.9328 (1.0050) loss 0.6668 (0.8223) grad_norm 10.7136 (8.2915/1.7411) mem 68106MB [2022-12-20 09:24:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][380/1519] eta 0:19:07 lr 0.000009 time 0.9278 (1.0072) model_time 0.9276 (1.0048) loss 0.6956 (0.8216) grad_norm 11.1403 (8.3221/1.7960) mem 68106MB [2022-12-20 09:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][390/1519] eta 0:18:57 lr 0.000009 time 0.9304 (1.0072) model_time 0.9302 (1.0048) loss 0.6822 (0.8219) grad_norm 9.1764 (8.3383/1.7901) mem 68106MB [2022-12-20 09:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][400/1519] eta 0:18:46 lr 0.000009 time 0.9345 (1.0070) model_time 0.9344 (1.0046) loss 1.0260 (0.8226) grad_norm 6.8196 (8.3451/1.8219) mem 68106MB [2022-12-20 09:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][410/1519] eta 0:18:36 lr 0.000009 time 0.9326 (1.0071) model_time 0.9325 (1.0048) loss 0.7730 (0.8236) grad_norm 7.4459 (8.3399/1.8336) mem 68106MB [2022-12-20 09:25:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][420/1519] eta 0:18:26 lr 0.000009 time 0.9436 (1.0072) model_time 0.9434 (1.0049) loss 1.1523 (0.8253) grad_norm 11.0654 (8.3482/1.8291) mem 68106MB [2022-12-20 09:25:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][430/1519] eta 0:18:16 lr 0.000009 time 0.9327 (1.0070) model_time 0.9325 (1.0048) loss 0.6843 (0.8238) grad_norm 5.5729 (8.3447/1.8333) mem 68106MB [2022-12-20 09:25:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][440/1519] eta 0:18:06 lr 0.000009 time 0.9286 (1.0069) model_time 0.9285 (1.0047) loss 0.7918 (0.8238) grad_norm 7.8280 (8.3240/1.8236) mem 68106MB [2022-12-20 09:25:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][450/1519] eta 0:17:56 lr 0.000009 time 0.9090 (1.0069) model_time 0.9088 (1.0048) loss 0.8046 (0.8235) grad_norm 8.1926 (8.3119/1.8185) mem 68106MB [2022-12-20 09:26:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][460/1519] eta 0:17:46 lr 0.000009 time 0.9296 (1.0067) model_time 0.9295 (1.0046) loss 0.9116 (0.8230) grad_norm 8.5659 (8.2863/1.8134) mem 68106MB [2022-12-20 09:26:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][470/1519] eta 0:17:35 lr 0.000009 time 0.9306 (1.0066) model_time 0.9305 (1.0046) loss 0.7728 (0.8237) grad_norm 8.0517 (8.2661/1.8030) mem 68106MB [2022-12-20 09:26:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][480/1519] eta 0:17:25 lr 0.000009 time 0.9294 (1.0066) model_time 0.9293 (1.0046) loss 0.6817 (0.8240) grad_norm 9.0451 (8.2576/1.7915) mem 68106MB [2022-12-20 09:26:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][490/1519] eta 0:17:15 lr 0.000009 time 0.9315 (1.0065) model_time 0.9313 (1.0045) loss 0.7664 (0.8245) grad_norm 8.0496 (8.2512/1.7771) mem 68106MB [2022-12-20 09:26:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][500/1519] eta 0:17:05 lr 0.000009 time 0.9307 (1.0065) model_time 0.9306 (1.0045) loss 0.8893 (0.8242) grad_norm 12.1486 (8.2717/1.7834) mem 68106MB [2022-12-20 09:26:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][510/1519] eta 0:16:55 lr 0.000009 time 0.9287 (1.0063) model_time 0.9286 (1.0044) loss 0.6705 (0.8244) grad_norm 9.4376 (8.2918/1.7783) mem 68106MB [2022-12-20 09:27:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][520/1519] eta 0:16:45 lr 0.000009 time 0.9334 (1.0065) model_time 0.9332 (1.0047) loss 0.8895 (0.8261) grad_norm 6.1551 (8.2910/1.7809) mem 68106MB [2022-12-20 09:27:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][530/1519] eta 0:16:35 lr 0.000009 time 0.9290 (1.0064) model_time 0.9289 (1.0046) loss 0.7006 (0.8265) grad_norm 6.6810 (8.2784/1.7735) mem 68106MB [2022-12-20 09:27:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][540/1519] eta 0:16:25 lr 0.000009 time 0.9289 (1.0062) model_time 0.9287 (1.0044) loss 1.0439 (0.8265) grad_norm 6.1027 (8.2461/1.7758) mem 68106MB [2022-12-20 09:27:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][550/1519] eta 0:16:14 lr 0.000009 time 0.9285 (1.0061) model_time 0.9284 (1.0043) loss 0.7732 (0.8256) grad_norm 13.9161 (8.2559/1.8096) mem 68106MB [2022-12-20 09:27:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][560/1519] eta 0:16:04 lr 0.000009 time 0.9302 (1.0060) model_time 0.9301 (1.0042) loss 0.9393 (0.8249) grad_norm 7.6226 (8.2505/1.8102) mem 68106MB [2022-12-20 09:27:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][570/1519] eta 0:15:54 lr 0.000009 time 0.9315 (1.0059) model_time 0.9314 (1.0041) loss 0.6831 (0.8239) grad_norm 8.6477 (8.2498/1.8007) mem 68106MB [2022-12-20 09:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][580/1519] eta 0:15:44 lr 0.000009 time 0.9281 (1.0058) model_time 0.9280 (1.0041) loss 0.8334 (0.8252) grad_norm 7.5780 (8.2641/1.8110) mem 68106MB [2022-12-20 09:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][590/1519] eta 0:15:34 lr 0.000009 time 0.9395 (1.0062) model_time 0.9394 (1.0045) loss 0.6788 (0.8251) grad_norm 6.7085 (8.2668/1.8091) mem 68106MB [2022-12-20 09:28:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][600/1519] eta 0:15:24 lr 0.000009 time 0.9298 (1.0062) model_time 0.9297 (1.0045) loss 0.8527 (0.8259) grad_norm 9.4911 (8.2782/1.8054) mem 68106MB [2022-12-20 09:28:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][610/1519] eta 0:15:14 lr 0.000009 time 0.9319 (1.0061) model_time 0.9318 (1.0044) loss 0.7805 (0.8258) grad_norm 8.4132 (8.2565/1.7821) mem 68106MB [2022-12-20 09:28:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][620/1519] eta 0:15:04 lr 0.000009 time 0.8858 (1.0061) model_time 0.8856 (1.0045) loss 0.7659 (0.8259) grad_norm 10.0365 (8.2572/1.7992) mem 68106MB [2022-12-20 09:28:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][630/1519] eta 0:14:54 lr 0.000009 time 0.9431 (1.0061) model_time 0.9430 (1.0045) loss 0.7008 (0.8245) grad_norm 8.1061 (8.2559/1.7817) mem 68106MB [2022-12-20 09:29:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][640/1519] eta 0:14:44 lr 0.000009 time 0.9314 (1.0060) model_time 0.9313 (1.0044) loss 0.8494 (0.8255) grad_norm 7.8374 (8.2607/1.7801) mem 68106MB [2022-12-20 09:29:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][650/1519] eta 0:14:34 lr 0.000009 time 0.9316 (1.0058) model_time 0.9314 (1.0043) loss 0.6760 (0.8250) grad_norm 8.5865 (8.2750/1.7671) mem 68106MB [2022-12-20 09:29:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][660/1519] eta 0:14:23 lr 0.000009 time 0.9322 (1.0058) model_time 0.9319 (1.0043) loss 0.7202 (0.8246) grad_norm 8.0478 (8.2808/1.7660) mem 68106MB [2022-12-20 09:29:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][670/1519] eta 0:14:14 lr 0.000009 time 0.9297 (1.0059) model_time 0.9294 (1.0044) loss 0.7914 (0.8233) grad_norm 6.6283 (8.2528/1.7656) mem 68106MB [2022-12-20 09:29:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][680/1519] eta 0:14:03 lr 0.000009 time 0.9266 (1.0059) model_time 0.9264 (1.0043) loss 0.6731 (0.8225) grad_norm 5.4464 (8.2521/1.7688) mem 68106MB [2022-12-20 09:30:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][690/1519] eta 0:13:53 lr 0.000009 time 0.9303 (1.0057) model_time 0.9301 (1.0042) loss 0.7389 (0.8228) grad_norm 10.5049 (8.2555/1.7541) mem 68106MB [2022-12-20 09:30:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][700/1519] eta 0:13:43 lr 0.000009 time 0.9609 (1.0060) model_time 0.9608 (1.0045) loss 0.7292 (0.8238) grad_norm 7.3759 (8.2911/1.7887) mem 68106MB [2022-12-20 09:30:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][710/1519] eta 0:13:33 lr 0.000009 time 0.9289 (1.0060) model_time 0.9288 (1.0045) loss 0.8318 (0.8244) grad_norm 7.8926 (8.2933/1.7979) mem 68106MB [2022-12-20 09:30:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][720/1519] eta 0:13:23 lr 0.000009 time 0.9289 (1.0059) model_time 0.9288 (1.0044) loss 1.0276 (0.8252) grad_norm 7.3491 (8.3003/1.7941) mem 68106MB [2022-12-20 09:30:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][730/1519] eta 0:13:13 lr 0.000009 time 0.9215 (1.0060) model_time 0.9213 (1.0045) loss 0.9983 (0.8256) grad_norm 8.4445 (8.2951/1.7936) mem 68106MB [2022-12-20 09:30:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][740/1519] eta 0:13:03 lr 0.000009 time 0.9345 (1.0059) model_time 0.9343 (1.0045) loss 0.8980 (0.8250) grad_norm 7.2234 (8.2928/1.7726) mem 68106MB [2022-12-20 09:31:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][750/1519] eta 0:12:53 lr 0.000009 time 0.9306 (1.0058) model_time 0.9304 (1.0044) loss 0.6864 (0.8244) grad_norm 8.0472 (8.2968/1.7718) mem 68106MB [2022-12-20 09:31:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][760/1519] eta 0:12:43 lr 0.000009 time 0.9285 (1.0057) model_time 0.9283 (1.0043) loss 0.6696 (0.8237) grad_norm 6.9133 (8.2826/1.7705) mem 68106MB [2022-12-20 09:31:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][770/1519] eta 0:12:33 lr 0.000009 time 0.9377 (1.0058) model_time 0.9374 (1.0044) loss 0.9175 (0.8235) grad_norm 7.9122 (8.2632/1.7409) mem 68106MB [2022-12-20 09:31:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][780/1519] eta 0:12:23 lr 0.000009 time 0.9310 (1.0057) model_time 0.9308 (1.0044) loss 0.6757 (0.8233) grad_norm 7.6906 (8.3293/1.7634) mem 68106MB [2022-12-20 09:31:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][790/1519] eta 0:12:13 lr 0.000009 time 0.9384 (1.0057) model_time 0.9382 (1.0043) loss 0.7078 (0.8231) grad_norm 7.7746 (8.3440/1.7685) mem 68106MB [2022-12-20 09:31:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][800/1519] eta 0:12:03 lr 0.000009 time 0.9289 (1.0056) model_time 0.9287 (1.0043) loss 1.1112 (0.8244) grad_norm 6.9617 (8.3489/1.7780) mem 68106MB [2022-12-20 09:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][810/1519] eta 0:11:52 lr 0.000009 time 0.9267 (1.0055) model_time 0.9265 (1.0042) loss 0.8362 (0.8241) grad_norm 13.0080 (8.3621/1.8024) mem 68106MB [2022-12-20 09:32:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][820/1519] eta 0:11:42 lr 0.000009 time 0.9288 (1.0054) model_time 0.9287 (1.0041) loss 0.9061 (0.8237) grad_norm 11.7527 (8.3932/1.8191) mem 68106MB [2022-12-20 09:32:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][830/1519] eta 0:11:32 lr 0.000009 time 0.9310 (1.0053) model_time 0.9308 (1.0040) loss 0.6699 (0.8231) grad_norm 8.4108 (8.3916/1.8050) mem 68106MB [2022-12-20 09:32:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][840/1519] eta 0:11:22 lr 0.000009 time 0.9397 (1.0053) model_time 0.9395 (1.0040) loss 0.8532 (0.8237) grad_norm 7.1058 (8.3932/1.8037) mem 68106MB [2022-12-20 09:32:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][850/1519] eta 0:11:12 lr 0.000009 time 0.9377 (1.0052) model_time 0.9376 (1.0039) loss 0.7108 (0.8228) grad_norm 10.8451 (8.3939/1.7892) mem 68106MB [2022-12-20 09:32:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][860/1519] eta 0:11:02 lr 0.000009 time 0.9287 (1.0051) model_time 0.9285 (1.0038) loss 0.7214 (0.8228) grad_norm 7.3604 (8.3689/1.7922) mem 68106MB [2022-12-20 09:33:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][870/1519] eta 0:10:52 lr 0.000009 time 0.9265 (1.0051) model_time 0.9263 (1.0038) loss 0.6649 (0.8227) grad_norm 9.4060 (8.3507/1.7876) mem 68106MB [2022-12-20 09:33:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][880/1519] eta 0:10:42 lr 0.000009 time 0.9299 (1.0050) model_time 0.9297 (1.0037) loss 0.7693 (0.8224) grad_norm 5.9163 (8.3609/1.8127) mem 68106MB [2022-12-20 09:33:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][890/1519] eta 0:10:32 lr 0.000009 time 0.9315 (1.0049) model_time 0.9313 (1.0036) loss 0.8449 (0.8218) grad_norm 7.4559 (8.3729/1.8204) mem 68106MB [2022-12-20 09:33:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][900/1519] eta 0:10:21 lr 0.000009 time 0.9254 (1.0048) model_time 0.9252 (1.0036) loss 0.8915 (0.8230) grad_norm 8.9390 (8.4034/1.8580) mem 68106MB [2022-12-20 09:33:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][910/1519] eta 0:10:11 lr 0.000009 time 1.0203 (1.0049) model_time 1.0202 (1.0037) loss 0.9671 (0.8233) grad_norm 16.1114 (8.4267/1.9185) mem 68106MB [2022-12-20 09:33:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][920/1519] eta 0:10:01 lr 0.000009 time 0.9350 (1.0048) model_time 0.9348 (1.0036) loss 0.7790 (0.8230) grad_norm 11.0642 (8.4632/1.9558) mem 68106MB [2022-12-20 09:34:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][930/1519] eta 0:09:51 lr 0.000009 time 0.9276 (1.0047) model_time 0.9274 (1.0035) loss 0.8564 (0.8237) grad_norm 8.7679 (8.4595/1.9619) mem 68106MB [2022-12-20 09:34:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][940/1519] eta 0:09:41 lr 0.000009 time 0.9281 (1.0047) model_time 0.9280 (1.0035) loss 0.7620 (0.8236) grad_norm 7.8384 (8.4729/1.9593) mem 68106MB [2022-12-20 09:34:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][950/1519] eta 0:09:31 lr 0.000009 time 0.9366 (1.0047) model_time 0.9364 (1.0035) loss 0.8020 (0.8227) grad_norm 9.2818 (8.4790/1.9518) mem 68106MB [2022-12-20 09:34:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][960/1519] eta 0:09:21 lr 0.000009 time 0.9242 (1.0048) model_time 0.9240 (1.0036) loss 0.9341 (0.8225) grad_norm 8.7033 (8.5001/1.9325) mem 68106MB [2022-12-20 09:34:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][970/1519] eta 0:09:11 lr 0.000009 time 0.9289 (1.0049) model_time 0.9288 (1.0037) loss 1.1665 (0.8225) grad_norm 6.0403 (8.4767/1.9193) mem 68106MB [2022-12-20 09:34:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][980/1519] eta 0:09:01 lr 0.000009 time 0.9283 (1.0050) model_time 0.9281 (1.0038) loss 0.6883 (0.8225) grad_norm 8.0418 (8.4435/1.8803) mem 68106MB [2022-12-20 09:35:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][990/1519] eta 0:08:51 lr 0.000009 time 0.9155 (1.0049) model_time 0.9154 (1.0037) loss 0.8565 (0.8231) grad_norm 7.0201 (8.4412/1.8795) mem 68106MB [2022-12-20 09:35:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1000/1519] eta 0:08:41 lr 0.000009 time 0.9284 (1.0048) model_time 0.9282 (1.0037) loss 0.9086 (0.8230) grad_norm 7.1452 (8.4291/1.8479) mem 68106MB [2022-12-20 09:35:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1010/1519] eta 0:08:31 lr 0.000009 time 0.9359 (1.0049) model_time 0.9357 (1.0037) loss 0.8437 (0.8231) grad_norm 7.5468 (8.4209/1.8279) mem 68106MB [2022-12-20 09:35:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1020/1519] eta 0:08:21 lr 0.000009 time 0.9251 (1.0050) model_time 0.9250 (1.0039) loss 0.7367 (0.8233) grad_norm 8.2092 (8.4126/1.8186) mem 68106MB [2022-12-20 09:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1030/1519] eta 0:08:11 lr 0.000009 time 0.9310 (1.0050) model_time 0.9308 (1.0039) loss 0.8416 (0.8239) grad_norm 9.9948 (8.4252/1.8167) mem 68106MB [2022-12-20 09:35:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1040/1519] eta 0:08:01 lr 0.000009 time 0.9434 (1.0051) model_time 0.9433 (1.0040) loss 0.6680 (0.8241) grad_norm 8.2805 (8.4540/1.8283) mem 68106MB [2022-12-20 09:36:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1050/1519] eta 0:07:51 lr 0.000009 time 0.9322 (1.0051) model_time 0.9321 (1.0040) loss 0.7626 (0.8239) grad_norm 7.8766 (8.4523/1.8211) mem 68106MB [2022-12-20 09:36:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1060/1519] eta 0:07:41 lr 0.000009 time 0.9321 (1.0051) model_time 0.9319 (1.0040) loss 0.7724 (0.8237) grad_norm 7.9088 (8.4615/1.8181) mem 68106MB [2022-12-20 09:36:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1070/1519] eta 0:07:31 lr 0.000009 time 0.9275 (1.0050) model_time 0.9273 (1.0039) loss 0.7100 (0.8235) grad_norm 9.5261 (8.4798/1.8151) mem 68106MB [2022-12-20 09:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1080/1519] eta 0:07:21 lr 0.000009 time 0.9348 (1.0053) model_time 0.9347 (1.0042) loss 0.7374 (0.8230) grad_norm 5.9224 (8.4845/1.8247) mem 68106MB [2022-12-20 09:36:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1090/1519] eta 0:07:11 lr 0.000009 time 0.9424 (1.0052) model_time 0.9422 (1.0041) loss 0.9451 (0.8224) grad_norm 8.4882 (8.4930/1.8312) mem 68106MB [2022-12-20 09:36:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1100/1519] eta 0:07:01 lr 0.000009 time 1.0087 (1.0052) model_time 1.0085 (1.0041) loss 0.9623 (0.8220) grad_norm 11.4565 (8.4923/1.8320) mem 68106MB [2022-12-20 09:37:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1110/1519] eta 0:06:51 lr 0.000009 time 0.9363 (1.0051) model_time 0.9362 (1.0041) loss 0.8694 (0.8225) grad_norm 10.1385 (8.4845/1.8290) mem 68106MB [2022-12-20 09:37:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1120/1519] eta 0:06:41 lr 0.000009 time 0.9346 (1.0053) model_time 0.9344 (1.0042) loss 0.9019 (0.8223) grad_norm 6.9721 (8.4785/1.8235) mem 68106MB [2022-12-20 09:37:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1130/1519] eta 0:06:31 lr 0.000009 time 0.9275 (1.0053) model_time 0.9273 (1.0042) loss 0.6889 (0.8220) grad_norm 7.9121 (8.4821/1.8187) mem 68106MB [2022-12-20 09:37:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1140/1519] eta 0:06:20 lr 0.000009 time 0.9308 (1.0052) model_time 0.9306 (1.0042) loss 0.6843 (0.8219) grad_norm 8.1363 (8.5014/1.8254) mem 68106MB [2022-12-20 09:37:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1150/1519] eta 0:06:10 lr 0.000009 time 0.9302 (1.0052) model_time 0.9300 (1.0041) loss 0.6775 (0.8221) grad_norm 9.9510 (8.5015/1.7961) mem 68106MB [2022-12-20 09:37:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1160/1519] eta 0:06:00 lr 0.000009 time 0.9312 (1.0053) model_time 0.9310 (1.0042) loss 0.8475 (0.8220) grad_norm 7.8848 (8.5128/1.7836) mem 68106MB [2022-12-20 09:38:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1170/1519] eta 0:05:50 lr 0.000009 time 0.9314 (1.0052) model_time 0.9312 (1.0042) loss 0.6854 (0.8221) grad_norm 7.7499 (8.5050/1.7854) mem 68106MB [2022-12-20 09:38:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1180/1519] eta 0:05:40 lr 0.000009 time 0.9251 (1.0052) model_time 0.9250 (1.0042) loss 1.1512 (0.8226) grad_norm 7.3261 (8.4828/1.7713) mem 68106MB [2022-12-20 09:38:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1190/1519] eta 0:05:30 lr 0.000009 time 0.9302 (1.0052) model_time 0.9300 (1.0041) loss 0.6882 (0.8221) grad_norm 7.7812 (8.4715/1.7649) mem 68106MB [2022-12-20 09:38:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1200/1519] eta 0:05:20 lr 0.000009 time 0.9229 (1.0052) model_time 0.9227 (1.0042) loss 0.7582 (0.8212) grad_norm 10.2953 (8.4822/1.7845) mem 68106MB [2022-12-20 09:38:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1210/1519] eta 0:05:10 lr 0.000009 time 0.9296 (1.0052) model_time 0.9294 (1.0042) loss 0.7538 (0.8213) grad_norm 8.5791 (8.4914/1.7806) mem 68106MB [2022-12-20 09:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1220/1519] eta 0:05:00 lr 0.000009 time 0.9265 (1.0052) model_time 0.9263 (1.0041) loss 0.7101 (0.8210) grad_norm 8.3178 (8.4908/1.7675) mem 68106MB [2022-12-20 09:39:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1230/1519] eta 0:04:50 lr 0.000009 time 0.9341 (1.0052) model_time 0.9340 (1.0042) loss 0.7080 (0.8214) grad_norm 6.7573 (8.5062/1.7802) mem 68106MB [2022-12-20 09:39:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1240/1519] eta 0:04:40 lr 0.000009 time 0.9187 (1.0052) model_time 0.9185 (1.0042) loss 0.6816 (0.8213) grad_norm 12.1269 (8.5485/1.8212) mem 68106MB [2022-12-20 09:39:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1250/1519] eta 0:04:30 lr 0.000009 time 0.9278 (1.0052) model_time 0.9275 (1.0042) loss 0.6995 (0.8206) grad_norm 7.0705 (8.5427/1.8129) mem 68106MB [2022-12-20 09:39:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1260/1519] eta 0:04:20 lr 0.000009 time 0.9318 (1.0052) model_time 0.9316 (1.0042) loss 0.8467 (0.8206) grad_norm 9.3231 (8.5317/1.8053) mem 68106MB [2022-12-20 09:39:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1270/1519] eta 0:04:10 lr 0.000009 time 0.9178 (1.0052) model_time 0.9177 (1.0042) loss 0.7199 (0.8209) grad_norm 6.9582 (8.5429/1.8056) mem 68106MB [2022-12-20 09:39:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1280/1519] eta 0:04:00 lr 0.000009 time 0.9911 (1.0052) model_time 0.9909 (1.0042) loss 0.8274 (0.8213) grad_norm 8.5080 (8.5678/1.8232) mem 68106MB [2022-12-20 09:40:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1290/1519] eta 0:03:50 lr 0.000009 time 0.9235 (1.0053) model_time 0.9232 (1.0041) loss 0.6811 (0.8211) grad_norm 8.2207 (8.5969/1.8410) mem 68106MB [2022-12-20 09:40:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1300/1519] eta 0:03:40 lr 0.000009 time 0.9344 (1.0052) model_time 0.9342 (1.0041) loss 0.6918 (0.8211) grad_norm 9.1513 (8.5623/1.8071) mem 68106MB [2022-12-20 09:40:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1310/1519] eta 0:03:30 lr 0.000009 time 0.9160 (1.0052) model_time 0.9159 (1.0040) loss 0.8432 (0.8212) grad_norm 11.0837 (8.5729/1.8017) mem 68106MB [2022-12-20 09:40:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1320/1519] eta 0:03:20 lr 0.000009 time 0.9297 (1.0052) model_time 0.9296 (1.0041) loss 0.8248 (0.8212) grad_norm 7.8711 (8.5796/1.8108) mem 68106MB [2022-12-20 09:40:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1330/1519] eta 0:03:10 lr 0.000009 time 0.9929 (1.0053) model_time 0.9927 (1.0042) loss 0.7270 (0.8210) grad_norm 8.8538 (8.5932/1.8072) mem 68106MB [2022-12-20 09:40:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1340/1519] eta 0:02:59 lr 0.000009 time 0.9252 (1.0052) model_time 0.9250 (1.0041) loss 0.8406 (0.8212) grad_norm 5.6442 (8.5828/1.8179) mem 68106MB [2022-12-20 09:41:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1350/1519] eta 0:02:49 lr 0.000009 time 0.9281 (1.0052) model_time 0.9279 (1.0041) loss 0.7214 (0.8213) grad_norm 9.3013 (8.6031/1.8203) mem 68106MB [2022-12-20 09:41:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1360/1519] eta 0:02:39 lr 0.000009 time 0.9297 (1.0052) model_time 0.9295 (1.0041) loss 0.8295 (0.8218) grad_norm 7.2712 (8.6071/1.8328) mem 68106MB [2022-12-20 09:41:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1370/1519] eta 0:02:29 lr 0.000009 time 0.9318 (1.0052) model_time 0.9316 (1.0041) loss 0.6988 (0.8217) grad_norm 13.4462 (8.6296/1.8481) mem 68106MB [2022-12-20 09:41:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1380/1519] eta 0:02:19 lr 0.000009 time 0.9329 (1.0052) model_time 0.9326 (1.0041) loss 0.6996 (0.8216) grad_norm 7.5484 (8.5971/1.8143) mem 68106MB [2022-12-20 09:41:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1390/1519] eta 0:02:09 lr 0.000009 time 0.9303 (1.0052) model_time 0.9300 (1.0041) loss 0.6972 (0.8212) grad_norm 9.3134 (8.6505/1.8946) mem 68106MB [2022-12-20 09:41:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1400/1519] eta 0:01:59 lr 0.000009 time 0.9299 (1.0051) model_time 0.9297 (1.0040) loss 0.7503 (0.8213) grad_norm 8.3584 (8.6425/1.8863) mem 68106MB [2022-12-20 09:42:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1410/1519] eta 0:01:49 lr 0.000009 time 0.9296 (1.0051) model_time 0.9294 (1.0040) loss 0.9625 (0.8212) grad_norm 7.5947 (8.6386/1.8833) mem 68106MB [2022-12-20 09:42:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1420/1519] eta 0:01:39 lr 0.000009 time 0.9312 (1.0051) model_time 0.9311 (1.0040) loss 0.7668 (0.8211) grad_norm 7.4649 (8.6129/1.8591) mem 68106MB [2022-12-20 09:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1430/1519] eta 0:01:29 lr 0.000009 time 0.9320 (1.0051) model_time 0.9319 (1.0041) loss 0.9130 (0.8217) grad_norm 10.6620 (8.5723/1.8473) mem 68106MB [2022-12-20 09:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1440/1519] eta 0:01:19 lr 0.000009 time 0.9107 (1.0051) model_time 0.9105 (1.0040) loss 1.0143 (0.8216) grad_norm 8.2348 (8.5476/1.8542) mem 68106MB [2022-12-20 09:42:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1450/1519] eta 0:01:09 lr 0.000009 time 0.9305 (1.0050) model_time 0.9303 (1.0040) loss 1.0420 (0.8218) grad_norm 7.3257 (8.5368/1.8530) mem 68106MB [2022-12-20 09:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1460/1519] eta 0:00:59 lr 0.000009 time 0.9344 (1.0050) model_time 0.9342 (1.0040) loss 0.8934 (0.8218) grad_norm 7.3036 (8.5317/1.8575) mem 68106MB [2022-12-20 09:43:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1470/1519] eta 0:00:49 lr 0.000009 time 0.9295 (1.0050) model_time 0.9293 (1.0040) loss 0.7074 (0.8217) grad_norm 13.2030 (8.5523/1.8382) mem 68106MB [2022-12-20 09:43:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1480/1519] eta 0:00:39 lr 0.000009 time 0.9281 (1.0050) model_time 0.9280 (1.0040) loss 0.6843 (0.8218) grad_norm 12.3290 (8.5316/1.8367) mem 68106MB [2022-12-20 09:43:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1490/1519] eta 0:00:29 lr 0.000009 time 0.9336 (1.0049) model_time 0.9335 (1.0039) loss 0.9326 (0.8221) grad_norm 6.6399 (8.5130/1.8298) mem 68106MB [2022-12-20 09:43:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1500/1519] eta 0:00:19 lr 0.000009 time 0.9340 (1.0050) model_time 0.9339 (1.0039) loss 0.6679 (0.8218) grad_norm 12.0389 (8.5149/1.8101) mem 68106MB [2022-12-20 09:43:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [66/100][1510/1519] eta 0:00:09 lr 0.000009 time 0.9329 (1.0050) model_time 0.9327 (1.0040) loss 0.7905 (0.8220) grad_norm 7.4182 (8.5274/1.8160) mem 68106MB [2022-12-20 09:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 66 training takes 0:25:26 [2022-12-20 09:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_66.pth saving...... [2022-12-20 09:44:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_66.pth saved !!! [2022-12-20 09:44:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.644 (0.644) Loss 0.5298 (0.5298) Acc@1 91.319 (91.319) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 09:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.331) Loss 0.5176 (0.4948) Acc@1 91.667 (92.708) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 09:44:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.314) Loss 0.4778 (0.4925) Acc@1 91.319 (92.675) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 09:44:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.310) Loss 0.6220 (0.5000) Acc@1 90.972 (92.361) Acc@5 98.264 (98.410) Mem 68106MB [2022-12-20 09:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.310 (0.312) Loss 0.4573 (0.4913) Acc@1 93.750 (92.412) Acc@5 99.306 (98.484) Mem 68106MB [2022-12-20 09:44:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.310) Loss 0.4684 (0.4890) Acc@1 91.667 (92.456) Acc@5 99.653 (98.509) Mem 68106MB [2022-12-20 09:44:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.301 (0.308) Loss 0.5047 (0.4885) Acc@1 91.319 (92.441) Acc@5 97.917 (98.486) Mem 68106MB [2022-12-20 09:44:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.307) Loss 0.5369 (0.4897) Acc@1 93.403 (92.400) Acc@5 97.917 (98.474) Mem 68106MB [2022-12-20 09:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.306) Loss 0.4244 (0.4882) Acc@1 93.750 (92.434) Acc@5 98.264 (98.504) Mem 68106MB [2022-12-20 09:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:66] * Acc@1 92.395 Acc@5 98.510 [2022-12-20 09:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 09:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 09:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 09:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.40% [2022-12-20 09:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][0/1519] eta 0:33:12 lr 0.000009 time 1.3119 (1.3119) model_time 0.9048 (0.9048) loss 0.6656 (0.6656) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 09:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][10/1519] eta 0:26:01 lr 0.000009 time 0.9646 (1.0348) model_time 0.9644 (0.9974) loss 0.6740 (0.7903) grad_norm 8.3019 (10.8885/2.4387) mem 68106MB [2022-12-20 09:45:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][20/1519] eta 0:25:34 lr 0.000009 time 0.9327 (1.0240) model_time 0.9326 (1.0042) loss 0.6745 (0.8259) grad_norm 6.1336 (9.3778/2.4758) mem 68106MB [2022-12-20 09:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][30/1519] eta 0:25:12 lr 0.000009 time 0.9211 (1.0155) model_time 0.9209 (1.0015) loss 1.1413 (0.8212) grad_norm 6.8314 (9.1222/2.1997) mem 68106MB [2022-12-20 09:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][40/1519] eta 0:24:58 lr 0.000009 time 0.9311 (1.0134) model_time 0.9309 (1.0027) loss 0.6890 (0.8081) grad_norm 8.0047 (8.8936/2.1477) mem 68106MB [2022-12-20 09:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][50/1519] eta 0:24:49 lr 0.000009 time 0.9347 (1.0143) model_time 0.9345 (1.0056) loss 0.7607 (0.8112) grad_norm 6.3769 (8.5883/2.1496) mem 68106MB [2022-12-20 09:46:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][60/1519] eta 0:24:37 lr 0.000009 time 0.9290 (1.0125) model_time 0.9289 (1.0052) loss 0.8287 (0.8022) grad_norm 10.1816 (8.5310/2.0363) mem 68106MB [2022-12-20 09:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][70/1519] eta 0:24:24 lr 0.000009 time 0.9290 (1.0107) model_time 0.9288 (1.0044) loss 0.6837 (0.8054) grad_norm 13.9699 (8.8054/2.1271) mem 68106MB [2022-12-20 09:46:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][80/1519] eta 0:24:13 lr 0.000009 time 0.9426 (1.0100) model_time 0.9425 (1.0044) loss 0.6775 (0.7985) grad_norm 9.1387 (8.6908/2.0797) mem 68106MB [2022-12-20 09:46:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][90/1519] eta 0:24:03 lr 0.000009 time 0.9704 (1.0098) model_time 0.9701 (1.0047) loss 0.7213 (0.7927) grad_norm 7.6262 (8.6974/2.0142) mem 68106MB [2022-12-20 09:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][100/1519] eta 0:23:52 lr 0.000009 time 0.9163 (1.0095) model_time 0.9161 (1.0048) loss 0.6750 (0.7911) grad_norm 10.7127 (8.9034/2.0241) mem 68106MB [2022-12-20 09:46:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][110/1519] eta 0:23:41 lr 0.000009 time 0.9257 (1.0087) model_time 0.9256 (1.0044) loss 0.7455 (0.7921) grad_norm 8.4045 (8.8898/1.9446) mem 68106MB [2022-12-20 09:47:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][120/1519] eta 0:23:31 lr 0.000009 time 0.9435 (1.0091) model_time 0.9434 (1.0051) loss 0.7261 (0.7922) grad_norm 10.4662 (8.8964/1.9004) mem 68106MB [2022-12-20 09:47:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][130/1519] eta 0:23:23 lr 0.000009 time 0.9835 (1.0108) model_time 0.9833 (1.0071) loss 0.7329 (0.7938) grad_norm 10.1505 (8.9038/1.8515) mem 68106MB [2022-12-20 09:47:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][140/1519] eta 0:23:12 lr 0.000008 time 0.9328 (1.0100) model_time 0.9325 (1.0066) loss 0.7622 (0.7954) grad_norm 10.1779 (8.8537/1.8199) mem 68106MB [2022-12-20 09:47:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][150/1519] eta 0:23:03 lr 0.000008 time 0.9220 (1.0107) model_time 0.9219 (1.0074) loss 0.7943 (0.7927) grad_norm 6.6819 (8.8075/1.8006) mem 68106MB [2022-12-20 09:47:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][160/1519] eta 0:22:52 lr 0.000008 time 0.9232 (1.0097) model_time 0.9231 (1.0066) loss 0.7634 (0.7935) grad_norm 5.8661 (8.7535/1.8212) mem 68106MB [2022-12-20 09:47:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][170/1519] eta 0:22:41 lr 0.000008 time 0.9210 (1.0092) model_time 0.9208 (1.0063) loss 0.6643 (0.7932) grad_norm 7.4736 (8.7232/1.7851) mem 68106MB [2022-12-20 09:48:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][180/1519] eta 0:22:32 lr 0.000008 time 0.9189 (1.0098) model_time 0.9188 (1.0070) loss 0.6637 (0.7928) grad_norm 7.5718 (8.7083/1.7506) mem 68106MB [2022-12-20 09:48:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][190/1519] eta 0:22:22 lr 0.000008 time 0.9985 (1.0104) model_time 0.9984 (1.0078) loss 0.7146 (0.7944) grad_norm 7.0306 (8.6541/1.7280) mem 68106MB [2022-12-20 09:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][200/1519] eta 0:22:12 lr 0.000008 time 0.9213 (1.0104) model_time 0.9212 (1.0078) loss 1.2404 (0.7998) grad_norm 8.9758 (8.6213/1.7414) mem 68106MB [2022-12-20 09:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][210/1519] eta 0:22:01 lr 0.000008 time 0.9262 (1.0098) model_time 0.9261 (1.0073) loss 0.8827 (0.7992) grad_norm 8.7711 (8.5569/1.7407) mem 68106MB [2022-12-20 09:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][220/1519] eta 0:21:51 lr 0.000008 time 0.9345 (1.0098) model_time 0.9343 (1.0075) loss 0.6831 (0.8042) grad_norm 9.2187 (8.5359/1.7519) mem 68106MB [2022-12-20 09:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][230/1519] eta 0:21:41 lr 0.000008 time 0.9260 (1.0100) model_time 0.9258 (1.0078) loss 0.8746 (0.8031) grad_norm 8.3614 (8.5286/1.7285) mem 68106MB [2022-12-20 09:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][240/1519] eta 0:21:31 lr 0.000008 time 0.9270 (1.0096) model_time 0.9268 (1.0074) loss 0.9525 (0.8010) grad_norm 8.1202 (8.5156/1.7125) mem 68106MB [2022-12-20 09:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][250/1519] eta 0:21:20 lr 0.000008 time 1.0117 (1.0094) model_time 1.0116 (1.0073) loss 1.2613 (0.8034) grad_norm 11.0199 (8.5071/1.7211) mem 68106MB [2022-12-20 09:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][260/1519] eta 0:21:11 lr 0.000008 time 0.9217 (1.0098) model_time 0.9216 (1.0077) loss 0.7823 (0.8036) grad_norm 8.5857 (8.5089/1.7317) mem 68106MB [2022-12-20 09:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][270/1519] eta 0:21:00 lr 0.000008 time 0.9264 (1.0093) model_time 0.9263 (1.0073) loss 0.9253 (0.8020) grad_norm 10.0693 (8.5450/1.8276) mem 68106MB [2022-12-20 09:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][280/1519] eta 0:20:50 lr 0.000008 time 0.9195 (1.0089) model_time 0.9194 (1.0070) loss 0.6504 (0.8006) grad_norm 5.7958 (8.5219/1.8140) mem 68106MB [2022-12-20 09:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][290/1519] eta 0:20:39 lr 0.000008 time 0.9384 (1.0086) model_time 0.9383 (1.0067) loss 0.6881 (0.8013) grad_norm 9.8128 (8.5424/1.7946) mem 68106MB [2022-12-20 09:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][300/1519] eta 0:20:29 lr 0.000008 time 0.9320 (1.0088) model_time 0.9319 (1.0070) loss 0.7622 (0.8027) grad_norm 11.9568 (8.5645/1.7980) mem 68106MB [2022-12-20 09:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][310/1519] eta 0:20:19 lr 0.000008 time 0.9340 (1.0084) model_time 0.9338 (1.0067) loss 0.8454 (0.8021) grad_norm 9.6209 (8.5197/1.8004) mem 68106MB [2022-12-20 09:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][320/1519] eta 0:20:08 lr 0.000008 time 0.9221 (1.0081) model_time 0.9220 (1.0064) loss 1.0474 (0.8062) grad_norm 5.6685 (8.5059/1.8008) mem 68106MB [2022-12-20 09:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][330/1519] eta 0:19:58 lr 0.000008 time 0.9285 (1.0081) model_time 0.9283 (1.0064) loss 0.9448 (0.8092) grad_norm 7.3851 (8.4871/1.7863) mem 68106MB [2022-12-20 09:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][340/1519] eta 0:19:48 lr 0.000008 time 0.9397 (1.0078) model_time 0.9391 (1.0062) loss 0.7360 (0.8071) grad_norm 7.7973 (8.4646/1.7693) mem 68106MB [2022-12-20 09:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][350/1519] eta 0:19:38 lr 0.000008 time 0.9254 (1.0078) model_time 0.9253 (1.0061) loss 0.8313 (0.8059) grad_norm 6.8600 (8.4359/1.7588) mem 68106MB [2022-12-20 09:51:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][360/1519] eta 0:19:28 lr 0.000008 time 0.9760 (1.0078) model_time 0.9759 (1.0062) loss 0.6772 (0.8085) grad_norm 6.1681 (8.4579/1.7977) mem 68106MB [2022-12-20 09:51:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][370/1519] eta 0:19:17 lr 0.000008 time 0.9019 (1.0075) model_time 0.9017 (1.0060) loss 0.9334 (0.8084) grad_norm 8.4820 (8.4545/1.7768) mem 68106MB [2022-12-20 09:51:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][380/1519] eta 0:19:07 lr 0.000008 time 0.9357 (1.0075) model_time 0.9355 (1.0060) loss 0.9079 (0.8075) grad_norm 7.2744 (8.4529/1.7625) mem 68106MB [2022-12-20 09:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][390/1519] eta 0:18:57 lr 0.000008 time 0.9352 (1.0073) model_time 0.9351 (1.0058) loss 0.6925 (0.8086) grad_norm 9.8966 (8.4673/1.7606) mem 68106MB [2022-12-20 09:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][400/1519] eta 0:18:47 lr 0.000008 time 0.9253 (1.0074) model_time 0.9249 (1.0059) loss 0.7173 (0.8094) grad_norm 12.3809 (8.4659/1.7735) mem 68106MB [2022-12-20 09:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][410/1519] eta 0:18:37 lr 0.000008 time 0.9292 (1.0075) model_time 0.9291 (1.0060) loss 1.0319 (0.8093) grad_norm 5.7358 (8.4607/1.7664) mem 68106MB [2022-12-20 09:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][420/1519] eta 0:18:26 lr 0.000008 time 0.9330 (1.0073) model_time 0.9328 (1.0058) loss 0.8493 (0.8084) grad_norm 6.9091 (8.4469/1.7563) mem 68106MB [2022-12-20 09:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][430/1519] eta 0:18:17 lr 0.000008 time 1.2252 (1.0077) model_time 1.2251 (1.0063) loss 0.8199 (0.8071) grad_norm 10.9305 (8.4805/1.7638) mem 68106MB [2022-12-20 09:52:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][440/1519] eta 0:18:07 lr 0.000008 time 0.9242 (1.0078) model_time 0.9240 (1.0065) loss 0.6885 (0.8063) grad_norm 9.3909 (8.5125/1.7728) mem 68106MB [2022-12-20 09:52:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][450/1519] eta 0:17:57 lr 0.000008 time 0.9185 (1.0077) model_time 0.9183 (1.0063) loss 0.7431 (0.8055) grad_norm 10.1936 (8.5147/1.7685) mem 68106MB [2022-12-20 09:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][460/1519] eta 0:17:47 lr 0.000008 time 0.9257 (1.0081) model_time 0.9256 (1.0068) loss 0.7159 (0.8066) grad_norm 6.9790 (8.5100/1.7679) mem 68106MB [2022-12-20 09:53:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][470/1519] eta 0:17:37 lr 0.000008 time 0.9316 (1.0079) model_time 0.9314 (1.0066) loss 0.8118 (0.8069) grad_norm 8.3768 (8.5268/1.7777) mem 68106MB [2022-12-20 09:53:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][480/1519] eta 0:17:27 lr 0.000008 time 0.9236 (1.0081) model_time 0.9235 (1.0068) loss 0.8826 (0.8068) grad_norm 9.4405 (8.5571/1.7985) mem 68106MB [2022-12-20 09:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][490/1519] eta 0:17:17 lr 0.000008 time 0.9194 (1.0080) model_time 0.9192 (1.0068) loss 0.6872 (0.8068) grad_norm 7.9564 (8.5749/1.7896) mem 68106MB [2022-12-20 09:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][500/1519] eta 0:17:07 lr 0.000008 time 0.9296 (1.0080) model_time 0.9294 (1.0068) loss 0.8931 (0.8074) grad_norm 6.0214 (8.5634/1.7807) mem 68106MB [2022-12-20 09:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][510/1519] eta 0:16:57 lr 0.000008 time 0.9229 (1.0082) model_time 0.9227 (1.0070) loss 0.8296 (0.8086) grad_norm 8.6628 (8.5734/1.8048) mem 68106MB [2022-12-20 09:53:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][520/1519] eta 0:16:47 lr 0.000008 time 0.9292 (1.0080) model_time 0.9290 (1.0068) loss 0.8532 (0.8089) grad_norm 6.7183 (8.5689/1.8117) mem 68106MB [2022-12-20 09:54:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][530/1519] eta 0:16:36 lr 0.000008 time 0.9242 (1.0080) model_time 0.9241 (1.0068) loss 1.2425 (0.8097) grad_norm 8.4316 (8.5714/1.8091) mem 68106MB [2022-12-20 09:54:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][540/1519] eta 0:16:26 lr 0.000008 time 0.9229 (1.0078) model_time 0.9228 (1.0067) loss 0.6784 (0.8108) grad_norm 7.4635 (8.5453/1.8031) mem 68106MB [2022-12-20 09:54:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][550/1519] eta 0:16:16 lr 0.000008 time 0.9339 (1.0079) model_time 0.9337 (1.0067) loss 0.7093 (0.8103) grad_norm 7.5933 (8.5713/1.8270) mem 68106MB [2022-12-20 09:54:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][560/1519] eta 0:16:06 lr 0.000008 time 0.9256 (1.0077) model_time 0.9255 (1.0065) loss 0.8320 (0.8105) grad_norm 9.0547 (8.5956/1.8383) mem 68106MB [2022-12-20 09:54:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][570/1519] eta 0:15:56 lr 0.000008 time 0.9311 (1.0080) model_time 0.9309 (1.0068) loss 0.6621 (0.8112) grad_norm 6.0063 (8.5713/1.8341) mem 68106MB [2022-12-20 09:54:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][580/1519] eta 0:15:46 lr 0.000008 time 0.9181 (1.0079) model_time 0.9179 (1.0067) loss 0.8422 (0.8121) grad_norm 9.3120 (8.5742/1.8265) mem 68106MB [2022-12-20 09:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][590/1519] eta 0:15:36 lr 0.000008 time 0.9213 (1.0077) model_time 0.9212 (1.0066) loss 0.8444 (0.8116) grad_norm 6.0147 (8.5625/1.8214) mem 68106MB [2022-12-20 09:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][600/1519] eta 0:15:25 lr 0.000008 time 0.9268 (1.0075) model_time 0.9267 (1.0064) loss 0.8886 (0.8126) grad_norm 7.1451 (8.5434/1.8201) mem 68106MB [2022-12-20 09:55:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][610/1519] eta 0:15:15 lr 0.000008 time 0.9317 (1.0075) model_time 0.9315 (1.0064) loss 0.8042 (0.8128) grad_norm 9.7285 (8.5054/1.7691) mem 68106MB [2022-12-20 09:55:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][620/1519] eta 0:15:05 lr 0.000008 time 0.9300 (1.0074) model_time 0.9298 (1.0063) loss 0.6610 (0.8123) grad_norm 7.1398 (8.4949/1.7690) mem 68106MB [2022-12-20 09:55:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][630/1519] eta 0:14:55 lr 0.000008 time 0.9252 (1.0072) model_time 0.9251 (1.0062) loss 1.0789 (0.8121) grad_norm 5.8551 (8.5058/1.7945) mem 68106MB [2022-12-20 09:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][640/1519] eta 0:14:45 lr 0.000008 time 0.9388 (1.0071) model_time 0.9387 (1.0061) loss 0.8283 (0.8124) grad_norm 8.3677 (8.5031/1.7846) mem 68106MB [2022-12-20 09:56:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][650/1519] eta 0:14:35 lr 0.000008 time 0.9214 (1.0070) model_time 0.9212 (1.0059) loss 0.8007 (0.8121) grad_norm 7.0799 (8.5061/1.7770) mem 68106MB [2022-12-20 09:56:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][660/1519] eta 0:14:24 lr 0.000008 time 0.9270 (1.0070) model_time 0.9268 (1.0059) loss 0.7087 (0.8110) grad_norm 9.4571 (8.4972/1.7759) mem 68106MB [2022-12-20 09:56:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][670/1519] eta 0:14:14 lr 0.000008 time 0.9243 (1.0068) model_time 0.9242 (1.0058) loss 0.7956 (0.8116) grad_norm 10.6832 (8.4478/1.7596) mem 68106MB [2022-12-20 09:56:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][680/1519] eta 0:14:04 lr 0.000008 time 0.9217 (1.0068) model_time 0.9216 (1.0058) loss 0.6734 (0.8115) grad_norm 8.2603 (8.4625/1.7497) mem 68106MB [2022-12-20 09:56:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][690/1519] eta 0:13:54 lr 0.000008 time 0.9073 (1.0071) model_time 0.9071 (1.0061) loss 0.7414 (0.8116) grad_norm 7.9512 (8.4623/1.7429) mem 68106MB [2022-12-20 09:56:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][700/1519] eta 0:13:44 lr 0.000008 time 0.9298 (1.0070) model_time 0.9297 (1.0060) loss 0.6662 (0.8105) grad_norm 8.1881 (8.4251/1.7162) mem 68106MB [2022-12-20 09:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][710/1519] eta 0:13:34 lr 0.000008 time 0.9178 (1.0069) model_time 0.9176 (1.0059) loss 0.9675 (0.8108) grad_norm 7.7087 (8.4243/1.7184) mem 68106MB [2022-12-20 09:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][720/1519] eta 0:13:24 lr 0.000008 time 0.9875 (1.0070) model_time 0.9873 (1.0061) loss 0.7298 (0.8113) grad_norm 7.5536 (8.4170/1.7261) mem 68106MB [2022-12-20 09:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][730/1519] eta 0:13:14 lr 0.000008 time 0.9269 (1.0069) model_time 0.9267 (1.0059) loss 0.8541 (0.8109) grad_norm 8.0083 (8.4249/1.7753) mem 68106MB [2022-12-20 09:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][740/1519] eta 0:13:04 lr 0.000008 time 0.9187 (1.0068) model_time 0.9186 (1.0058) loss 0.8359 (0.8108) grad_norm 9.4307 (8.4331/1.7759) mem 68106MB [2022-12-20 09:57:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][750/1519] eta 0:12:54 lr 0.000008 time 0.9328 (1.0070) model_time 0.9326 (1.0061) loss 0.7814 (0.8111) grad_norm 6.8040 (8.4645/1.8133) mem 68106MB [2022-12-20 09:57:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][760/1519] eta 0:12:44 lr 0.000008 time 0.9155 (1.0069) model_time 0.9153 (1.0060) loss 0.8610 (0.8103) grad_norm 9.1157 (8.4785/1.7962) mem 68106MB [2022-12-20 09:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][770/1519] eta 0:12:34 lr 0.000008 time 0.9244 (1.0068) model_time 0.9242 (1.0058) loss 0.8998 (0.8103) grad_norm 7.9773 (8.4635/1.8002) mem 68106MB [2022-12-20 09:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][780/1519] eta 0:12:24 lr 0.000008 time 0.9213 (1.0068) model_time 0.9211 (1.0059) loss 0.8901 (0.8099) grad_norm 8.3895 (8.5031/1.8993) mem 68106MB [2022-12-20 09:58:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][790/1519] eta 0:12:13 lr 0.000008 time 0.9214 (1.0068) model_time 0.9212 (1.0059) loss 0.7067 (0.8090) grad_norm 7.7026 (8.5015/1.8993) mem 68106MB [2022-12-20 09:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][800/1519] eta 0:12:04 lr 0.000008 time 0.9482 (1.0072) model_time 0.9480 (1.0063) loss 0.7184 (0.8084) grad_norm 11.8240 (8.5228/1.9047) mem 68106MB [2022-12-20 09:58:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][810/1519] eta 0:11:54 lr 0.000008 time 0.9241 (1.0072) model_time 0.9239 (1.0063) loss 0.7854 (0.8082) grad_norm 6.6292 (8.5504/1.9124) mem 68106MB [2022-12-20 09:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][820/1519] eta 0:11:44 lr 0.000008 time 0.9165 (1.0073) model_time 0.9163 (1.0064) loss 0.6740 (0.8082) grad_norm 7.4679 (8.5557/1.9003) mem 68106MB [2022-12-20 09:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][830/1519] eta 0:11:33 lr 0.000008 time 0.9215 (1.0072) model_time 0.9214 (1.0063) loss 0.6954 (0.8080) grad_norm 11.8211 (8.5647/1.9131) mem 68106MB [2022-12-20 09:59:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][840/1519] eta 0:11:23 lr 0.000008 time 0.9255 (1.0072) model_time 0.9254 (1.0063) loss 0.8820 (0.8077) grad_norm 5.6784 (8.5665/1.9166) mem 68106MB [2022-12-20 09:59:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][850/1519] eta 0:11:13 lr 0.000008 time 0.9205 (1.0071) model_time 0.9204 (1.0062) loss 0.7624 (0.8078) grad_norm 8.7951 (8.5802/1.9133) mem 68106MB [2022-12-20 09:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][860/1519] eta 0:11:03 lr 0.000008 time 0.9356 (1.0071) model_time 0.9355 (1.0063) loss 0.7025 (0.8081) grad_norm 7.4214 (8.5712/1.9001) mem 68106MB [2022-12-20 09:59:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][870/1519] eta 0:10:53 lr 0.000008 time 0.9238 (1.0070) model_time 0.9236 (1.0062) loss 0.8637 (0.8092) grad_norm 6.7641 (8.5339/1.8555) mem 68106MB [2022-12-20 09:59:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][880/1519] eta 0:10:43 lr 0.000008 time 0.9739 (1.0070) model_time 0.9738 (1.0061) loss 0.7250 (0.8096) grad_norm 7.9719 (8.5490/1.8533) mem 68106MB [2022-12-20 10:00:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][890/1519] eta 0:10:33 lr 0.000008 time 0.9216 (1.0069) model_time 0.9215 (1.0060) loss 0.7490 (0.8097) grad_norm 7.6245 (8.5179/1.8566) mem 68106MB [2022-12-20 10:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][900/1519] eta 0:10:23 lr 0.000008 time 0.9824 (1.0069) model_time 0.9823 (1.0061) loss 0.6878 (0.8095) grad_norm 10.0104 (8.5021/1.8546) mem 68106MB [2022-12-20 10:00:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][910/1519] eta 0:10:13 lr 0.000008 time 0.9372 (1.0069) model_time 0.9371 (1.0060) loss 0.7600 (0.8090) grad_norm 8.8831 (8.5296/1.8438) mem 68106MB [2022-12-20 10:00:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][920/1519] eta 0:10:03 lr 0.000008 time 0.9210 (1.0068) model_time 0.9208 (1.0060) loss 0.7580 (0.8093) grad_norm 7.9581 (8.5321/1.8389) mem 68106MB [2022-12-20 10:00:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][930/1519] eta 0:09:53 lr 0.000008 time 0.9211 (1.0071) model_time 0.9210 (1.0063) loss 0.7639 (0.8093) grad_norm 7.1872 (8.5413/1.8355) mem 68106MB [2022-12-20 10:00:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][940/1519] eta 0:09:43 lr 0.000008 time 0.9223 (1.0070) model_time 0.9221 (1.0062) loss 0.8006 (0.8090) grad_norm 7.9755 (8.5306/1.8406) mem 68106MB [2022-12-20 10:01:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][950/1519] eta 0:09:32 lr 0.000008 time 0.9271 (1.0069) model_time 0.9270 (1.0061) loss 1.0717 (0.8093) grad_norm 8.7215 (8.5475/1.8355) mem 68106MB [2022-12-20 10:01:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][960/1519] eta 0:09:22 lr 0.000008 time 0.9219 (1.0068) model_time 0.9218 (1.0060) loss 0.9938 (0.8101) grad_norm 11.1165 (8.5441/1.8140) mem 68106MB [2022-12-20 10:01:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][970/1519] eta 0:09:12 lr 0.000008 time 1.0015 (1.0068) model_time 1.0014 (1.0060) loss 0.8809 (0.8095) grad_norm 7.7854 (8.5365/1.8146) mem 68106MB [2022-12-20 10:01:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][980/1519] eta 0:09:02 lr 0.000008 time 0.9205 (1.0068) model_time 0.9204 (1.0060) loss 0.7179 (0.8095) grad_norm 7.4541 (8.5266/1.8212) mem 68106MB [2022-12-20 10:01:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][990/1519] eta 0:08:52 lr 0.000008 time 0.9191 (1.0067) model_time 0.9190 (1.0059) loss 1.0276 (0.8100) grad_norm 8.1200 (8.5128/1.8201) mem 68106MB [2022-12-20 10:01:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1000/1519] eta 0:08:42 lr 0.000008 time 0.9232 (1.0066) model_time 0.9231 (1.0058) loss 0.9075 (0.8098) grad_norm 9.5984 (8.4997/1.8074) mem 68106MB [2022-12-20 10:02:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1010/1519] eta 0:08:32 lr 0.000008 time 0.9275 (1.0066) model_time 0.9274 (1.0058) loss 0.8612 (0.8097) grad_norm 6.8462 (8.5034/1.8015) mem 68106MB [2022-12-20 10:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1020/1519] eta 0:08:22 lr 0.000008 time 0.9225 (1.0065) model_time 0.9223 (1.0057) loss 1.0829 (0.8095) grad_norm 9.2477 (8.5210/1.8150) mem 68106MB [2022-12-20 10:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1030/1519] eta 0:08:12 lr 0.000008 time 0.9230 (1.0065) model_time 0.9229 (1.0057) loss 0.6800 (0.8096) grad_norm 7.9781 (8.4972/1.8146) mem 68106MB [2022-12-20 10:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1040/1519] eta 0:08:02 lr 0.000008 time 0.9142 (1.0064) model_time 0.9140 (1.0057) loss 0.8566 (0.8097) grad_norm 11.0405 (8.4592/1.8157) mem 68106MB [2022-12-20 10:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1050/1519] eta 0:07:51 lr 0.000008 time 0.9624 (1.0064) model_time 0.9622 (1.0056) loss 0.7137 (0.8091) grad_norm 6.1308 (8.4580/1.8152) mem 68106MB [2022-12-20 10:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1060/1519] eta 0:07:41 lr 0.000008 time 0.9293 (1.0063) model_time 0.9291 (1.0056) loss 0.9342 (0.8102) grad_norm 8.9096 (8.4640/1.8125) mem 68106MB [2022-12-20 10:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1070/1519] eta 0:07:31 lr 0.000008 time 0.9238 (1.0064) model_time 0.9236 (1.0056) loss 1.2439 (0.8107) grad_norm 12.7810 (8.4971/1.8468) mem 68106MB [2022-12-20 10:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1080/1519] eta 0:07:21 lr 0.000008 time 0.9259 (1.0063) model_time 0.9258 (1.0055) loss 0.7999 (0.8110) grad_norm 6.7623 (8.4859/1.8472) mem 68106MB [2022-12-20 10:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1090/1519] eta 0:07:11 lr 0.000008 time 0.9236 (1.0063) model_time 0.9235 (1.0055) loss 0.6921 (0.8104) grad_norm 8.9835 (8.4641/1.8437) mem 68106MB [2022-12-20 10:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1100/1519] eta 0:07:01 lr 0.000008 time 0.9190 (1.0063) model_time 0.9189 (1.0055) loss 0.9546 (0.8104) grad_norm 7.0217 (8.4772/1.8504) mem 68106MB [2022-12-20 10:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1110/1519] eta 0:06:51 lr 0.000008 time 0.9732 (1.0063) model_time 0.9730 (1.0056) loss 0.7834 (0.8109) grad_norm 8.5379 (8.4612/1.8175) mem 68106MB [2022-12-20 10:03:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1120/1519] eta 0:06:41 lr 0.000008 time 1.0016 (1.0063) model_time 1.0015 (1.0056) loss 0.8617 (0.8111) grad_norm 11.6899 (8.4804/1.8243) mem 68106MB [2022-12-20 10:04:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1130/1519] eta 0:06:31 lr 0.000008 time 0.9989 (1.0064) model_time 0.9988 (1.0056) loss 0.7291 (0.8116) grad_norm 8.5729 (8.4732/1.8187) mem 68106MB [2022-12-20 10:04:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1140/1519] eta 0:06:21 lr 0.000008 time 0.9231 (1.0064) model_time 0.9229 (1.0056) loss 0.7441 (0.8121) grad_norm 7.1249 (8.5032/1.8206) mem 68106MB [2022-12-20 10:04:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1150/1519] eta 0:06:11 lr 0.000008 time 1.0087 (1.0064) model_time 1.0084 (1.0057) loss 0.6563 (0.8125) grad_norm 6.9034 (8.4741/1.8052) mem 68106MB [2022-12-20 10:04:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1160/1519] eta 0:06:01 lr 0.000008 time 0.9287 (1.0064) model_time 0.9286 (1.0056) loss 0.6850 (0.8126) grad_norm 7.8593 (8.4582/1.7818) mem 68106MB [2022-12-20 10:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1170/1519] eta 0:05:51 lr 0.000008 time 0.9181 (1.0064) model_time 0.9179 (1.0057) loss 0.7816 (0.8124) grad_norm 6.4334 (8.4701/1.7814) mem 68106MB [2022-12-20 10:04:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1180/1519] eta 0:05:41 lr 0.000008 time 0.9237 (1.0063) model_time 0.9236 (1.0056) loss 0.7196 (0.8125) grad_norm 11.0178 (8.4728/1.7805) mem 68106MB [2022-12-20 10:05:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1190/1519] eta 0:05:31 lr 0.000008 time 0.9242 (1.0063) model_time 0.9240 (1.0056) loss 0.9639 (0.8124) grad_norm 8.3592 (8.4789/1.7844) mem 68106MB [2022-12-20 10:05:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1200/1519] eta 0:05:21 lr 0.000008 time 0.9221 (1.0063) model_time 0.9220 (1.0056) loss 0.8263 (0.8125) grad_norm 6.9840 (8.4844/1.7884) mem 68106MB [2022-12-20 10:05:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1210/1519] eta 0:05:10 lr 0.000008 time 0.9230 (1.0063) model_time 0.9229 (1.0056) loss 0.6633 (0.8121) grad_norm 7.6052 (8.4789/1.7874) mem 68106MB [2022-12-20 10:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1220/1519] eta 0:05:00 lr 0.000008 time 0.9345 (1.0063) model_time 0.9344 (1.0056) loss 0.7504 (0.8120) grad_norm 9.6489 (8.5112/1.7839) mem 68106MB [2022-12-20 10:05:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1230/1519] eta 0:04:50 lr 0.000008 time 0.9732 (1.0063) model_time 0.9730 (1.0055) loss 0.8199 (0.8121) grad_norm 9.4438 (8.4995/1.7586) mem 68106MB [2022-12-20 10:05:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1240/1519] eta 0:04:40 lr 0.000008 time 0.9195 (1.0062) model_time 0.9193 (1.0055) loss 0.6551 (0.8120) grad_norm 6.3560 (8.5092/1.7641) mem 68106MB [2022-12-20 10:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1250/1519] eta 0:04:30 lr 0.000008 time 0.9435 (1.0064) model_time 0.9433 (1.0056) loss 0.6662 (0.8123) grad_norm 7.9700 (8.5156/1.7547) mem 68106MB [2022-12-20 10:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1260/1519] eta 0:04:20 lr 0.000008 time 0.9258 (1.0063) model_time 0.9256 (1.0056) loss 0.9744 (0.8120) grad_norm 8.6895 (8.5258/1.7652) mem 68106MB [2022-12-20 10:06:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1270/1519] eta 0:04:10 lr 0.000008 time 0.9291 (1.0063) model_time 0.9289 (1.0056) loss 0.9128 (0.8118) grad_norm 7.3334 (8.5541/1.7519) mem 68106MB [2022-12-20 10:06:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1280/1519] eta 0:04:00 lr 0.000008 time 0.9353 (1.0063) model_time 0.9352 (1.0056) loss 0.9166 (0.8122) grad_norm 10.1983 (8.5522/1.7665) mem 68106MB [2022-12-20 10:06:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1290/1519] eta 0:03:50 lr 0.000008 time 0.9735 (1.0063) model_time 0.9733 (1.0056) loss 0.8198 (0.8120) grad_norm 10.3089 (8.5421/1.7810) mem 68106MB [2022-12-20 10:06:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1300/1519] eta 0:03:40 lr 0.000008 time 0.9994 (1.0063) model_time 0.9993 (1.0056) loss 1.0229 (0.8120) grad_norm 9.3934 (8.5283/1.7895) mem 68106MB [2022-12-20 10:07:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1310/1519] eta 0:03:30 lr 0.000008 time 1.0030 (1.0064) model_time 1.0029 (1.0057) loss 0.9273 (0.8120) grad_norm 10.8018 (8.5345/1.7917) mem 68106MB [2022-12-20 10:07:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1320/1519] eta 0:03:20 lr 0.000008 time 0.9237 (1.0064) model_time 0.9236 (1.0057) loss 0.8520 (0.8117) grad_norm 7.9484 (8.5501/1.7889) mem 68106MB [2022-12-20 10:07:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1330/1519] eta 0:03:10 lr 0.000008 time 1.2138 (1.0066) model_time 1.2136 (1.0059) loss 0.9375 (0.8118) grad_norm 8.7036 (8.5351/1.7427) mem 68106MB [2022-12-20 10:07:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1340/1519] eta 0:03:00 lr 0.000008 time 0.9252 (1.0066) model_time 0.9251 (1.0059) loss 0.8639 (0.8116) grad_norm 10.6266 (8.5668/1.8312) mem 68106MB [2022-12-20 10:07:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1350/1519] eta 0:02:50 lr 0.000008 time 0.9180 (1.0066) model_time 0.9178 (1.0059) loss 0.6870 (0.8115) grad_norm 8.9985 (8.5334/1.7958) mem 68106MB [2022-12-20 10:07:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1360/1519] eta 0:02:40 lr 0.000008 time 0.9391 (1.0065) model_time 0.9389 (1.0058) loss 0.6856 (0.8113) grad_norm 8.0403 (8.5219/1.8019) mem 68106MB [2022-12-20 10:08:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1370/1519] eta 0:02:29 lr 0.000008 time 0.9210 (1.0065) model_time 0.9209 (1.0058) loss 0.9266 (0.8117) grad_norm 6.4983 (8.5414/1.8033) mem 68106MB [2022-12-20 10:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1380/1519] eta 0:02:19 lr 0.000008 time 0.9225 (1.0067) model_time 0.9224 (1.0060) loss 0.7118 (0.8116) grad_norm 7.7006 (8.4941/1.7019) mem 68106MB [2022-12-20 10:08:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1390/1519] eta 0:02:09 lr 0.000008 time 0.9220 (1.0066) model_time 0.9218 (1.0059) loss 1.0778 (0.8115) grad_norm 10.2786 (8.5030/1.7076) mem 68106MB [2022-12-20 10:08:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1400/1519] eta 0:01:59 lr 0.000008 time 0.9181 (1.0066) model_time 0.9180 (1.0059) loss 0.6849 (0.8114) grad_norm 7.6224 (8.4739/1.6958) mem 68106MB [2022-12-20 10:08:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1410/1519] eta 0:01:49 lr 0.000008 time 1.0205 (1.0066) model_time 1.0203 (1.0059) loss 0.7668 (0.8115) grad_norm 6.1881 (8.4563/1.6803) mem 68106MB [2022-12-20 10:08:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1420/1519] eta 0:01:39 lr 0.000008 time 0.9269 (1.0065) model_time 0.9268 (1.0058) loss 0.6826 (0.8116) grad_norm 10.0206 (8.4535/1.6810) mem 68106MB [2022-12-20 10:09:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1430/1519] eta 0:01:29 lr 0.000008 time 0.9194 (1.0065) model_time 0.9192 (1.0058) loss 0.7114 (0.8114) grad_norm 8.0046 (8.4387/1.6637) mem 68106MB [2022-12-20 10:09:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1440/1519] eta 0:01:19 lr 0.000008 time 0.9334 (1.0064) model_time 0.9333 (1.0057) loss 0.8366 (0.8118) grad_norm 6.0804 (8.4719/1.7749) mem 68106MB [2022-12-20 10:09:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1450/1519] eta 0:01:09 lr 0.000008 time 0.9182 (1.0064) model_time 0.9180 (1.0057) loss 0.7027 (0.8115) grad_norm 8.7723 (8.4680/1.7717) mem 68106MB [2022-12-20 10:09:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1460/1519] eta 0:00:59 lr 0.000008 time 0.9208 (1.0064) model_time 0.9207 (1.0057) loss 0.6984 (0.8109) grad_norm 7.1846 (8.4620/1.7731) mem 68106MB [2022-12-20 10:09:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1470/1519] eta 0:00:49 lr 0.000008 time 0.9254 (1.0064) model_time 0.9253 (1.0057) loss 0.8264 (0.8108) grad_norm 12.8135 (8.4740/1.7975) mem 68106MB [2022-12-20 10:09:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1480/1519] eta 0:00:39 lr 0.000008 time 0.9265 (1.0064) model_time 0.9262 (1.0057) loss 0.9930 (0.8107) grad_norm 8.1065 (8.4635/1.7917) mem 68106MB [2022-12-20 10:10:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1490/1519] eta 0:00:29 lr 0.000008 time 0.9352 (1.0063) model_time 0.9350 (1.0057) loss 0.7239 (0.8106) grad_norm 7.8521 (8.4756/1.7872) mem 68106MB [2022-12-20 10:10:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1500/1519] eta 0:00:19 lr 0.000008 time 0.9210 (1.0063) model_time 0.9209 (1.0057) loss 0.7400 (0.8106) grad_norm 7.4544 (8.4766/1.7761) mem 68106MB [2022-12-20 10:10:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [67/100][1510/1519] eta 0:00:09 lr 0.000008 time 0.9206 (1.0063) model_time 0.9205 (1.0056) loss 0.6913 (0.8107) grad_norm 7.7967 (8.4814/1.7830) mem 68106MB [2022-12-20 10:10:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 67 training takes 0:25:28 [2022-12-20 10:10:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_67.pth saving...... [2022-12-20 10:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_67.pth saved !!! [2022-12-20 10:11:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.692 (0.692) Loss 0.5234 (0.5234) Acc@1 91.319 (91.319) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 10:11:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.306 (0.334) Loss 0.5228 (0.4975) Acc@1 92.708 (92.771) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 10:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.317) Loss 0.4691 (0.4937) Acc@1 91.667 (92.841) Acc@5 99.306 (98.446) Mem 68106MB [2022-12-20 10:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.311) Loss 0.6365 (0.5009) Acc@1 89.583 (92.451) Acc@5 97.569 (98.376) Mem 68106MB [2022-12-20 10:11:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.311) Loss 0.4505 (0.4920) Acc@1 93.750 (92.505) Acc@5 99.306 (98.467) Mem 68106MB [2022-12-20 10:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.308) Loss 0.4830 (0.4897) Acc@1 91.319 (92.497) Acc@5 99.306 (98.509) Mem 68106MB [2022-12-20 10:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.307) Loss 0.5057 (0.4893) Acc@1 90.625 (92.464) Acc@5 97.917 (98.480) Mem 68106MB [2022-12-20 10:11:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.305) Loss 0.5378 (0.4907) Acc@1 92.708 (92.449) Acc@5 97.917 (98.479) Mem 68106MB [2022-12-20 10:11:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.304) Loss 0.4248 (0.4895) Acc@1 93.056 (92.464) Acc@5 98.611 (98.525) Mem 68106MB [2022-12-20 10:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:67] * Acc@1 92.424 Acc@5 98.527 [2022-12-20 10:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 10:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 10:11:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 10:11:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.42% [2022-12-20 10:11:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][0/1519] eta 0:34:45 lr 0.000008 time 1.3727 (1.3727) model_time 0.9655 (0.9655) loss 0.8600 (0.8600) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 10:12:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][10/1519] eta 0:26:05 lr 0.000008 time 0.9221 (1.0376) model_time 0.9220 (1.0002) loss 0.7424 (0.8449) grad_norm 8.9075 (7.9643/0.8053) mem 68106MB [2022-12-20 10:12:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][20/1519] eta 0:25:46 lr 0.000008 time 0.9386 (1.0314) model_time 0.9384 (1.0118) loss 1.0859 (0.8212) grad_norm 7.7897 (7.9673/0.6802) mem 68106MB [2022-12-20 10:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][30/1519] eta 0:25:19 lr 0.000008 time 0.9204 (1.0205) model_time 0.9203 (1.0071) loss 0.7215 (0.8053) grad_norm 7.8408 (8.0060/1.1520) mem 68106MB [2022-12-20 10:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][40/1519] eta 0:25:03 lr 0.000008 time 0.9247 (1.0165) model_time 0.9246 (1.0063) loss 0.6828 (0.8069) grad_norm 7.1086 (7.9938/1.4111) mem 68106MB [2022-12-20 10:12:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][50/1519] eta 0:24:50 lr 0.000008 time 0.9710 (1.0145) model_time 0.9708 (1.0062) loss 0.8009 (0.8008) grad_norm 9.5490 (8.4722/1.8262) mem 68106MB [2022-12-20 10:12:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][60/1519] eta 0:24:39 lr 0.000008 time 0.9371 (1.0140) model_time 0.9370 (1.0070) loss 0.7949 (0.7953) grad_norm 8.7351 (8.4045/1.8164) mem 68106MB [2022-12-20 10:13:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][70/1519] eta 0:24:26 lr 0.000008 time 0.9301 (1.0118) model_time 0.9299 (1.0058) loss 0.9957 (0.7940) grad_norm 8.0699 (8.4496/1.7623) mem 68106MB [2022-12-20 10:13:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][80/1519] eta 0:24:14 lr 0.000008 time 0.9217 (1.0106) model_time 0.9216 (1.0053) loss 0.8105 (0.7986) grad_norm 10.2087 (8.3544/1.7420) mem 68106MB [2022-12-20 10:13:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][90/1519] eta 0:24:03 lr 0.000008 time 0.9216 (1.0102) model_time 0.9215 (1.0055) loss 0.7151 (0.8019) grad_norm 7.1984 (8.3747/1.7228) mem 68106MB [2022-12-20 10:13:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][100/1519] eta 0:23:52 lr 0.000008 time 0.9365 (1.0097) model_time 0.9364 (1.0053) loss 0.6684 (0.7996) grad_norm 7.8178 (8.3009/1.6799) mem 68106MB [2022-12-20 10:13:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][110/1519] eta 0:23:42 lr 0.000008 time 0.9308 (1.0094) model_time 0.9307 (1.0055) loss 0.7579 (0.7994) grad_norm 8.6660 (8.2996/1.6173) mem 68106MB [2022-12-20 10:13:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][120/1519] eta 0:23:32 lr 0.000008 time 1.0325 (1.0097) model_time 1.0324 (1.0060) loss 0.8153 (0.7993) grad_norm 8.5920 (8.2702/1.5792) mem 68106MB [2022-12-20 10:14:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][130/1519] eta 0:23:22 lr 0.000008 time 0.9280 (1.0094) model_time 0.9278 (1.0060) loss 0.7855 (0.7988) grad_norm 8.2085 (8.2788/1.5399) mem 68106MB [2022-12-20 10:14:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][140/1519] eta 0:23:11 lr 0.000008 time 0.9814 (1.0094) model_time 0.9812 (1.0062) loss 0.7354 (0.7978) grad_norm 7.3455 (8.2598/1.6781) mem 68106MB [2022-12-20 10:14:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][150/1519] eta 0:23:01 lr 0.000008 time 0.9167 (1.0090) model_time 0.9166 (1.0060) loss 1.4336 (0.8041) grad_norm 11.8313 (8.3901/1.8567) mem 68106MB [2022-12-20 10:14:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][160/1519] eta 0:22:50 lr 0.000008 time 0.9298 (1.0082) model_time 0.9297 (1.0054) loss 0.7659 (0.8011) grad_norm 9.1287 (8.3618/1.8453) mem 68106MB [2022-12-20 10:14:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][170/1519] eta 0:22:40 lr 0.000008 time 0.9299 (1.0083) model_time 0.9297 (1.0057) loss 0.8278 (0.8011) grad_norm 9.1246 (8.4328/1.8294) mem 68106MB [2022-12-20 10:14:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][180/1519] eta 0:22:29 lr 0.000008 time 0.9236 (1.0079) model_time 0.9235 (1.0054) loss 0.9786 (0.8025) grad_norm 6.8364 (8.4279/1.8088) mem 68106MB [2022-12-20 10:15:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][190/1519] eta 0:22:19 lr 0.000008 time 0.9455 (1.0078) model_time 0.9454 (1.0054) loss 0.7204 (0.8055) grad_norm 7.8813 (8.3885/1.8017) mem 68106MB [2022-12-20 10:15:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][200/1519] eta 0:22:08 lr 0.000008 time 0.9231 (1.0074) model_time 0.9230 (1.0050) loss 0.7314 (0.8049) grad_norm 8.4826 (8.3655/1.7830) mem 68106MB [2022-12-20 10:15:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][210/1519] eta 0:21:58 lr 0.000008 time 0.9160 (1.0070) model_time 0.9159 (1.0048) loss 0.7745 (0.8027) grad_norm 9.6821 (8.4053/1.7843) mem 68106MB [2022-12-20 10:15:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][220/1519] eta 0:21:47 lr 0.000008 time 0.9211 (1.0068) model_time 0.9209 (1.0047) loss 1.0241 (0.8048) grad_norm 8.3621 (8.4079/1.7567) mem 68106MB [2022-12-20 10:15:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][230/1519] eta 0:21:37 lr 0.000008 time 0.9238 (1.0064) model_time 0.9236 (1.0043) loss 0.8477 (0.8067) grad_norm 6.2684 (8.3877/1.7355) mem 68106MB [2022-12-20 10:15:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][240/1519] eta 0:21:27 lr 0.000008 time 0.9393 (1.0067) model_time 0.9392 (1.0047) loss 0.7420 (0.8079) grad_norm 9.3786 (8.3854/1.7030) mem 68106MB [2022-12-20 10:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][250/1519] eta 0:21:17 lr 0.000008 time 0.9300 (1.0064) model_time 0.9299 (1.0044) loss 0.7316 (0.8100) grad_norm 7.2958 (8.3776/1.6905) mem 68106MB [2022-12-20 10:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][260/1519] eta 0:21:06 lr 0.000008 time 0.9199 (1.0061) model_time 0.9197 (1.0042) loss 1.0001 (0.8105) grad_norm 8.0106 (8.4158/1.6874) mem 68106MB [2022-12-20 10:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][270/1519] eta 0:20:56 lr 0.000008 time 0.9282 (1.0061) model_time 0.9281 (1.0043) loss 0.6862 (0.8096) grad_norm 12.9978 (8.4436/1.7052) mem 68106MB [2022-12-20 10:16:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][280/1519] eta 0:20:46 lr 0.000008 time 0.9287 (1.0062) model_time 0.9286 (1.0044) loss 0.7364 (0.8073) grad_norm 8.6942 (8.4484/1.6956) mem 68106MB [2022-12-20 10:16:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][290/1519] eta 0:20:37 lr 0.000008 time 0.9293 (1.0070) model_time 0.9291 (1.0053) loss 0.8672 (0.8085) grad_norm 7.6920 (8.4733/1.7147) mem 68106MB [2022-12-20 10:16:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][300/1519] eta 0:20:27 lr 0.000008 time 0.9287 (1.0068) model_time 0.9285 (1.0052) loss 0.6681 (0.8082) grad_norm 7.2991 (8.5265/1.7377) mem 68106MB [2022-12-20 10:17:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][310/1519] eta 0:20:17 lr 0.000008 time 0.9159 (1.0074) model_time 0.9158 (1.0058) loss 0.7149 (0.8086) grad_norm 7.7249 (8.5058/1.7173) mem 68106MB [2022-12-20 10:17:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][320/1519] eta 0:20:08 lr 0.000008 time 0.9858 (1.0081) model_time 0.9856 (1.0065) loss 0.8231 (0.8079) grad_norm 7.5410 (8.5299/1.7346) mem 68106MB [2022-12-20 10:17:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][330/1519] eta 0:19:58 lr 0.000008 time 0.9247 (1.0082) model_time 0.9245 (1.0066) loss 0.9594 (0.8071) grad_norm 7.8606 (8.4797/1.7367) mem 68106MB [2022-12-20 10:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][340/1519] eta 0:19:48 lr 0.000008 time 0.9261 (1.0079) model_time 0.9260 (1.0064) loss 0.6892 (0.8077) grad_norm 10.0943 (8.4902/1.7244) mem 68106MB [2022-12-20 10:17:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][350/1519] eta 0:19:38 lr 0.000008 time 0.9293 (1.0083) model_time 0.9292 (1.0068) loss 0.6746 (0.8086) grad_norm 7.6736 (8.4709/1.7161) mem 68106MB [2022-12-20 10:17:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][360/1519] eta 0:19:28 lr 0.000008 time 0.9708 (1.0081) model_time 0.9706 (1.0067) loss 0.6883 (0.8080) grad_norm 7.3044 (8.4693/1.7042) mem 68106MB [2022-12-20 10:18:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][370/1519] eta 0:19:18 lr 0.000008 time 0.9278 (1.0080) model_time 0.9276 (1.0066) loss 0.8324 (0.8079) grad_norm 9.9282 (8.4553/1.6977) mem 68106MB [2022-12-20 10:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][380/1519] eta 0:19:07 lr 0.000008 time 0.9274 (1.0079) model_time 0.9272 (1.0065) loss 0.8093 (0.8084) grad_norm 9.7137 (8.4573/1.6821) mem 68106MB [2022-12-20 10:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][390/1519] eta 0:18:57 lr 0.000008 time 0.9390 (1.0079) model_time 0.9389 (1.0065) loss 0.6844 (0.8068) grad_norm 9.6536 (8.4539/1.6794) mem 68106MB [2022-12-20 10:18:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][400/1519] eta 0:18:47 lr 0.000008 time 0.9272 (1.0077) model_time 0.9270 (1.0063) loss 0.8222 (0.8069) grad_norm 8.4775 (8.4512/1.6663) mem 68106MB [2022-12-20 10:18:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][410/1519] eta 0:18:37 lr 0.000008 time 0.9288 (1.0075) model_time 0.9287 (1.0062) loss 0.7023 (0.8061) grad_norm 8.9261 (8.4612/1.6477) mem 68106MB [2022-12-20 10:18:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][420/1519] eta 0:18:27 lr 0.000008 time 0.9266 (1.0075) model_time 0.9264 (1.0062) loss 1.1037 (0.8070) grad_norm 8.1884 (8.4766/1.6405) mem 68106MB [2022-12-20 10:19:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][430/1519] eta 0:18:16 lr 0.000008 time 0.9187 (1.0073) model_time 0.9185 (1.0060) loss 0.8429 (0.8074) grad_norm 8.0502 (8.4834/1.6420) mem 68106MB [2022-12-20 10:19:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][440/1519] eta 0:18:06 lr 0.000008 time 0.9333 (1.0074) model_time 0.9331 (1.0062) loss 0.7755 (0.8065) grad_norm 6.3055 (8.4645/1.6425) mem 68106MB [2022-12-20 10:19:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][450/1519] eta 0:17:56 lr 0.000008 time 0.9229 (1.0073) model_time 0.9228 (1.0061) loss 0.6707 (0.8063) grad_norm 13.8959 (8.4773/1.6758) mem 68106MB [2022-12-20 10:19:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][460/1519] eta 0:17:46 lr 0.000008 time 0.9286 (1.0071) model_time 0.9284 (1.0059) loss 0.9252 (0.8063) grad_norm 5.4012 (8.4605/1.6739) mem 68106MB [2022-12-20 10:19:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][470/1519] eta 0:17:36 lr 0.000008 time 0.9206 (1.0074) model_time 0.9205 (1.0062) loss 0.6574 (0.8054) grad_norm 8.7805 (8.4609/1.6568) mem 68106MB [2022-12-20 10:19:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][480/1519] eta 0:17:26 lr 0.000008 time 0.9241 (1.0072) model_time 0.9239 (1.0061) loss 0.7540 (0.8058) grad_norm 8.8003 (8.4734/1.6534) mem 68106MB [2022-12-20 10:20:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][490/1519] eta 0:17:16 lr 0.000008 time 0.9351 (1.0071) model_time 0.9349 (1.0060) loss 0.7859 (0.8065) grad_norm 10.9876 (8.4809/1.6533) mem 68106MB [2022-12-20 10:20:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][500/1519] eta 0:17:06 lr 0.000008 time 0.9251 (1.0073) model_time 0.9249 (1.0062) loss 0.8497 (0.8068) grad_norm 9.9225 (8.4772/1.6513) mem 68106MB [2022-12-20 10:20:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][510/1519] eta 0:16:56 lr 0.000008 time 0.9395 (1.0072) model_time 0.9393 (1.0061) loss 0.7538 (0.8064) grad_norm 11.7173 (8.4915/1.6517) mem 68106MB [2022-12-20 10:20:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][520/1519] eta 0:16:46 lr 0.000008 time 0.9278 (1.0070) model_time 0.9276 (1.0059) loss 0.7744 (0.8056) grad_norm 7.1837 (8.4954/1.6499) mem 68106MB [2022-12-20 10:20:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][530/1519] eta 0:16:35 lr 0.000008 time 0.9093 (1.0069) model_time 0.9092 (1.0059) loss 0.6762 (0.8065) grad_norm 10.5697 (8.5072/1.6539) mem 68106MB [2022-12-20 10:20:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][540/1519] eta 0:16:25 lr 0.000008 time 0.9833 (1.0069) model_time 0.9831 (1.0058) loss 1.2444 (0.8071) grad_norm 11.3134 (8.5881/1.9440) mem 68106MB [2022-12-20 10:21:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][550/1519] eta 0:16:15 lr 0.000008 time 0.9745 (1.0068) model_time 0.9743 (1.0057) loss 0.6916 (0.8084) grad_norm 10.0375 (8.5877/1.9390) mem 68106MB [2022-12-20 10:21:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][560/1519] eta 0:16:05 lr 0.000008 time 0.9333 (1.0069) model_time 0.9331 (1.0058) loss 0.9061 (0.8091) grad_norm 7.5900 (8.5738/1.9393) mem 68106MB [2022-12-20 10:21:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][570/1519] eta 0:15:55 lr 0.000008 time 0.9230 (1.0067) model_time 0.9228 (1.0057) loss 0.7287 (0.8101) grad_norm 16.2677 (8.6108/1.9789) mem 68106MB [2022-12-20 10:21:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][580/1519] eta 0:15:45 lr 0.000008 time 0.9307 (1.0066) model_time 0.9306 (1.0056) loss 0.6875 (0.8100) grad_norm 9.0252 (8.6135/1.9641) mem 68106MB [2022-12-20 10:21:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][590/1519] eta 0:15:35 lr 0.000008 time 0.9286 (1.0066) model_time 0.9285 (1.0056) loss 0.7075 (0.8099) grad_norm 7.4603 (8.6014/1.9725) mem 68106MB [2022-12-20 10:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][600/1519] eta 0:15:25 lr 0.000008 time 0.9271 (1.0068) model_time 0.9270 (1.0058) loss 0.9696 (0.8094) grad_norm 7.2058 (8.5888/1.9676) mem 68106MB [2022-12-20 10:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][610/1519] eta 0:15:15 lr 0.000008 time 0.9276 (1.0066) model_time 0.9275 (1.0057) loss 0.9602 (0.8098) grad_norm 6.3820 (8.5917/1.9762) mem 68106MB [2022-12-20 10:22:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][620/1519] eta 0:15:05 lr 0.000008 time 0.9223 (1.0069) model_time 0.9222 (1.0059) loss 0.8206 (0.8107) grad_norm 9.1277 (8.6178/1.9866) mem 68106MB [2022-12-20 10:22:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][630/1519] eta 0:14:55 lr 0.000008 time 0.9308 (1.0068) model_time 0.9306 (1.0058) loss 0.7163 (0.8115) grad_norm 6.7097 (8.6089/1.9823) mem 68106MB [2022-12-20 10:22:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][640/1519] eta 0:14:45 lr 0.000008 time 0.9300 (1.0068) model_time 0.9298 (1.0059) loss 0.7775 (0.8116) grad_norm 6.2210 (8.6175/1.9742) mem 68106MB [2022-12-20 10:22:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][650/1519] eta 0:14:34 lr 0.000008 time 0.9267 (1.0068) model_time 0.9265 (1.0058) loss 1.2279 (0.8123) grad_norm 8.3706 (8.5918/1.9556) mem 68106MB [2022-12-20 10:22:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][660/1519] eta 0:14:24 lr 0.000008 time 0.9097 (1.0067) model_time 0.9096 (1.0058) loss 0.7300 (0.8119) grad_norm 7.7851 (8.6104/1.9937) mem 68106MB [2022-12-20 10:23:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][670/1519] eta 0:14:14 lr 0.000008 time 0.9269 (1.0066) model_time 0.9268 (1.0057) loss 0.8671 (0.8118) grad_norm 11.2072 (8.6153/2.0064) mem 68106MB [2022-12-20 10:23:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][680/1519] eta 0:14:04 lr 0.000008 time 0.9308 (1.0065) model_time 0.9307 (1.0056) loss 1.1846 (0.8120) grad_norm 8.6675 (8.6047/2.0103) mem 68106MB [2022-12-20 10:23:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][690/1519] eta 0:13:54 lr 0.000008 time 0.9343 (1.0064) model_time 0.9341 (1.0055) loss 0.9065 (0.8122) grad_norm 7.3368 (8.5888/2.0083) mem 68106MB [2022-12-20 10:23:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][700/1519] eta 0:13:44 lr 0.000008 time 0.9948 (1.0066) model_time 0.9947 (1.0057) loss 1.0374 (0.8127) grad_norm 7.4494 (8.6080/2.0103) mem 68106MB [2022-12-20 10:23:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][710/1519] eta 0:13:34 lr 0.000008 time 0.9391 (1.0065) model_time 0.9390 (1.0056) loss 1.0715 (0.8126) grad_norm 10.4804 (8.6056/2.0184) mem 68106MB [2022-12-20 10:23:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][720/1519] eta 0:13:24 lr 0.000008 time 0.9243 (1.0064) model_time 0.9241 (1.0056) loss 0.7672 (0.8129) grad_norm 8.5965 (8.6010/2.0189) mem 68106MB [2022-12-20 10:24:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][730/1519] eta 0:13:14 lr 0.000008 time 0.9486 (1.0064) model_time 0.9485 (1.0055) loss 0.8934 (0.8131) grad_norm 8.1551 (8.6113/2.0294) mem 68106MB [2022-12-20 10:24:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][740/1519] eta 0:13:03 lr 0.000008 time 0.9273 (1.0063) model_time 0.9271 (1.0054) loss 0.8122 (0.8131) grad_norm 8.3591 (8.6251/1.9954) mem 68106MB [2022-12-20 10:24:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][750/1519] eta 0:12:53 lr 0.000008 time 0.9321 (1.0062) model_time 0.9320 (1.0053) loss 0.8929 (0.8131) grad_norm 5.9760 (8.5893/1.9649) mem 68106MB [2022-12-20 10:24:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][760/1519] eta 0:12:43 lr 0.000008 time 0.9260 (1.0061) model_time 0.9258 (1.0052) loss 0.6889 (0.8125) grad_norm 7.7456 (8.5939/1.9549) mem 68106MB [2022-12-20 10:24:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][770/1519] eta 0:12:33 lr 0.000008 time 0.9342 (1.0060) model_time 0.9336 (1.0051) loss 0.8859 (0.8118) grad_norm 8.0925 (8.5607/1.9609) mem 68106MB [2022-12-20 10:24:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][780/1519] eta 0:12:23 lr 0.000008 time 0.9267 (1.0060) model_time 0.9266 (1.0052) loss 0.7219 (0.8110) grad_norm 7.8447 (8.5804/1.9617) mem 68106MB [2022-12-20 10:25:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][790/1519] eta 0:12:13 lr 0.000008 time 0.9268 (1.0060) model_time 0.9265 (1.0052) loss 0.9782 (0.8110) grad_norm 10.1051 (8.5859/1.9569) mem 68106MB [2022-12-20 10:25:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][800/1519] eta 0:12:03 lr 0.000008 time 0.9314 (1.0060) model_time 0.9312 (1.0052) loss 0.6659 (0.8105) grad_norm 8.4591 (8.5965/1.9617) mem 68106MB [2022-12-20 10:25:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][810/1519] eta 0:11:53 lr 0.000008 time 0.9285 (1.0061) model_time 0.9281 (1.0052) loss 0.8372 (0.8106) grad_norm 8.0286 (8.6031/1.9589) mem 68106MB [2022-12-20 10:25:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][820/1519] eta 0:11:43 lr 0.000008 time 0.9271 (1.0060) model_time 0.9269 (1.0051) loss 0.7247 (0.8101) grad_norm 7.8132 (8.5999/1.9646) mem 68106MB [2022-12-20 10:25:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][830/1519] eta 0:11:33 lr 0.000008 time 0.9356 (1.0059) model_time 0.9355 (1.0050) loss 0.6817 (0.8096) grad_norm 7.1038 (8.6195/1.9992) mem 68106MB [2022-12-20 10:25:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][840/1519] eta 0:11:22 lr 0.000008 time 0.9263 (1.0058) model_time 0.9262 (1.0049) loss 0.9291 (0.8101) grad_norm 7.2085 (8.5939/2.0132) mem 68106MB [2022-12-20 10:26:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][850/1519] eta 0:11:12 lr 0.000008 time 0.9192 (1.0057) model_time 0.9191 (1.0049) loss 1.1038 (0.8104) grad_norm 8.1258 (8.6040/2.0096) mem 68106MB [2022-12-20 10:26:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][860/1519] eta 0:11:02 lr 0.000008 time 0.9201 (1.0057) model_time 0.9199 (1.0049) loss 0.9533 (0.8112) grad_norm 7.5897 (8.6116/2.0211) mem 68106MB [2022-12-20 10:26:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][870/1519] eta 0:10:52 lr 0.000008 time 0.9284 (1.0057) model_time 0.9283 (1.0049) loss 0.9074 (0.8121) grad_norm 10.7418 (8.6232/2.0201) mem 68106MB [2022-12-20 10:26:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][880/1519] eta 0:10:42 lr 0.000008 time 0.9237 (1.0057) model_time 0.9236 (1.0049) loss 0.8618 (0.8122) grad_norm 11.6221 (8.6320/2.0276) mem 68106MB [2022-12-20 10:26:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][890/1519] eta 0:10:32 lr 0.000008 time 0.9560 (1.0056) model_time 0.9558 (1.0048) loss 0.7629 (0.8121) grad_norm 6.4571 (8.6184/2.0199) mem 68106MB [2022-12-20 10:26:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][900/1519] eta 0:10:22 lr 0.000008 time 0.9584 (1.0057) model_time 0.9583 (1.0050) loss 0.7047 (0.8120) grad_norm 9.4046 (8.5780/2.0079) mem 68106MB [2022-12-20 10:27:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][910/1519] eta 0:10:12 lr 0.000008 time 0.9582 (1.0059) model_time 0.9581 (1.0051) loss 1.0721 (0.8119) grad_norm 8.7944 (8.6058/2.0118) mem 68106MB [2022-12-20 10:27:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][920/1519] eta 0:10:02 lr 0.000008 time 0.9289 (1.0058) model_time 0.9288 (1.0050) loss 0.9343 (0.8123) grad_norm 7.1773 (8.5846/1.9976) mem 68106MB [2022-12-20 10:27:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][930/1519] eta 0:09:52 lr 0.000008 time 0.9455 (1.0059) model_time 0.9452 (1.0052) loss 0.6793 (0.8119) grad_norm 7.3015 (8.6158/1.9938) mem 68106MB [2022-12-20 10:27:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][940/1519] eta 0:09:42 lr 0.000008 time 0.9245 (1.0059) model_time 0.9244 (1.0052) loss 1.0513 (0.8127) grad_norm 5.7800 (8.5929/1.9964) mem 68106MB [2022-12-20 10:27:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][950/1519] eta 0:09:32 lr 0.000008 time 0.9249 (1.0060) model_time 0.9247 (1.0052) loss 0.8067 (0.8123) grad_norm 8.3110 (8.5932/1.9916) mem 68106MB [2022-12-20 10:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][960/1519] eta 0:09:22 lr 0.000008 time 0.9352 (1.0059) model_time 0.9350 (1.0052) loss 0.7659 (0.8124) grad_norm 6.6490 (8.5964/1.9901) mem 68106MB [2022-12-20 10:28:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][970/1519] eta 0:09:12 lr 0.000008 time 0.9195 (1.0059) model_time 0.9194 (1.0051) loss 0.8778 (0.8121) grad_norm 9.6671 (8.6005/1.9936) mem 68106MB [2022-12-20 10:28:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][980/1519] eta 0:09:02 lr 0.000008 time 0.9308 (1.0059) model_time 0.9301 (1.0051) loss 0.6646 (0.8120) grad_norm 9.2893 (8.5942/1.9950) mem 68106MB [2022-12-20 10:28:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][990/1519] eta 0:08:52 lr 0.000008 time 0.9119 (1.0059) model_time 0.9117 (1.0052) loss 0.7822 (0.8118) grad_norm 10.5884 (8.5931/1.9923) mem 68106MB [2022-12-20 10:28:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1000/1519] eta 0:08:42 lr 0.000008 time 0.9337 (1.0058) model_time 0.9335 (1.0051) loss 0.7960 (0.8116) grad_norm 6.6311 (8.5842/1.9971) mem 68106MB [2022-12-20 10:28:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1010/1519] eta 0:08:31 lr 0.000008 time 0.9730 (1.0058) model_time 0.9728 (1.0051) loss 0.7276 (0.8114) grad_norm 7.9824 (8.5636/2.0089) mem 68106MB [2022-12-20 10:28:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1020/1519] eta 0:08:21 lr 0.000008 time 0.9286 (1.0058) model_time 0.9284 (1.0050) loss 0.7752 (0.8114) grad_norm 8.6249 (8.5500/2.0229) mem 68106MB [2022-12-20 10:29:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1030/1519] eta 0:08:11 lr 0.000008 time 0.9309 (1.0057) model_time 0.9307 (1.0050) loss 0.8388 (0.8121) grad_norm 7.2423 (8.5298/2.0165) mem 68106MB [2022-12-20 10:29:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1040/1519] eta 0:08:01 lr 0.000008 time 0.9320 (1.0056) model_time 0.9318 (1.0049) loss 0.6673 (0.8117) grad_norm 7.4993 (8.5579/2.0249) mem 68106MB [2022-12-20 10:29:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1050/1519] eta 0:07:51 lr 0.000008 time 0.9331 (1.0056) model_time 0.9329 (1.0048) loss 0.8256 (0.8113) grad_norm 9.0992 (8.5346/2.0007) mem 68106MB [2022-12-20 10:29:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1060/1519] eta 0:07:41 lr 0.000008 time 0.9251 (1.0055) model_time 0.9250 (1.0048) loss 0.8492 (0.8116) grad_norm 9.5787 (8.5559/1.9923) mem 68106MB [2022-12-20 10:29:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1070/1519] eta 0:07:31 lr 0.000008 time 0.9330 (1.0054) model_time 0.9329 (1.0047) loss 0.7143 (0.8123) grad_norm 7.3879 (8.5540/2.0066) mem 68106MB [2022-12-20 10:29:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1080/1519] eta 0:07:21 lr 0.000008 time 0.9319 (1.0056) model_time 0.9315 (1.0049) loss 0.8284 (0.8124) grad_norm 8.7019 (8.5264/2.0059) mem 68106MB [2022-12-20 10:30:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1090/1519] eta 0:07:11 lr 0.000008 time 0.9347 (1.0057) model_time 0.9345 (1.0049) loss 0.8045 (0.8125) grad_norm 6.9241 (8.4972/2.0049) mem 68106MB [2022-12-20 10:30:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1100/1519] eta 0:07:01 lr 0.000008 time 0.9292 (1.0056) model_time 0.9290 (1.0049) loss 1.0452 (0.8136) grad_norm 9.3907 (8.4961/1.9991) mem 68106MB [2022-12-20 10:30:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1110/1519] eta 0:06:51 lr 0.000008 time 0.9016 (1.0058) model_time 0.9014 (1.0051) loss 0.6931 (0.8141) grad_norm 8.1825 (8.4692/1.9920) mem 68106MB [2022-12-20 10:30:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1120/1519] eta 0:06:41 lr 0.000008 time 0.9319 (1.0058) model_time 0.9317 (1.0051) loss 0.6674 (0.8137) grad_norm 7.8929 (8.4592/1.9837) mem 68106MB [2022-12-20 10:30:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1130/1519] eta 0:06:31 lr 0.000008 time 0.9246 (1.0058) model_time 0.9244 (1.0050) loss 0.7076 (0.8138) grad_norm 6.6065 (8.4424/1.9752) mem 68106MB [2022-12-20 10:30:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1140/1519] eta 0:06:21 lr 0.000008 time 0.9369 (1.0058) model_time 0.9367 (1.0050) loss 0.8463 (0.8138) grad_norm 7.0630 (8.3483/1.7210) mem 68106MB [2022-12-20 10:31:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1150/1519] eta 0:06:11 lr 0.000008 time 0.9208 (1.0057) model_time 0.9207 (1.0050) loss 0.9355 (0.8137) grad_norm 6.4512 (8.3435/1.7133) mem 68106MB [2022-12-20 10:31:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1160/1519] eta 0:06:01 lr 0.000008 time 0.9221 (1.0058) model_time 0.9219 (1.0051) loss 0.6637 (0.8141) grad_norm 7.5449 (8.3499/1.7017) mem 68106MB [2022-12-20 10:31:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1170/1519] eta 0:05:51 lr 0.000008 time 0.9300 (1.0058) model_time 0.9298 (1.0051) loss 0.7013 (0.8143) grad_norm 8.1304 (8.2994/1.6428) mem 68106MB [2022-12-20 10:31:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1180/1519] eta 0:05:40 lr 0.000008 time 0.9293 (1.0058) model_time 0.9291 (1.0051) loss 1.0562 (0.8143) grad_norm 8.6954 (8.2861/1.6418) mem 68106MB [2022-12-20 10:31:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1190/1519] eta 0:05:30 lr 0.000008 time 0.9302 (1.0057) model_time 0.9301 (1.0050) loss 0.6767 (0.8136) grad_norm 11.6056 (8.2958/1.6296) mem 68106MB [2022-12-20 10:31:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1200/1519] eta 0:05:20 lr 0.000008 time 0.9304 (1.0057) model_time 0.9303 (1.0050) loss 0.9157 (0.8138) grad_norm 13.1555 (8.3504/1.6956) mem 68106MB [2022-12-20 10:32:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1210/1519] eta 0:05:10 lr 0.000008 time 0.9328 (1.0056) model_time 0.9326 (1.0049) loss 0.6645 (0.8131) grad_norm 7.4713 (8.3608/1.6927) mem 68106MB [2022-12-20 10:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1220/1519] eta 0:05:00 lr 0.000008 time 0.9850 (1.0056) model_time 0.9848 (1.0049) loss 0.6964 (0.8126) grad_norm 8.3235 (8.3500/1.6801) mem 68106MB [2022-12-20 10:32:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1230/1519] eta 0:04:50 lr 0.000008 time 0.9304 (1.0058) model_time 0.9303 (1.0051) loss 0.6735 (0.8124) grad_norm 7.5592 (8.3757/1.6954) mem 68106MB [2022-12-20 10:32:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1240/1519] eta 0:04:40 lr 0.000008 time 0.9355 (1.0058) model_time 0.9353 (1.0051) loss 1.0756 (0.8122) grad_norm 5.8059 (8.3672/1.7053) mem 68106MB [2022-12-20 10:32:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1250/1519] eta 0:04:30 lr 0.000008 time 0.9495 (1.0058) model_time 0.9493 (1.0051) loss 1.0161 (0.8127) grad_norm 8.8149 (8.3655/1.6935) mem 68106MB [2022-12-20 10:32:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1260/1519] eta 0:04:20 lr 0.000008 time 0.9332 (1.0058) model_time 0.9329 (1.0051) loss 0.8429 (0.8125) grad_norm 7.5932 (8.3397/1.6348) mem 68106MB [2022-12-20 10:33:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1270/1519] eta 0:04:10 lr 0.000008 time 0.9354 (1.0058) model_time 0.9352 (1.0051) loss 0.6809 (0.8126) grad_norm 7.0306 (8.3279/1.6193) mem 68106MB [2022-12-20 10:33:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1280/1519] eta 0:04:00 lr 0.000008 time 0.9349 (1.0057) model_time 0.9348 (1.0051) loss 1.0624 (0.8130) grad_norm 7.8006 (8.3463/1.6086) mem 68106MB [2022-12-20 10:33:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1290/1519] eta 0:03:50 lr 0.000008 time 0.9326 (1.0058) model_time 0.9325 (1.0051) loss 0.7396 (0.8135) grad_norm 7.5945 (8.3713/1.6198) mem 68106MB [2022-12-20 10:33:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1300/1519] eta 0:03:40 lr 0.000008 time 0.9292 (1.0057) model_time 0.9291 (1.0051) loss 0.8954 (0.8140) grad_norm 14.2258 (8.3865/1.6490) mem 68106MB [2022-12-20 10:33:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1310/1519] eta 0:03:30 lr 0.000008 time 0.9305 (1.0057) model_time 0.9304 (1.0050) loss 0.7024 (0.8139) grad_norm 9.3600 (8.3873/1.6508) mem 68106MB [2022-12-20 10:33:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1320/1519] eta 0:03:20 lr 0.000008 time 0.9324 (1.0056) model_time 0.9322 (1.0050) loss 0.6754 (0.8138) grad_norm 6.4445 (8.3855/1.6521) mem 68106MB [2022-12-20 10:34:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1330/1519] eta 0:03:10 lr 0.000008 time 0.9263 (1.0057) model_time 0.9261 (1.0050) loss 0.8895 (0.8137) grad_norm 5.8242 (8.3548/1.6449) mem 68106MB [2022-12-20 10:34:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1340/1519] eta 0:03:00 lr 0.000008 time 0.9332 (1.0056) model_time 0.9331 (1.0050) loss 1.0144 (0.8139) grad_norm 7.8816 (8.3413/1.6469) mem 68106MB [2022-12-20 10:34:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1350/1519] eta 0:02:49 lr 0.000008 time 0.9360 (1.0056) model_time 0.9359 (1.0049) loss 0.7873 (0.8133) grad_norm 6.7063 (8.3453/1.6680) mem 68106MB [2022-12-20 10:34:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1360/1519] eta 0:02:39 lr 0.000008 time 0.9318 (1.0056) model_time 0.9315 (1.0050) loss 1.0510 (0.8131) grad_norm 8.9122 (8.3543/1.6804) mem 68106MB [2022-12-20 10:34:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1370/1519] eta 0:02:29 lr 0.000008 time 0.9308 (1.0056) model_time 0.9307 (1.0049) loss 0.6697 (0.8131) grad_norm 6.2732 (8.3633/1.6710) mem 68106MB [2022-12-20 10:34:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1380/1519] eta 0:02:19 lr 0.000008 time 0.9309 (1.0055) model_time 0.9308 (1.0049) loss 0.8677 (0.8130) grad_norm 15.0588 (8.3504/1.7093) mem 68106MB [2022-12-20 10:35:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1390/1519] eta 0:02:09 lr 0.000008 time 0.9107 (1.0056) model_time 0.9106 (1.0049) loss 1.2774 (0.8133) grad_norm 11.2234 (8.3502/1.7146) mem 68106MB [2022-12-20 10:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1400/1519] eta 0:01:59 lr 0.000008 time 0.9792 (1.0056) model_time 0.9791 (1.0049) loss 0.8130 (0.8132) grad_norm 7.3701 (8.3448/1.7089) mem 68106MB [2022-12-20 10:35:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1410/1519] eta 0:01:49 lr 0.000008 time 0.9295 (1.0056) model_time 0.9294 (1.0050) loss 1.1139 (0.8131) grad_norm 8.8202 (8.3212/1.7008) mem 68106MB [2022-12-20 10:35:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1420/1519] eta 0:01:39 lr 0.000008 time 0.9293 (1.0056) model_time 0.9291 (1.0049) loss 0.7054 (0.8133) grad_norm 8.4114 (8.3171/1.7014) mem 68106MB [2022-12-20 10:35:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1430/1519] eta 0:01:29 lr 0.000008 time 1.0024 (1.0056) model_time 1.0023 (1.0050) loss 0.6683 (0.8128) grad_norm 9.3864 (8.3164/1.6599) mem 68106MB [2022-12-20 10:36:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1440/1519] eta 0:01:19 lr 0.000008 time 0.9346 (1.0056) model_time 0.9345 (1.0050) loss 0.6673 (0.8133) grad_norm 13.2686 (8.3627/1.6858) mem 68106MB [2022-12-20 10:36:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1450/1519] eta 0:01:09 lr 0.000008 time 0.9320 (1.0057) model_time 0.9319 (1.0050) loss 0.6879 (0.8127) grad_norm 6.4563 (8.3432/1.6863) mem 68106MB [2022-12-20 10:36:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1460/1519] eta 0:00:59 lr 0.000008 time 0.9256 (1.0056) model_time 0.9254 (1.0050) loss 0.7765 (0.8127) grad_norm 6.4635 (8.3018/1.6643) mem 68106MB [2022-12-20 10:36:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1470/1519] eta 0:00:49 lr 0.000008 time 0.9348 (1.0056) model_time 0.9347 (1.0050) loss 0.8951 (0.8130) grad_norm 7.5634 (8.2730/1.6447) mem 68106MB [2022-12-20 10:36:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1480/1519] eta 0:00:39 lr 0.000008 time 0.9343 (1.0056) model_time 0.9341 (1.0050) loss 0.7435 (0.8125) grad_norm 13.1404 (8.2686/1.6521) mem 68106MB [2022-12-20 10:36:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1490/1519] eta 0:00:29 lr 0.000008 time 0.9313 (1.0056) model_time 0.9311 (1.0050) loss 0.6514 (0.8126) grad_norm 10.5947 (8.2709/1.6487) mem 68106MB [2022-12-20 10:37:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1500/1519] eta 0:00:19 lr 0.000008 time 0.9304 (1.0056) model_time 0.9302 (1.0050) loss 0.7209 (0.8123) grad_norm 6.5632 (8.2844/1.6594) mem 68106MB [2022-12-20 10:37:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [68/100][1510/1519] eta 0:00:09 lr 0.000008 time 0.9119 (1.0056) model_time 0.9117 (1.0049) loss 0.7372 (0.8117) grad_norm 8.6998 (8.3010/1.6881) mem 68106MB [2022-12-20 10:37:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 68 training takes 0:25:27 [2022-12-20 10:37:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_68.pth saving...... [2022-12-20 10:37:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_68.pth saved !!! [2022-12-20 10:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.709 (0.709) Loss 0.5236 (0.5236) Acc@1 92.014 (92.014) Acc@5 98.958 (98.958) Mem 68106MB [2022-12-20 10:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.301 (0.334) Loss 0.5270 (0.4943) Acc@1 92.014 (92.771) Acc@5 97.917 (98.516) Mem 68106MB [2022-12-20 10:37:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.317) Loss 0.4706 (0.4918) Acc@1 92.361 (92.774) Acc@5 98.958 (98.462) Mem 68106MB [2022-12-20 10:37:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.311) Loss 0.6248 (0.4988) Acc@1 89.583 (92.440) Acc@5 97.569 (98.376) Mem 68106MB [2022-12-20 10:37:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.308) Loss 0.4509 (0.4900) Acc@1 93.750 (92.530) Acc@5 99.306 (98.467) Mem 68106MB [2022-12-20 10:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.4719 (0.4874) Acc@1 92.014 (92.531) Acc@5 99.653 (98.529) Mem 68106MB [2022-12-20 10:38:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.305) Loss 0.5041 (0.4869) Acc@1 90.972 (92.509) Acc@5 97.917 (98.526) Mem 68106MB [2022-12-20 10:38:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5377 (0.4884) Acc@1 93.056 (92.469) Acc@5 97.917 (98.523) Mem 68106MB [2022-12-20 10:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.303 (0.303) Loss 0.4247 (0.4869) Acc@1 92.708 (92.503) Acc@5 98.611 (98.555) Mem 68106MB [2022-12-20 10:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:68] * Acc@1 92.461 Acc@5 98.551 [2022-12-20 10:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 10:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 10:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 10:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.46% [2022-12-20 10:38:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][0/1519] eta 0:35:15 lr 0.000008 time 1.3924 (1.3924) model_time 1.0039 (1.0039) loss 0.9987 (0.9987) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 10:38:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][10/1519] eta 0:25:57 lr 0.000008 time 0.9319 (1.0319) model_time 0.9317 (0.9961) loss 0.8451 (0.8894) grad_norm 8.4439 (8.1423/0.4196) mem 68106MB [2022-12-20 10:38:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][20/1519] eta 0:25:31 lr 0.000008 time 0.9252 (1.0216) model_time 0.9251 (1.0027) loss 0.7481 (0.8305) grad_norm 10.3839 (9.1670/1.9166) mem 68106MB [2022-12-20 10:39:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][30/1519] eta 0:25:25 lr 0.000008 time 0.9667 (1.0242) model_time 0.9666 (1.0113) loss 0.8495 (0.8200) grad_norm 11.6585 (9.0184/1.8933) mem 68106MB [2022-12-20 10:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][40/1519] eta 0:25:05 lr 0.000008 time 0.9268 (1.0176) model_time 0.9267 (1.0078) loss 0.8266 (0.8177) grad_norm 11.1859 (8.6913/1.9682) mem 68106MB [2022-12-20 10:39:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][50/1519] eta 0:24:50 lr 0.000008 time 0.9466 (1.0143) model_time 0.9464 (1.0063) loss 0.9985 (0.8205) grad_norm 6.0671 (9.3024/3.7783) mem 68106MB [2022-12-20 10:39:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][60/1519] eta 0:24:39 lr 0.000008 time 0.9330 (1.0142) model_time 0.9328 (1.0074) loss 0.8018 (0.8096) grad_norm 7.0441 (8.9321/3.5548) mem 68106MB [2022-12-20 10:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][70/1519] eta 0:24:27 lr 0.000008 time 0.9244 (1.0125) model_time 0.9243 (1.0066) loss 0.9260 (0.8112) grad_norm 8.8127 (8.7867/3.3310) mem 68106MB [2022-12-20 10:39:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][80/1519] eta 0:24:14 lr 0.000008 time 0.9330 (1.0108) model_time 0.9329 (1.0057) loss 0.7110 (0.8151) grad_norm 6.9758 (8.6179/3.1914) mem 68106MB [2022-12-20 10:40:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][90/1519] eta 0:24:03 lr 0.000008 time 0.9376 (1.0098) model_time 0.9375 (1.0051) loss 1.0478 (0.8211) grad_norm 7.4308 (8.7367/3.2191) mem 68106MB [2022-12-20 10:40:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][100/1519] eta 0:23:51 lr 0.000008 time 0.9398 (1.0091) model_time 0.9396 (1.0049) loss 0.8365 (0.8144) grad_norm 8.2731 (8.7372/3.1078) mem 68106MB [2022-12-20 10:40:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][110/1519] eta 0:23:40 lr 0.000008 time 0.9247 (1.0084) model_time 0.9246 (1.0045) loss 0.8286 (0.8219) grad_norm 9.4746 (8.6976/2.9792) mem 68106MB [2022-12-20 10:40:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][120/1519] eta 0:23:31 lr 0.000008 time 0.9249 (1.0089) model_time 0.9248 (1.0053) loss 0.8907 (0.8159) grad_norm 6.2123 (8.6217/2.8820) mem 68106MB [2022-12-20 10:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][130/1519] eta 0:23:22 lr 0.000008 time 0.9250 (1.0096) model_time 0.9249 (1.0063) loss 0.9182 (0.8142) grad_norm 8.9705 (8.5945/2.7964) mem 68106MB [2022-12-20 10:40:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][140/1519] eta 0:23:11 lr 0.000008 time 0.9210 (1.0088) model_time 0.9208 (1.0057) loss 0.8012 (0.8159) grad_norm 10.2853 (8.5946/2.7303) mem 68106MB [2022-12-20 10:41:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][150/1519] eta 0:23:00 lr 0.000008 time 1.0112 (1.0087) model_time 1.0110 (1.0057) loss 0.9857 (0.8150) grad_norm 10.8477 (8.6570/2.7385) mem 68106MB [2022-12-20 10:41:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][160/1519] eta 0:22:51 lr 0.000008 time 0.9243 (1.0091) model_time 0.9242 (1.0063) loss 0.6900 (0.8130) grad_norm 9.3203 (8.6243/2.6778) mem 68106MB [2022-12-20 10:41:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][170/1519] eta 0:22:40 lr 0.000008 time 0.9282 (1.0085) model_time 0.9280 (1.0058) loss 0.7106 (0.8127) grad_norm 7.5888 (8.5503/2.6420) mem 68106MB [2022-12-20 10:41:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][180/1519] eta 0:22:29 lr 0.000008 time 0.9189 (1.0081) model_time 0.9188 (1.0056) loss 1.1652 (0.8145) grad_norm 10.2784 (8.5993/2.5847) mem 68106MB [2022-12-20 10:41:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][190/1519] eta 0:22:21 lr 0.000008 time 0.9899 (1.0095) model_time 0.9897 (1.0071) loss 0.7885 (0.8151) grad_norm 5.6563 (8.5571/2.5478) mem 68106MB [2022-12-20 10:41:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][200/1519] eta 0:22:10 lr 0.000008 time 0.9246 (1.0091) model_time 0.9245 (1.0068) loss 0.6992 (0.8149) grad_norm 8.9655 (8.5745/2.5073) mem 68106MB [2022-12-20 10:42:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][210/1519] eta 0:22:00 lr 0.000008 time 0.9243 (1.0086) model_time 0.9241 (1.0064) loss 0.6830 (0.8122) grad_norm 5.9811 (8.5095/2.4991) mem 68106MB [2022-12-20 10:42:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][220/1519] eta 0:21:49 lr 0.000008 time 0.9196 (1.0082) model_time 0.9195 (1.0061) loss 0.7628 (0.8129) grad_norm 6.7055 (8.4407/2.4709) mem 68106MB [2022-12-20 10:42:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][230/1519] eta 0:21:39 lr 0.000008 time 0.9792 (1.0080) model_time 0.9790 (1.0060) loss 0.7069 (0.8097) grad_norm 8.6587 (8.3841/2.4396) mem 68106MB [2022-12-20 10:42:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][240/1519] eta 0:21:30 lr 0.000008 time 1.1986 (1.0090) model_time 1.1985 (1.0071) loss 1.1368 (0.8111) grad_norm 6.9347 (8.3909/2.4058) mem 68106MB [2022-12-20 10:42:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][250/1519] eta 0:21:20 lr 0.000008 time 0.9264 (1.0088) model_time 0.9263 (1.0068) loss 0.7242 (0.8082) grad_norm 11.7236 (8.3623/2.4005) mem 68106MB [2022-12-20 10:42:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][260/1519] eta 0:21:09 lr 0.000008 time 0.9317 (1.0086) model_time 0.9316 (1.0067) loss 1.0032 (0.8090) grad_norm 7.8637 (8.2920/2.3866) mem 68106MB [2022-12-20 10:43:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][270/1519] eta 0:20:59 lr 0.000008 time 0.9188 (1.0081) model_time 0.9186 (1.0064) loss 0.8642 (0.8089) grad_norm 8.8399 (8.3280/2.3705) mem 68106MB [2022-12-20 10:43:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][280/1519] eta 0:20:49 lr 0.000008 time 0.9233 (1.0087) model_time 0.9232 (1.0070) loss 0.7408 (0.8085) grad_norm 8.7342 (8.3418/2.3605) mem 68106MB [2022-12-20 10:43:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][290/1519] eta 0:20:39 lr 0.000008 time 0.9220 (1.0084) model_time 0.9219 (1.0067) loss 0.9254 (0.8112) grad_norm 11.2369 (8.3510/2.3371) mem 68106MB [2022-12-20 10:43:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][300/1519] eta 0:20:29 lr 0.000008 time 0.9261 (1.0083) model_time 0.9260 (1.0067) loss 0.7394 (0.8112) grad_norm 7.8404 (8.3239/2.3029) mem 68106MB [2022-12-20 10:43:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][310/1519] eta 0:20:18 lr 0.000008 time 0.9249 (1.0081) model_time 0.9247 (1.0065) loss 0.6989 (0.8087) grad_norm 7.1395 (8.3047/2.2787) mem 68106MB [2022-12-20 10:43:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][320/1519] eta 0:20:08 lr 0.000008 time 0.9383 (1.0079) model_time 0.9381 (1.0063) loss 0.7061 (0.8082) grad_norm 8.9290 (8.2952/2.2486) mem 68106MB [2022-12-20 10:44:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][330/1519] eta 0:19:58 lr 0.000008 time 0.9225 (1.0076) model_time 0.9224 (1.0061) loss 0.8283 (0.8083) grad_norm 8.5588 (8.2904/2.2208) mem 68106MB [2022-12-20 10:44:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][340/1519] eta 0:19:48 lr 0.000008 time 0.9294 (1.0079) model_time 0.9291 (1.0064) loss 1.0898 (0.8091) grad_norm 12.9790 (8.3617/2.3553) mem 68106MB [2022-12-20 10:44:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][350/1519] eta 0:19:38 lr 0.000008 time 0.9327 (1.0078) model_time 0.9325 (1.0064) loss 1.1905 (0.8090) grad_norm 7.4784 (8.3676/2.3239) mem 68106MB [2022-12-20 10:44:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][360/1519] eta 0:19:27 lr 0.000008 time 0.9222 (1.0076) model_time 0.9221 (1.0062) loss 0.9012 (0.8096) grad_norm 9.6218 (8.3796/2.3119) mem 68106MB [2022-12-20 10:44:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][370/1519] eta 0:19:18 lr 0.000008 time 0.9245 (1.0079) model_time 0.9243 (1.0065) loss 0.6520 (0.8088) grad_norm 6.6895 (8.3492/2.2942) mem 68106MB [2022-12-20 10:44:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][380/1519] eta 0:19:07 lr 0.000008 time 0.9193 (1.0077) model_time 0.9191 (1.0063) loss 0.8637 (0.8097) grad_norm 8.4767 (8.3484/2.2680) mem 68106MB [2022-12-20 10:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][390/1519] eta 0:18:57 lr 0.000008 time 0.9330 (1.0075) model_time 0.9329 (1.0061) loss 0.8272 (0.8092) grad_norm 7.2996 (8.3336/2.2458) mem 68106MB [2022-12-20 10:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][400/1519] eta 0:18:47 lr 0.000008 time 0.9234 (1.0072) model_time 0.9232 (1.0059) loss 0.9795 (0.8096) grad_norm 6.9935 (8.2963/2.2385) mem 68106MB [2022-12-20 10:45:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][410/1519] eta 0:18:36 lr 0.000008 time 0.9806 (1.0072) model_time 0.9805 (1.0059) loss 0.9092 (0.8114) grad_norm 10.7177 (8.3116/2.2357) mem 68106MB [2022-12-20 10:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][420/1519] eta 0:18:26 lr 0.000008 time 1.0082 (1.0072) model_time 1.0080 (1.0060) loss 0.8039 (0.8113) grad_norm 6.9182 (8.2943/2.2250) mem 68106MB [2022-12-20 10:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][430/1519] eta 0:18:16 lr 0.000008 time 0.9229 (1.0071) model_time 0.9228 (1.0059) loss 0.7195 (0.8111) grad_norm 6.1936 (8.2914/2.2111) mem 68106MB [2022-12-20 10:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][440/1519] eta 0:18:06 lr 0.000008 time 0.9807 (1.0071) model_time 0.9805 (1.0059) loss 0.8644 (0.8122) grad_norm 12.0418 (8.3358/2.2349) mem 68106MB [2022-12-20 10:46:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][450/1519] eta 0:17:56 lr 0.000008 time 0.9251 (1.0070) model_time 0.9249 (1.0058) loss 0.7980 (0.8112) grad_norm 6.4748 (8.3063/2.2201) mem 68106MB [2022-12-20 10:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][460/1519] eta 0:17:46 lr 0.000008 time 0.9515 (1.0070) model_time 0.9514 (1.0059) loss 0.7025 (0.8110) grad_norm 8.8795 (8.3097/2.2015) mem 68106MB [2022-12-20 10:46:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][470/1519] eta 0:17:36 lr 0.000008 time 0.9680 (1.0071) model_time 0.9678 (1.0060) loss 0.6725 (0.8109) grad_norm 6.7450 (8.2866/2.1880) mem 68106MB [2022-12-20 10:46:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][480/1519] eta 0:17:26 lr 0.000008 time 0.8940 (1.0070) model_time 0.8938 (1.0058) loss 0.8964 (0.8113) grad_norm 7.4690 (8.2840/2.1961) mem 68106MB [2022-12-20 10:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][490/1519] eta 0:17:16 lr 0.000008 time 0.9320 (1.0069) model_time 0.9319 (1.0058) loss 1.0778 (0.8122) grad_norm 20.6383 (8.3553/2.3461) mem 68106MB [2022-12-20 10:46:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][500/1519] eta 0:17:06 lr 0.000008 time 0.9233 (1.0069) model_time 0.9231 (1.0058) loss 0.8167 (0.8121) grad_norm 8.4778 (8.3860/2.3428) mem 68106MB [2022-12-20 10:47:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][510/1519] eta 0:16:56 lr 0.000008 time 0.9253 (1.0069) model_time 0.9251 (1.0059) loss 0.7263 (0.8125) grad_norm 6.6574 (8.3652/2.3270) mem 68106MB [2022-12-20 10:47:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][520/1519] eta 0:16:45 lr 0.000008 time 0.9237 (1.0068) model_time 0.9236 (1.0057) loss 0.6655 (0.8119) grad_norm 5.9698 (8.3532/2.3138) mem 68106MB [2022-12-20 10:47:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][530/1519] eta 0:16:35 lr 0.000008 time 0.9267 (1.0067) model_time 0.9265 (1.0057) loss 0.6784 (0.8113) grad_norm 10.0343 (8.3841/2.3261) mem 68106MB [2022-12-20 10:47:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][540/1519] eta 0:16:25 lr 0.000008 time 0.9202 (1.0066) model_time 0.9201 (1.0055) loss 1.0667 (0.8113) grad_norm 6.4652 (8.3585/2.3137) mem 68106MB [2022-12-20 10:47:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][550/1519] eta 0:16:15 lr 0.000007 time 0.9257 (1.0067) model_time 0.9255 (1.0057) loss 0.7583 (0.8101) grad_norm 9.8061 (8.3490/2.2993) mem 68106MB [2022-12-20 10:47:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][560/1519] eta 0:16:05 lr 0.000007 time 0.9332 (1.0068) model_time 0.9331 (1.0058) loss 0.8796 (0.8092) grad_norm 7.2409 (8.3791/2.3312) mem 68106MB [2022-12-20 10:48:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][570/1519] eta 0:15:55 lr 0.000007 time 1.0751 (1.0070) model_time 1.0750 (1.0060) loss 1.0280 (0.8087) grad_norm 7.6503 (8.3717/2.3167) mem 68106MB [2022-12-20 10:48:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][580/1519] eta 0:15:45 lr 0.000007 time 0.9270 (1.0068) model_time 0.9269 (1.0058) loss 0.7684 (0.8089) grad_norm 6.2955 (8.3649/2.3058) mem 68106MB [2022-12-20 10:48:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][590/1519] eta 0:15:35 lr 0.000007 time 0.9337 (1.0067) model_time 0.9335 (1.0057) loss 0.8869 (0.8092) grad_norm 9.4697 (8.3733/2.2984) mem 68106MB [2022-12-20 10:48:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][600/1519] eta 0:15:25 lr 0.000007 time 0.9310 (1.0066) model_time 0.9308 (1.0056) loss 0.7841 (0.8093) grad_norm 11.2476 (8.3734/2.2881) mem 68106MB [2022-12-20 10:48:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][610/1519] eta 0:15:14 lr 0.000007 time 0.9229 (1.0066) model_time 0.9227 (1.0056) loss 0.6536 (0.8092) grad_norm 12.2355 (8.4023/2.3038) mem 68106MB [2022-12-20 10:48:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][620/1519] eta 0:15:04 lr 0.000007 time 0.9795 (1.0065) model_time 0.9793 (1.0056) loss 1.0001 (0.8091) grad_norm 8.8915 (8.3847/2.2851) mem 68106MB [2022-12-20 10:49:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][630/1519] eta 0:14:54 lr 0.000007 time 0.9351 (1.0064) model_time 0.9350 (1.0055) loss 0.8778 (0.8080) grad_norm 8.8163 (8.3801/2.2776) mem 68106MB [2022-12-20 10:49:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][640/1519] eta 0:14:44 lr 0.000007 time 0.9233 (1.0063) model_time 0.9232 (1.0054) loss 1.0504 (0.8090) grad_norm 15.7171 (8.4207/2.3062) mem 68106MB [2022-12-20 10:49:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][650/1519] eta 0:14:34 lr 0.000007 time 0.9248 (1.0067) model_time 0.9247 (1.0058) loss 0.7774 (0.8093) grad_norm 10.5309 (8.3697/2.0947) mem 68106MB [2022-12-20 10:49:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][660/1519] eta 0:14:24 lr 0.000007 time 0.9234 (1.0066) model_time 0.9233 (1.0057) loss 0.7194 (0.8087) grad_norm 6.3586 (8.3769/2.0938) mem 68106MB [2022-12-20 10:49:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][670/1519] eta 0:14:14 lr 0.000007 time 0.9276 (1.0066) model_time 0.9274 (1.0057) loss 1.0795 (0.8092) grad_norm 13.4913 (8.4068/2.1141) mem 68106MB [2022-12-20 10:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][680/1519] eta 0:14:04 lr 0.000007 time 0.9253 (1.0069) model_time 0.9252 (1.0060) loss 0.7531 (0.8085) grad_norm 9.4997 (8.4152/2.1055) mem 68106MB [2022-12-20 10:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][690/1519] eta 0:13:54 lr 0.000007 time 0.9230 (1.0070) model_time 0.9229 (1.0061) loss 0.6980 (0.8089) grad_norm 8.8991 (8.3879/2.0609) mem 68106MB [2022-12-20 10:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][700/1519] eta 0:13:44 lr 0.000007 time 0.9326 (1.0069) model_time 0.9325 (1.0060) loss 0.7641 (0.8095) grad_norm 8.1708 (8.3852/2.0555) mem 68106MB [2022-12-20 10:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][710/1519] eta 0:13:34 lr 0.000007 time 0.9242 (1.0071) model_time 0.9241 (1.0062) loss 0.9196 (0.8095) grad_norm 9.0253 (8.3895/2.0572) mem 68106MB [2022-12-20 10:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][720/1519] eta 0:13:24 lr 0.000007 time 0.9301 (1.0070) model_time 0.9299 (1.0061) loss 0.7646 (0.8099) grad_norm 10.3235 (8.4239/2.0741) mem 68106MB [2022-12-20 10:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][730/1519] eta 0:13:14 lr 0.000007 time 0.9301 (1.0070) model_time 0.9300 (1.0062) loss 0.9710 (0.8098) grad_norm 7.5897 (8.4035/2.0778) mem 68106MB [2022-12-20 10:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][740/1519] eta 0:13:04 lr 0.000007 time 1.0582 (1.0073) model_time 1.0581 (1.0064) loss 0.8065 (0.8092) grad_norm 7.0202 (8.3958/2.0759) mem 68106MB [2022-12-20 10:51:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][750/1519] eta 0:12:54 lr 0.000007 time 0.9788 (1.0072) model_time 0.9786 (1.0064) loss 0.7753 (0.8090) grad_norm 9.6295 (8.3997/2.0823) mem 68106MB [2022-12-20 10:51:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][760/1519] eta 0:12:44 lr 0.000007 time 0.9784 (1.0072) model_time 0.9783 (1.0063) loss 0.8838 (0.8088) grad_norm 7.5378 (8.3901/2.0799) mem 68106MB [2022-12-20 10:51:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][770/1519] eta 0:12:34 lr 0.000007 time 0.9391 (1.0072) model_time 0.9390 (1.0063) loss 0.6913 (0.8086) grad_norm 6.5982 (8.4012/2.0742) mem 68106MB [2022-12-20 10:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][780/1519] eta 0:12:24 lr 0.000007 time 0.9236 (1.0070) model_time 0.9234 (1.0062) loss 0.8550 (0.8097) grad_norm 8.4520 (8.3838/2.0699) mem 68106MB [2022-12-20 10:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][790/1519] eta 0:12:14 lr 0.000007 time 0.9274 (1.0071) model_time 0.9273 (1.0063) loss 0.8263 (0.8093) grad_norm 7.6326 (8.3778/2.0625) mem 68106MB [2022-12-20 10:52:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][800/1519] eta 0:12:03 lr 0.000007 time 0.9213 (1.0070) model_time 0.9212 (1.0061) loss 0.7469 (0.8091) grad_norm 6.9565 (8.3872/2.1137) mem 68106MB [2022-12-20 10:52:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][810/1519] eta 0:11:53 lr 0.000007 time 0.9208 (1.0069) model_time 0.9207 (1.0061) loss 0.7167 (0.8094) grad_norm 8.6682 (8.4098/2.0955) mem 68106MB [2022-12-20 10:52:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][820/1519] eta 0:11:43 lr 0.000007 time 0.9274 (1.0068) model_time 0.9273 (1.0060) loss 0.7412 (0.8089) grad_norm 11.6652 (8.4243/2.1039) mem 68106MB [2022-12-20 10:52:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][830/1519] eta 0:11:33 lr 0.000007 time 0.9229 (1.0067) model_time 0.9228 (1.0059) loss 1.1368 (0.8089) grad_norm 7.9927 (8.4366/2.0955) mem 68106MB [2022-12-20 10:52:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][840/1519] eta 0:11:23 lr 0.000007 time 0.9256 (1.0066) model_time 0.9255 (1.0058) loss 0.9509 (0.8089) grad_norm 15.4631 (8.4752/2.1359) mem 68106MB [2022-12-20 10:52:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][850/1519] eta 0:11:13 lr 0.000007 time 0.9573 (1.0064) model_time 0.9572 (1.0056) loss 0.6555 (0.8085) grad_norm 6.9264 (8.4938/2.1331) mem 68106MB [2022-12-20 10:53:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][860/1519] eta 0:11:03 lr 0.000007 time 0.9095 (1.0064) model_time 0.9094 (1.0056) loss 0.8222 (0.8078) grad_norm 7.0880 (8.5216/2.1168) mem 68106MB [2022-12-20 10:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][870/1519] eta 0:10:53 lr 0.000007 time 0.9176 (1.0063) model_time 0.9175 (1.0056) loss 0.6755 (0.8087) grad_norm 6.2478 (8.5279/2.1281) mem 68106MB [2022-12-20 10:53:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][880/1519] eta 0:10:43 lr 0.000007 time 0.9332 (1.0063) model_time 0.9331 (1.0055) loss 0.8193 (0.8086) grad_norm 9.1101 (8.5360/2.1201) mem 68106MB [2022-12-20 10:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][890/1519] eta 0:10:32 lr 0.000007 time 0.9220 (1.0062) model_time 0.9219 (1.0054) loss 0.8357 (0.8083) grad_norm 10.5834 (8.5391/2.1257) mem 68106MB [2022-12-20 10:53:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][900/1519] eta 0:10:22 lr 0.000007 time 0.9246 (1.0061) model_time 0.9245 (1.0053) loss 0.7123 (0.8080) grad_norm 9.9395 (8.5568/2.1342) mem 68106MB [2022-12-20 10:53:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][910/1519] eta 0:10:12 lr 0.000007 time 0.9292 (1.0060) model_time 0.9291 (1.0053) loss 0.6854 (0.8082) grad_norm 8.1428 (8.5699/2.1279) mem 68106MB [2022-12-20 10:54:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][920/1519] eta 0:10:02 lr 0.000007 time 1.0206 (1.0061) model_time 1.0203 (1.0053) loss 0.8904 (0.8082) grad_norm 11.0071 (8.5905/2.1346) mem 68106MB [2022-12-20 10:54:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][930/1519] eta 0:09:52 lr 0.000007 time 0.9213 (1.0060) model_time 0.9212 (1.0052) loss 0.7581 (0.8082) grad_norm 10.5205 (8.6076/2.1437) mem 68106MB [2022-12-20 10:54:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][940/1519] eta 0:09:42 lr 0.000007 time 0.9251 (1.0059) model_time 0.9249 (1.0052) loss 0.6636 (0.8078) grad_norm 8.5410 (8.5779/2.0691) mem 68106MB [2022-12-20 10:54:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][950/1519] eta 0:09:32 lr 0.000007 time 0.9231 (1.0059) model_time 0.9229 (1.0051) loss 0.8904 (0.8077) grad_norm 6.3678 (8.5616/2.0658) mem 68106MB [2022-12-20 10:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][960/1519] eta 0:09:22 lr 0.000007 time 0.8884 (1.0061) model_time 0.8882 (1.0054) loss 0.7840 (0.8070) grad_norm 10.3466 (8.5848/2.0593) mem 68106MB [2022-12-20 10:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][970/1519] eta 0:09:12 lr 0.000007 time 0.9216 (1.0063) model_time 0.9215 (1.0056) loss 0.8080 (0.8072) grad_norm 9.4566 (8.5868/2.0578) mem 68106MB [2022-12-20 10:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][980/1519] eta 0:09:02 lr 0.000007 time 0.9287 (1.0062) model_time 0.9285 (1.0055) loss 0.8698 (0.8067) grad_norm 11.6252 (8.6376/2.0878) mem 68106MB [2022-12-20 10:55:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][990/1519] eta 0:08:52 lr 0.000007 time 0.9987 (1.0063) model_time 0.9986 (1.0056) loss 0.6763 (0.8066) grad_norm 7.9203 (8.6410/2.0885) mem 68106MB [2022-12-20 10:55:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1000/1519] eta 0:08:42 lr 0.000007 time 0.9110 (1.0064) model_time 0.9109 (1.0057) loss 0.8664 (0.8067) grad_norm 6.2288 (8.6855/2.0933) mem 68106MB [2022-12-20 10:55:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1010/1519] eta 0:08:32 lr 0.000007 time 0.9270 (1.0063) model_time 0.9269 (1.0056) loss 0.7383 (0.8071) grad_norm 10.4072 (8.6841/2.0917) mem 68106MB [2022-12-20 10:55:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1020/1519] eta 0:08:22 lr 0.000007 time 0.9204 (1.0063) model_time 0.9203 (1.0056) loss 1.2019 (0.8074) grad_norm 8.3189 (8.6863/2.0797) mem 68106MB [2022-12-20 10:55:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1030/1519] eta 0:08:12 lr 0.000007 time 0.9333 (1.0063) model_time 0.9331 (1.0056) loss 0.6885 (0.8068) grad_norm 10.3484 (8.7222/2.1292) mem 68106MB [2022-12-20 10:56:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1040/1519] eta 0:08:02 lr 0.000007 time 0.9253 (1.0063) model_time 0.9252 (1.0056) loss 0.6611 (0.8070) grad_norm 7.3643 (8.7010/2.1087) mem 68106MB [2022-12-20 10:56:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1050/1519] eta 0:07:51 lr 0.000007 time 0.8891 (1.0063) model_time 0.8890 (1.0056) loss 0.9194 (0.8070) grad_norm 7.6251 (8.7015/2.0939) mem 68106MB [2022-12-20 10:56:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1060/1519] eta 0:07:41 lr 0.000007 time 0.9245 (1.0062) model_time 0.9244 (1.0056) loss 0.9318 (0.8072) grad_norm 9.7797 (8.7283/2.1474) mem 68106MB [2022-12-20 10:56:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1070/1519] eta 0:07:31 lr 0.000007 time 0.9339 (1.0063) model_time 0.9338 (1.0056) loss 0.8456 (0.8072) grad_norm 9.7979 (8.7738/2.1760) mem 68106MB [2022-12-20 10:56:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1080/1519] eta 0:07:21 lr 0.000007 time 1.1710 (1.0065) model_time 1.1708 (1.0058) loss 0.8271 (0.8071) grad_norm 5.9791 (8.7784/2.1572) mem 68106MB [2022-12-20 10:56:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1090/1519] eta 0:07:11 lr 0.000007 time 0.9288 (1.0064) model_time 0.9286 (1.0057) loss 0.7190 (0.8076) grad_norm 7.6427 (8.7551/2.1367) mem 68106MB [2022-12-20 10:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1100/1519] eta 0:07:01 lr 0.000007 time 0.9749 (1.0064) model_time 0.9747 (1.0057) loss 0.7240 (0.8072) grad_norm 9.9905 (8.7199/2.0300) mem 68106MB [2022-12-20 10:57:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1110/1519] eta 0:06:51 lr 0.000007 time 0.9242 (1.0064) model_time 0.9241 (1.0057) loss 0.7247 (0.8073) grad_norm 6.1187 (8.7150/2.0322) mem 68106MB [2022-12-20 10:57:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1120/1519] eta 0:06:41 lr 0.000007 time 0.9193 (1.0063) model_time 0.9192 (1.0057) loss 0.7295 (0.8076) grad_norm 7.8765 (8.7371/2.0373) mem 68106MB [2022-12-20 10:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1130/1519] eta 0:06:31 lr 0.000007 time 0.9241 (1.0063) model_time 0.9239 (1.0056) loss 0.9087 (0.8082) grad_norm 8.3676 (8.7514/2.0227) mem 68106MB [2022-12-20 10:57:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1140/1519] eta 0:06:21 lr 0.000007 time 1.0004 (1.0064) model_time 1.0002 (1.0058) loss 0.6715 (0.8077) grad_norm 6.5998 (8.7586/2.0180) mem 68106MB [2022-12-20 10:57:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1150/1519] eta 0:06:11 lr 0.000007 time 0.9227 (1.0064) model_time 0.9226 (1.0057) loss 1.0844 (0.8080) grad_norm 6.9723 (8.7580/2.0174) mem 68106MB [2022-12-20 10:58:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1160/1519] eta 0:06:01 lr 0.000007 time 0.9228 (1.0063) model_time 0.9227 (1.0057) loss 0.6884 (0.8084) grad_norm 9.8611 (8.7373/1.9754) mem 68106MB [2022-12-20 10:58:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1170/1519] eta 0:05:51 lr 0.000007 time 0.9999 (1.0063) model_time 0.9997 (1.0057) loss 0.6794 (0.8079) grad_norm 12.9254 (8.7500/1.9899) mem 68106MB [2022-12-20 10:58:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1180/1519] eta 0:05:41 lr 0.000007 time 0.9173 (1.0063) model_time 0.9172 (1.0057) loss 0.7165 (0.8083) grad_norm 6.9506 (8.7552/1.9862) mem 68106MB [2022-12-20 10:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1190/1519] eta 0:05:31 lr 0.000007 time 0.9256 (1.0063) model_time 0.9254 (1.0056) loss 0.6953 (0.8081) grad_norm 12.3696 (8.7988/2.0316) mem 68106MB [2022-12-20 10:58:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1200/1519] eta 0:05:20 lr 0.000007 time 0.9213 (1.0062) model_time 0.9211 (1.0056) loss 0.6903 (0.8080) grad_norm 22.7048 (8.8407/2.1844) mem 68106MB [2022-12-20 10:58:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1210/1519] eta 0:05:10 lr 0.000007 time 0.9227 (1.0061) model_time 0.9225 (1.0055) loss 0.9500 (0.8075) grad_norm 7.0023 (8.8145/2.1917) mem 68106MB [2022-12-20 10:59:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1220/1519] eta 0:05:00 lr 0.000007 time 0.9345 (1.0061) model_time 0.9344 (1.0054) loss 0.8693 (0.8074) grad_norm 8.0740 (8.7938/2.1817) mem 68106MB [2022-12-20 10:59:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1230/1519] eta 0:04:50 lr 0.000007 time 0.9259 (1.0060) model_time 0.9257 (1.0054) loss 1.1970 (0.8082) grad_norm 10.5898 (8.7917/2.1877) mem 68106MB [2022-12-20 10:59:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1240/1519] eta 0:04:40 lr 0.000007 time 0.9174 (1.0060) model_time 0.9173 (1.0054) loss 0.9158 (0.8081) grad_norm 10.3867 (8.8032/2.1965) mem 68106MB [2022-12-20 10:59:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1250/1519] eta 0:04:30 lr 0.000007 time 0.9233 (1.0060) model_time 0.9232 (1.0054) loss 0.7036 (0.8075) grad_norm 5.5421 (8.7515/2.1687) mem 68106MB [2022-12-20 10:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1260/1519] eta 0:04:20 lr 0.000007 time 0.9185 (1.0060) model_time 0.9183 (1.0054) loss 0.7226 (0.8079) grad_norm 10.6474 (8.7686/2.1665) mem 68106MB [2022-12-20 10:59:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1270/1519] eta 0:04:10 lr 0.000007 time 0.9281 (1.0060) model_time 0.9280 (1.0054) loss 1.0212 (0.8072) grad_norm 6.8401 (8.7665/2.1674) mem 68106MB [2022-12-20 11:00:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1280/1519] eta 0:04:00 lr 0.000007 time 0.9189 (1.0060) model_time 0.9188 (1.0054) loss 1.0152 (0.8071) grad_norm 22.8686 (8.8108/2.2974) mem 68106MB [2022-12-20 11:00:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1290/1519] eta 0:03:50 lr 0.000007 time 0.9261 (1.0059) model_time 0.9260 (1.0053) loss 0.7184 (0.8073) grad_norm 7.7260 (8.8139/2.3010) mem 68106MB [2022-12-20 11:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1300/1519] eta 0:03:40 lr 0.000007 time 0.9974 (1.0060) model_time 0.9973 (1.0053) loss 0.6628 (0.8075) grad_norm 7.6052 (8.7974/2.3038) mem 68106MB [2022-12-20 11:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1310/1519] eta 0:03:30 lr 0.000007 time 0.9288 (1.0060) model_time 0.9286 (1.0053) loss 0.7480 (0.8074) grad_norm 11.8237 (8.8072/2.3086) mem 68106MB [2022-12-20 11:00:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1320/1519] eta 0:03:20 lr 0.000007 time 0.9281 (1.0060) model_time 0.9279 (1.0053) loss 0.6678 (0.8073) grad_norm 6.9353 (8.7981/2.3032) mem 68106MB [2022-12-20 11:00:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1330/1519] eta 0:03:10 lr 0.000007 time 0.9207 (1.0059) model_time 0.9206 (1.0053) loss 0.7699 (0.8076) grad_norm 8.2607 (8.7998/2.2948) mem 68106MB [2022-12-20 11:01:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1340/1519] eta 0:03:00 lr 0.000007 time 0.9222 (1.0060) model_time 0.9221 (1.0054) loss 0.6630 (0.8074) grad_norm 10.7915 (8.7976/2.2959) mem 68106MB [2022-12-20 11:01:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1350/1519] eta 0:02:50 lr 0.000007 time 0.9924 (1.0060) model_time 0.9923 (1.0054) loss 1.0301 (0.8072) grad_norm 7.7889 (8.7712/2.2681) mem 68106MB [2022-12-20 11:01:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1360/1519] eta 0:02:39 lr 0.000007 time 0.9287 (1.0060) model_time 0.9286 (1.0054) loss 0.6914 (0.8066) grad_norm 6.3804 (8.7663/2.2715) mem 68106MB [2022-12-20 11:01:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1370/1519] eta 0:02:29 lr 0.000007 time 0.9238 (1.0060) model_time 0.9236 (1.0054) loss 0.8466 (0.8069) grad_norm 9.6683 (8.7683/2.2681) mem 68106MB [2022-12-20 11:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1380/1519] eta 0:02:19 lr 0.000007 time 0.9285 (1.0060) model_time 0.9283 (1.0054) loss 0.8326 (0.8068) grad_norm 6.9472 (8.7810/2.2797) mem 68106MB [2022-12-20 11:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1390/1519] eta 0:02:09 lr 0.000007 time 0.9236 (1.0059) model_time 0.9235 (1.0053) loss 0.7375 (0.8064) grad_norm 6.6748 (8.7988/2.2928) mem 68106MB [2022-12-20 11:02:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1400/1519] eta 0:01:59 lr 0.000007 time 0.9256 (1.0059) model_time 0.9255 (1.0054) loss 0.8771 (0.8061) grad_norm 8.9943 (8.7677/2.2478) mem 68106MB [2022-12-20 11:02:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1410/1519] eta 0:01:49 lr 0.000007 time 0.9220 (1.0059) model_time 0.9219 (1.0053) loss 0.8632 (0.8061) grad_norm 8.8445 (8.7822/2.2515) mem 68106MB [2022-12-20 11:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1420/1519] eta 0:01:39 lr 0.000007 time 0.9278 (1.0059) model_time 0.9276 (1.0053) loss 0.6803 (0.8062) grad_norm 7.7903 (8.8072/2.2456) mem 68106MB [2022-12-20 11:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1430/1519] eta 0:01:29 lr 0.000007 time 0.9287 (1.0058) model_time 0.9285 (1.0052) loss 0.7122 (0.8065) grad_norm 9.5743 (8.8046/2.2387) mem 68106MB [2022-12-20 11:02:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1440/1519] eta 0:01:19 lr 0.000007 time 0.9285 (1.0058) model_time 0.9283 (1.0052) loss 0.6859 (0.8059) grad_norm 9.0848 (8.7849/2.2365) mem 68106MB [2022-12-20 11:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1450/1519] eta 0:01:09 lr 0.000007 time 0.9202 (1.0058) model_time 0.9200 (1.0052) loss 0.8146 (0.8061) grad_norm 10.9496 (8.7883/2.2060) mem 68106MB [2022-12-20 11:03:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1460/1519] eta 0:00:59 lr 0.000007 time 0.9175 (1.0058) model_time 0.9174 (1.0052) loss 0.8378 (0.8066) grad_norm 10.3578 (8.8003/2.2142) mem 68106MB [2022-12-20 11:03:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1470/1519] eta 0:00:49 lr 0.000007 time 0.9298 (1.0058) model_time 0.9297 (1.0052) loss 0.6962 (0.8067) grad_norm 10.2607 (8.8217/2.2321) mem 68106MB [2022-12-20 11:03:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1480/1519] eta 0:00:39 lr 0.000007 time 0.9916 (1.0058) model_time 0.9914 (1.0052) loss 0.9090 (0.8068) grad_norm 8.5219 (8.8015/2.2307) mem 68106MB [2022-12-20 11:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1490/1519] eta 0:00:29 lr 0.000007 time 0.9338 (1.0059) model_time 0.9337 (1.0053) loss 0.6705 (0.8065) grad_norm 10.0115 (8.8002/2.2273) mem 68106MB [2022-12-20 11:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1500/1519] eta 0:00:19 lr 0.000007 time 0.9404 (1.0058) model_time 0.9403 (1.0052) loss 0.8596 (0.8063) grad_norm 7.6253 (8.7638/2.2244) mem 68106MB [2022-12-20 11:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [69/100][1510/1519] eta 0:00:09 lr 0.000007 time 0.9233 (1.0058) model_time 0.9232 (1.0052) loss 0.7157 (0.8068) grad_norm 7.5395 (8.7809/2.3122) mem 68106MB [2022-12-20 11:04:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 69 training takes 0:25:27 [2022-12-20 11:04:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_69.pth saving...... [2022-12-20 11:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_69.pth saved !!! [2022-12-20 11:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.621 (0.621) Loss 0.5139 (0.5139) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 11:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.328) Loss 0.5297 (0.4931) Acc@1 92.014 (92.866) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 11:04:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.306 (0.314) Loss 0.4779 (0.4909) Acc@1 92.708 (92.907) Acc@5 98.958 (98.429) Mem 68106MB [2022-12-20 11:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.6206 (0.4980) Acc@1 88.542 (92.451) Acc@5 97.222 (98.342) Mem 68106MB [2022-12-20 11:04:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.308) Loss 0.4487 (0.4893) Acc@1 94.097 (92.480) Acc@5 99.306 (98.450) Mem 68106MB [2022-12-20 11:04:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.306) Loss 0.4766 (0.4867) Acc@1 92.014 (92.484) Acc@5 99.653 (98.482) Mem 68106MB [2022-12-20 11:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.303 (0.305) Loss 0.5031 (0.4869) Acc@1 90.625 (92.401) Acc@5 98.264 (98.463) Mem 68106MB [2022-12-20 11:04:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.305) Loss 0.5408 (0.4883) Acc@1 93.056 (92.356) Acc@5 97.917 (98.455) Mem 68106MB [2022-12-20 11:04:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.299 (0.304) Loss 0.4394 (0.4869) Acc@1 92.708 (92.404) Acc@5 98.958 (98.504) Mem 68106MB [2022-12-20 11:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:69] * Acc@1 92.358 Acc@5 98.510 [2022-12-20 11:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 11:04:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.46% [2022-12-20 11:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][0/1519] eta 0:47:16 lr 0.000007 time 1.8674 (1.8674) model_time 1.1029 (1.1029) loss 0.9596 (0.9596) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 11:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][10/1519] eta 0:27:06 lr 0.000007 time 0.9254 (1.0781) model_time 0.9253 (1.0083) loss 0.6721 (0.8059) grad_norm 7.2722 (8.0571/1.5456) mem 68106MB [2022-12-20 11:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][20/1519] eta 0:26:06 lr 0.000007 time 0.9829 (1.0452) model_time 0.9828 (1.0085) loss 0.6807 (0.8015) grad_norm 8.6008 (8.3721/1.7786) mem 68106MB [2022-12-20 11:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][30/1519] eta 0:25:35 lr 0.000007 time 0.9225 (1.0314) model_time 0.9223 (1.0064) loss 0.6538 (0.8272) grad_norm 8.8625 (8.3509/1.6323) mem 68106MB [2022-12-20 11:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][40/1519] eta 0:25:21 lr 0.000007 time 0.9274 (1.0290) model_time 0.9273 (1.0100) loss 0.7698 (0.8354) grad_norm 9.0849 (8.9391/2.0575) mem 68106MB [2022-12-20 11:05:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][50/1519] eta 0:25:05 lr 0.000007 time 0.9763 (1.0248) model_time 0.9761 (1.0095) loss 0.8779 (0.8319) grad_norm 7.5639 (8.7793/2.2188) mem 68106MB [2022-12-20 11:05:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][60/1519] eta 0:24:52 lr 0.000007 time 0.9774 (1.0232) model_time 0.9773 (1.0103) loss 0.7499 (0.8345) grad_norm 8.8297 (8.7719/2.1278) mem 68106MB [2022-12-20 11:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][70/1519] eta 0:24:42 lr 0.000007 time 0.9336 (1.0231) model_time 0.9335 (1.0121) loss 0.8038 (0.8410) grad_norm 9.3397 (8.9060/2.1443) mem 68106MB [2022-12-20 11:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][80/1519] eta 0:24:29 lr 0.000007 time 0.9263 (1.0213) model_time 0.9262 (1.0115) loss 0.8353 (0.8370) grad_norm 8.4186 (8.8379/2.1674) mem 68106MB [2022-12-20 11:06:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][90/1519] eta 0:24:18 lr 0.000007 time 0.9341 (1.0204) model_time 0.9340 (1.0117) loss 0.9633 (0.8357) grad_norm 8.7029 (8.8134/2.0864) mem 68106MB [2022-12-20 11:06:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][100/1519] eta 0:24:05 lr 0.000007 time 0.9201 (1.0189) model_time 0.9200 (1.0110) loss 0.9522 (0.8334) grad_norm 9.2079 (8.6474/2.0999) mem 68106MB [2022-12-20 11:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][110/1519] eta 0:23:53 lr 0.000007 time 0.9240 (1.0172) model_time 0.9239 (1.0100) loss 0.7972 (0.8306) grad_norm 6.4790 (8.6569/2.0508) mem 68106MB [2022-12-20 11:06:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][120/1519] eta 0:23:41 lr 0.000007 time 0.9205 (1.0162) model_time 0.9203 (1.0096) loss 1.1367 (0.8339) grad_norm 9.1293 (8.6058/2.0454) mem 68106MB [2022-12-20 11:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][130/1519] eta 0:23:29 lr 0.000007 time 0.9314 (1.0150) model_time 0.9313 (1.0088) loss 0.8068 (0.8294) grad_norm 8.2821 (8.4928/2.0175) mem 68106MB [2022-12-20 11:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][140/1519] eta 0:23:18 lr 0.000007 time 0.9253 (1.0138) model_time 0.9250 (1.0081) loss 0.6590 (0.8244) grad_norm 13.0347 (8.5769/2.0631) mem 68106MB [2022-12-20 11:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][150/1519] eta 0:23:06 lr 0.000007 time 0.9279 (1.0130) model_time 0.9278 (1.0077) loss 0.8015 (0.8170) grad_norm 6.4664 (8.5844/2.0983) mem 68106MB [2022-12-20 11:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][160/1519] eta 0:22:55 lr 0.000007 time 0.9306 (1.0125) model_time 0.9304 (1.0074) loss 0.9547 (0.8188) grad_norm 6.5689 (8.5138/2.0541) mem 68106MB [2022-12-20 11:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][170/1519] eta 0:22:44 lr 0.000007 time 0.9231 (1.0116) model_time 0.9230 (1.0068) loss 0.7429 (0.8184) grad_norm 9.3200 (8.5132/2.0007) mem 68106MB [2022-12-20 11:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][180/1519] eta 0:22:33 lr 0.000007 time 0.9286 (1.0108) model_time 0.9284 (1.0062) loss 0.6837 (0.8155) grad_norm 9.0658 (8.6104/2.0399) mem 68106MB [2022-12-20 11:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][190/1519] eta 0:22:22 lr 0.000007 time 0.9222 (1.0101) model_time 0.9221 (1.0058) loss 0.7008 (0.8172) grad_norm 8.9132 (8.5867/2.0135) mem 68106MB [2022-12-20 11:08:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][200/1519] eta 0:22:11 lr 0.000007 time 0.9223 (1.0095) model_time 0.9222 (1.0054) loss 1.1952 (0.8138) grad_norm 7.8807 (8.6212/2.0145) mem 68106MB [2022-12-20 11:08:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][210/1519] eta 0:22:01 lr 0.000007 time 0.9264 (1.0095) model_time 0.9263 (1.0056) loss 0.7056 (0.8137) grad_norm 8.4517 (8.6189/1.9688) mem 68106MB [2022-12-20 11:08:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][220/1519] eta 0:21:50 lr 0.000007 time 0.9293 (1.0090) model_time 0.9291 (1.0053) loss 0.7289 (0.8128) grad_norm 7.9026 (8.5935/1.9882) mem 68106MB [2022-12-20 11:08:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][230/1519] eta 0:21:40 lr 0.000007 time 0.9216 (1.0086) model_time 0.9214 (1.0049) loss 0.8545 (0.8133) grad_norm 6.0699 (8.7022/2.2299) mem 68106MB [2022-12-20 11:08:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][240/1519] eta 0:21:29 lr 0.000007 time 0.9164 (1.0081) model_time 0.9162 (1.0046) loss 0.7101 (0.8129) grad_norm 7.1423 (8.6801/2.1997) mem 68106MB [2022-12-20 11:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][250/1519] eta 0:21:19 lr 0.000007 time 0.9234 (1.0083) model_time 0.9232 (1.0049) loss 0.6611 (0.8121) grad_norm 6.3181 (8.6416/2.1684) mem 68106MB [2022-12-20 11:09:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][260/1519] eta 0:21:09 lr 0.000007 time 0.9203 (1.0085) model_time 0.9202 (1.0053) loss 0.6966 (0.8144) grad_norm 9.7138 (8.6490/2.1357) mem 68106MB [2022-12-20 11:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][270/1519] eta 0:20:59 lr 0.000007 time 0.9226 (1.0086) model_time 0.9225 (1.0054) loss 1.0409 (0.8143) grad_norm 9.7585 (8.6426/2.1005) mem 68106MB [2022-12-20 11:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][280/1519] eta 0:20:49 lr 0.000007 time 0.9209 (1.0081) model_time 0.9208 (1.0051) loss 0.6723 (0.8153) grad_norm 6.3113 (8.5970/2.0925) mem 68106MB [2022-12-20 11:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][290/1519] eta 0:20:39 lr 0.000007 time 0.9263 (1.0081) model_time 0.9261 (1.0052) loss 0.8213 (0.8163) grad_norm 8.5908 (8.6518/2.1968) mem 68106MB [2022-12-20 11:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][300/1519] eta 0:20:28 lr 0.000007 time 0.9313 (1.0079) model_time 0.9311 (1.0050) loss 0.9360 (0.8186) grad_norm 9.1960 (8.6312/2.1763) mem 68106MB [2022-12-20 11:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][310/1519] eta 0:20:18 lr 0.000007 time 0.9344 (1.0076) model_time 0.9343 (1.0048) loss 0.7017 (0.8185) grad_norm 7.2687 (8.6137/2.1490) mem 68106MB [2022-12-20 11:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][320/1519] eta 0:20:07 lr 0.000007 time 0.9240 (1.0072) model_time 0.9239 (1.0045) loss 1.0307 (0.8190) grad_norm 8.8706 (8.6311/2.1452) mem 68106MB [2022-12-20 11:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][330/1519] eta 0:19:57 lr 0.000007 time 0.9299 (1.0070) model_time 0.9297 (1.0044) loss 0.9350 (0.8205) grad_norm 6.5550 (8.6686/2.1741) mem 68106MB [2022-12-20 11:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][340/1519] eta 0:19:47 lr 0.000007 time 0.9269 (1.0071) model_time 0.9267 (1.0045) loss 0.9225 (0.8205) grad_norm 7.1815 (8.6466/2.1658) mem 68106MB [2022-12-20 11:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][350/1519] eta 0:19:37 lr 0.000007 time 0.9312 (1.0076) model_time 0.9310 (1.0051) loss 0.8737 (0.8219) grad_norm 7.7614 (8.6269/2.1605) mem 68106MB [2022-12-20 11:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][360/1519] eta 0:19:27 lr 0.000007 time 0.9322 (1.0073) model_time 0.9320 (1.0049) loss 0.9690 (0.8211) grad_norm 9.8821 (8.6230/2.1343) mem 68106MB [2022-12-20 11:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][370/1519] eta 0:19:17 lr 0.000007 time 0.9014 (1.0073) model_time 0.9012 (1.0050) loss 0.8306 (0.8206) grad_norm 6.9500 (8.5941/2.1140) mem 68106MB [2022-12-20 11:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][380/1519] eta 0:19:07 lr 0.000007 time 0.9326 (1.0075) model_time 0.9324 (1.0052) loss 1.0203 (0.8198) grad_norm 6.3480 (8.5870/2.0947) mem 68106MB [2022-12-20 11:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][390/1519] eta 0:18:57 lr 0.000007 time 0.9199 (1.0072) model_time 0.9198 (1.0050) loss 0.8619 (0.8197) grad_norm 10.8669 (8.5753/2.0895) mem 68106MB [2022-12-20 11:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][400/1519] eta 0:18:47 lr 0.000007 time 0.9265 (1.0073) model_time 0.9264 (1.0051) loss 0.8578 (0.8179) grad_norm 5.8193 (8.5687/2.1216) mem 68106MB [2022-12-20 11:11:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][410/1519] eta 0:18:37 lr 0.000007 time 0.8854 (1.0076) model_time 0.8853 (1.0054) loss 0.6603 (0.8162) grad_norm 5.5936 (8.5317/2.1344) mem 68106MB [2022-12-20 11:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][420/1519] eta 0:18:27 lr 0.000007 time 0.9243 (1.0074) model_time 0.9242 (1.0052) loss 0.6947 (0.8171) grad_norm 8.0069 (8.5202/2.1288) mem 68106MB [2022-12-20 11:12:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][430/1519] eta 0:18:17 lr 0.000007 time 0.9125 (1.0075) model_time 0.9124 (1.0054) loss 0.6765 (0.8155) grad_norm 5.4168 (8.5176/2.1382) mem 68106MB [2022-12-20 11:12:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][440/1519] eta 0:18:07 lr 0.000007 time 0.9212 (1.0074) model_time 0.9211 (1.0054) loss 0.6802 (0.8162) grad_norm 9.7449 (8.5130/2.1250) mem 68106MB [2022-12-20 11:12:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][450/1519] eta 0:17:56 lr 0.000007 time 0.9322 (1.0073) model_time 0.9320 (1.0052) loss 0.7035 (0.8167) grad_norm 7.3993 (8.5109/2.1041) mem 68106MB [2022-12-20 11:12:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][460/1519] eta 0:17:46 lr 0.000007 time 0.9246 (1.0071) model_time 0.9245 (1.0051) loss 0.8746 (0.8176) grad_norm 6.4974 (8.4860/2.0928) mem 68106MB [2022-12-20 11:12:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][470/1519] eta 0:17:36 lr 0.000007 time 0.9301 (1.0070) model_time 0.9300 (1.0051) loss 0.7416 (0.8166) grad_norm 7.0392 (8.4888/2.0844) mem 68106MB [2022-12-20 11:12:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][480/1519] eta 0:17:26 lr 0.000007 time 0.9380 (1.0068) model_time 0.9378 (1.0049) loss 0.8417 (0.8181) grad_norm 7.9980 (8.4954/2.0719) mem 68106MB [2022-12-20 11:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][490/1519] eta 0:17:15 lr 0.000007 time 0.9246 (1.0067) model_time 0.9244 (1.0048) loss 0.7429 (0.8178) grad_norm 5.7665 (8.4961/2.0682) mem 68106MB [2022-12-20 11:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][500/1519] eta 0:17:05 lr 0.000007 time 0.9243 (1.0065) model_time 0.9242 (1.0047) loss 0.7336 (0.8179) grad_norm 7.7328 (8.4866/2.0531) mem 68106MB [2022-12-20 11:13:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][510/1519] eta 0:16:55 lr 0.000007 time 0.9263 (1.0064) model_time 0.9262 (1.0046) loss 0.8625 (0.8171) grad_norm 8.1429 (8.5200/2.0770) mem 68106MB [2022-12-20 11:13:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][520/1519] eta 0:16:45 lr 0.000007 time 0.9091 (1.0068) model_time 0.9090 (1.0050) loss 0.6724 (0.8165) grad_norm 7.4746 (8.5395/2.0889) mem 68106MB [2022-12-20 11:13:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][530/1519] eta 0:16:35 lr 0.000007 time 0.9204 (1.0066) model_time 0.9202 (1.0049) loss 0.9514 (0.8167) grad_norm 9.9848 (8.5485/2.0763) mem 68106MB [2022-12-20 11:13:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][540/1519] eta 0:16:25 lr 0.000007 time 0.9255 (1.0066) model_time 0.9253 (1.0048) loss 0.7480 (0.8170) grad_norm 9.5981 (8.5707/2.1017) mem 68106MB [2022-12-20 11:14:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][550/1519] eta 0:16:15 lr 0.000007 time 0.9332 (1.0064) model_time 0.9331 (1.0047) loss 0.7263 (0.8175) grad_norm 6.9593 (8.5852/2.1067) mem 68106MB [2022-12-20 11:14:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][560/1519] eta 0:16:05 lr 0.000007 time 0.9270 (1.0063) model_time 0.9269 (1.0046) loss 0.9182 (0.8175) grad_norm 6.1612 (8.5730/2.1015) mem 68106MB [2022-12-20 11:14:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][570/1519] eta 0:15:55 lr 0.000007 time 0.9266 (1.0064) model_time 0.9264 (1.0048) loss 0.6612 (0.8174) grad_norm 6.7973 (8.5656/2.0865) mem 68106MB [2022-12-20 11:14:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][580/1519] eta 0:15:45 lr 0.000007 time 1.1796 (1.0067) model_time 1.1794 (1.0051) loss 0.6688 (0.8179) grad_norm 10.9496 (8.5747/2.0747) mem 68106MB [2022-12-20 11:14:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][590/1519] eta 0:15:35 lr 0.000007 time 1.0362 (1.0068) model_time 1.0361 (1.0051) loss 0.8099 (0.8180) grad_norm 7.9220 (8.5678/2.0599) mem 68106MB [2022-12-20 11:14:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][600/1519] eta 0:15:25 lr 0.000007 time 0.9194 (1.0066) model_time 0.9193 (1.0050) loss 0.9520 (0.8180) grad_norm 9.5166 (8.5619/2.0453) mem 68106MB [2022-12-20 11:15:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][610/1519] eta 0:15:14 lr 0.000007 time 0.9343 (1.0066) model_time 0.9342 (1.0050) loss 1.0581 (0.8172) grad_norm 8.8765 (8.5770/2.0473) mem 68106MB [2022-12-20 11:15:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][620/1519] eta 0:15:04 lr 0.000007 time 0.9258 (1.0064) model_time 0.9256 (1.0049) loss 0.8217 (0.8163) grad_norm 6.8388 (8.5515/2.0422) mem 68106MB [2022-12-20 11:15:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][630/1519] eta 0:14:54 lr 0.000007 time 0.9382 (1.0063) model_time 0.9381 (1.0048) loss 1.0575 (0.8170) grad_norm 9.4995 (8.5437/2.0430) mem 68106MB [2022-12-20 11:15:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][640/1519] eta 0:14:44 lr 0.000007 time 0.9791 (1.0063) model_time 0.9789 (1.0048) loss 0.9720 (0.8174) grad_norm 15.0484 (8.5352/2.0490) mem 68106MB [2022-12-20 11:15:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][650/1519] eta 0:14:34 lr 0.000007 time 0.9309 (1.0063) model_time 0.9307 (1.0048) loss 1.0509 (0.8190) grad_norm 16.1746 (8.5596/2.0761) mem 68106MB [2022-12-20 11:15:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][660/1519] eta 0:14:24 lr 0.000007 time 0.9352 (1.0064) model_time 0.9350 (1.0049) loss 0.7236 (0.8180) grad_norm 11.9390 (8.5804/2.0844) mem 68106MB [2022-12-20 11:16:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][670/1519] eta 0:14:14 lr 0.000007 time 0.9345 (1.0065) model_time 0.9344 (1.0050) loss 0.6898 (0.8170) grad_norm 7.6843 (8.5711/2.0770) mem 68106MB [2022-12-20 11:16:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][680/1519] eta 0:14:04 lr 0.000007 time 0.9180 (1.0065) model_time 0.9179 (1.0051) loss 0.8089 (0.8168) grad_norm 6.8854 (8.5586/2.0623) mem 68106MB [2022-12-20 11:16:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][690/1519] eta 0:13:54 lr 0.000007 time 0.9235 (1.0066) model_time 0.9234 (1.0052) loss 0.7001 (0.8167) grad_norm 6.4115 (8.5524/2.0623) mem 68106MB [2022-12-20 11:16:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][700/1519] eta 0:13:44 lr 0.000007 time 0.9224 (1.0065) model_time 0.9223 (1.0051) loss 0.7682 (0.8157) grad_norm 6.8681 (8.5744/2.0651) mem 68106MB [2022-12-20 11:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][710/1519] eta 0:13:34 lr 0.000007 time 0.9244 (1.0064) model_time 0.9243 (1.0050) loss 0.8372 (0.8158) grad_norm 9.0508 (8.5810/2.0633) mem 68106MB [2022-12-20 11:16:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][720/1519] eta 0:13:24 lr 0.000007 time 0.9265 (1.0068) model_time 0.9264 (1.0054) loss 0.7658 (0.8155) grad_norm 11.6086 (8.6086/2.0633) mem 68106MB [2022-12-20 11:17:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][730/1519] eta 0:13:14 lr 0.000007 time 0.9391 (1.0069) model_time 0.9390 (1.0055) loss 0.7032 (0.8162) grad_norm 9.4441 (8.6400/2.0604) mem 68106MB [2022-12-20 11:17:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][740/1519] eta 0:13:04 lr 0.000007 time 0.9341 (1.0070) model_time 0.9340 (1.0056) loss 0.8408 (0.8157) grad_norm 7.9165 (8.6286/2.0592) mem 68106MB [2022-12-20 11:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][750/1519] eta 0:12:54 lr 0.000007 time 0.9338 (1.0070) model_time 0.9335 (1.0056) loss 0.8705 (0.8178) grad_norm 9.2592 (8.6126/2.0442) mem 68106MB [2022-12-20 11:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][760/1519] eta 0:12:44 lr 0.000007 time 1.0116 (1.0070) model_time 1.0115 (1.0057) loss 0.9285 (0.8178) grad_norm 5.9709 (8.6144/2.0561) mem 68106MB [2022-12-20 11:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][770/1519] eta 0:12:34 lr 0.000007 time 0.9184 (1.0069) model_time 0.9183 (1.0056) loss 0.8353 (0.8183) grad_norm 12.4022 (8.6173/2.0724) mem 68106MB [2022-12-20 11:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][780/1519] eta 0:12:24 lr 0.000007 time 0.9231 (1.0069) model_time 0.9230 (1.0056) loss 0.7078 (0.8177) grad_norm 8.6693 (8.6037/2.0709) mem 68106MB [2022-12-20 11:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][790/1519] eta 0:12:13 lr 0.000007 time 0.9297 (1.0068) model_time 0.9296 (1.0055) loss 0.6722 (0.8173) grad_norm 8.1932 (8.6050/2.0688) mem 68106MB [2022-12-20 11:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][800/1519] eta 0:12:03 lr 0.000007 time 0.9191 (1.0067) model_time 0.9189 (1.0054) loss 0.8268 (0.8173) grad_norm 8.7602 (8.5989/2.0579) mem 68106MB [2022-12-20 11:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][810/1519] eta 0:11:53 lr 0.000007 time 0.9372 (1.0066) model_time 0.9371 (1.0053) loss 0.7818 (0.8172) grad_norm 7.8593 (8.5804/2.0676) mem 68106MB [2022-12-20 11:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][820/1519] eta 0:11:43 lr 0.000007 time 0.9971 (1.0066) model_time 0.9968 (1.0053) loss 0.8031 (0.8173) grad_norm 9.6205 (8.5928/2.0489) mem 68106MB [2022-12-20 11:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][830/1519] eta 0:11:33 lr 0.000007 time 0.9227 (1.0065) model_time 0.9226 (1.0052) loss 0.6974 (0.8166) grad_norm 8.3047 (8.5561/1.9442) mem 68106MB [2022-12-20 11:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][840/1519] eta 0:11:23 lr 0.000007 time 0.9370 (1.0065) model_time 0.9348 (1.0053) loss 0.7416 (0.8163) grad_norm 8.2688 (8.5778/1.9493) mem 68106MB [2022-12-20 11:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][850/1519] eta 0:11:13 lr 0.000007 time 0.9255 (1.0065) model_time 0.9253 (1.0053) loss 0.7003 (0.8154) grad_norm 7.9535 (8.5929/1.9531) mem 68106MB [2022-12-20 11:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][860/1519] eta 0:11:03 lr 0.000007 time 0.9354 (1.0065) model_time 0.9352 (1.0052) loss 0.8350 (0.8157) grad_norm 7.1468 (8.5912/1.9571) mem 68106MB [2022-12-20 11:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][870/1519] eta 0:10:53 lr 0.000007 time 0.9198 (1.0064) model_time 0.9196 (1.0052) loss 0.6879 (0.8164) grad_norm 8.0752 (8.6118/1.9721) mem 68106MB [2022-12-20 11:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][880/1519] eta 0:10:43 lr 0.000007 time 0.9260 (1.0064) model_time 0.9258 (1.0052) loss 0.6604 (0.8156) grad_norm 6.1107 (8.6403/1.9724) mem 68106MB [2022-12-20 11:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][890/1519] eta 0:10:32 lr 0.000007 time 0.9209 (1.0063) model_time 0.9207 (1.0052) loss 0.7526 (0.8161) grad_norm 16.8894 (8.6375/1.9582) mem 68106MB [2022-12-20 11:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][900/1519] eta 0:10:22 lr 0.000007 time 0.9319 (1.0064) model_time 0.9318 (1.0052) loss 0.6878 (0.8163) grad_norm 9.0057 (8.6316/1.9578) mem 68106MB [2022-12-20 11:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][910/1519] eta 0:10:12 lr 0.000007 time 0.9264 (1.0064) model_time 0.9262 (1.0052) loss 0.9817 (0.8159) grad_norm 9.2840 (8.6297/1.9597) mem 68106MB [2022-12-20 11:20:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][920/1519] eta 0:10:02 lr 0.000007 time 0.9323 (1.0064) model_time 0.9321 (1.0052) loss 0.8582 (0.8162) grad_norm 8.8466 (8.6175/1.9568) mem 68106MB [2022-12-20 11:20:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][930/1519] eta 0:09:52 lr 0.000007 time 0.9319 (1.0063) model_time 0.9318 (1.0051) loss 1.0408 (0.8157) grad_norm 17.3927 (8.6769/2.0943) mem 68106MB [2022-12-20 11:20:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][940/1519] eta 0:09:42 lr 0.000007 time 0.9229 (1.0062) model_time 0.9228 (1.0051) loss 0.7175 (0.8153) grad_norm 8.5338 (8.6946/2.0862) mem 68106MB [2022-12-20 11:20:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][950/1519] eta 0:09:32 lr 0.000007 time 0.9290 (1.0061) model_time 0.9289 (1.0050) loss 0.9763 (0.8158) grad_norm 7.0979 (8.6896/2.0772) mem 68106MB [2022-12-20 11:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][960/1519] eta 0:09:22 lr 0.000007 time 0.9309 (1.0062) model_time 0.9308 (1.0050) loss 1.0127 (0.8169) grad_norm 7.7231 (8.6925/2.0885) mem 68106MB [2022-12-20 11:21:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][970/1519] eta 0:09:12 lr 0.000007 time 0.9253 (1.0061) model_time 0.9251 (1.0050) loss 0.7819 (0.8170) grad_norm 7.5496 (8.7044/2.0921) mem 68106MB [2022-12-20 11:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][980/1519] eta 0:09:02 lr 0.000007 time 0.9224 (1.0063) model_time 0.9223 (1.0052) loss 0.8726 (0.8170) grad_norm 15.2139 (8.7665/2.1548) mem 68106MB [2022-12-20 11:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][990/1519] eta 0:08:52 lr 0.000007 time 0.9294 (1.0063) model_time 0.9292 (1.0052) loss 0.7949 (0.8171) grad_norm 6.5227 (8.7965/2.1744) mem 68106MB [2022-12-20 11:21:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1000/1519] eta 0:08:42 lr 0.000007 time 0.9312 (1.0066) model_time 0.9311 (1.0055) loss 0.8673 (0.8179) grad_norm 11.7853 (8.8199/2.1513) mem 68106MB [2022-12-20 11:21:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1010/1519] eta 0:08:32 lr 0.000007 time 0.9208 (1.0065) model_time 0.9207 (1.0054) loss 0.6623 (0.8177) grad_norm 6.4006 (8.8667/2.1577) mem 68106MB [2022-12-20 11:22:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1020/1519] eta 0:08:22 lr 0.000007 time 0.9894 (1.0065) model_time 0.9893 (1.0055) loss 0.9484 (0.8176) grad_norm 8.1541 (8.8670/2.1479) mem 68106MB [2022-12-20 11:22:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1030/1519] eta 0:08:12 lr 0.000007 time 0.9230 (1.0065) model_time 0.9229 (1.0054) loss 0.8058 (0.8178) grad_norm 11.3676 (8.8902/2.1398) mem 68106MB [2022-12-20 11:22:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1040/1519] eta 0:08:02 lr 0.000007 time 0.9256 (1.0064) model_time 0.9255 (1.0054) loss 0.8658 (0.8179) grad_norm 8.3220 (8.9063/2.1375) mem 68106MB [2022-12-20 11:22:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1050/1519] eta 0:07:52 lr 0.000007 time 0.9226 (1.0065) model_time 0.9224 (1.0054) loss 0.9244 (0.8178) grad_norm 7.9995 (8.9003/2.1413) mem 68106MB [2022-12-20 11:22:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1060/1519] eta 0:07:42 lr 0.000007 time 0.9800 (1.0066) model_time 0.9799 (1.0055) loss 0.6674 (0.8173) grad_norm 8.7022 (8.9081/2.1414) mem 68106MB [2022-12-20 11:22:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1070/1519] eta 0:07:31 lr 0.000007 time 0.9284 (1.0065) model_time 0.9283 (1.0055) loss 0.6707 (0.8175) grad_norm 7.7076 (8.9056/2.1332) mem 68106MB [2022-12-20 11:23:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1080/1519] eta 0:07:21 lr 0.000007 time 0.9399 (1.0065) model_time 0.9398 (1.0055) loss 1.0869 (0.8181) grad_norm 10.2746 (8.9184/2.1343) mem 68106MB [2022-12-20 11:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1090/1519] eta 0:07:11 lr 0.000007 time 0.9224 (1.0065) model_time 0.9223 (1.0055) loss 0.7398 (0.8175) grad_norm 8.8491 (8.9246/2.1423) mem 68106MB [2022-12-20 11:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1100/1519] eta 0:07:01 lr 0.000007 time 0.9375 (1.0064) model_time 0.9374 (1.0054) loss 0.7796 (0.8177) grad_norm 8.0408 (8.9418/2.1379) mem 68106MB [2022-12-20 11:23:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1110/1519] eta 0:06:51 lr 0.000007 time 0.9187 (1.0064) model_time 0.9186 (1.0053) loss 0.9816 (0.8176) grad_norm 8.7676 (8.9094/2.1118) mem 68106MB [2022-12-20 11:23:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1120/1519] eta 0:06:41 lr 0.000007 time 0.9329 (1.0063) model_time 0.9327 (1.0053) loss 0.9465 (0.8173) grad_norm 9.6164 (8.9036/2.0930) mem 68106MB [2022-12-20 11:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1130/1519] eta 0:06:31 lr 0.000007 time 0.9229 (1.0062) model_time 0.9228 (1.0052) loss 0.6818 (0.8168) grad_norm 5.9653 (8.8864/2.1039) mem 68106MB [2022-12-20 11:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1140/1519] eta 0:06:21 lr 0.000007 time 0.9243 (1.0062) model_time 0.9241 (1.0053) loss 0.6855 (0.8168) grad_norm 6.7708 (8.8535/2.0746) mem 68106MB [2022-12-20 11:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1150/1519] eta 0:06:11 lr 0.000007 time 0.9055 (1.0063) model_time 0.9053 (1.0053) loss 0.8417 (0.8170) grad_norm 6.4346 (8.8533/2.0727) mem 68106MB [2022-12-20 11:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1160/1519] eta 0:06:01 lr 0.000007 time 0.9322 (1.0063) model_time 0.9320 (1.0053) loss 0.8708 (0.8168) grad_norm 5.7420 (8.8541/2.0744) mem 68106MB [2022-12-20 11:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1170/1519] eta 0:05:51 lr 0.000007 time 0.9321 (1.0063) model_time 0.9319 (1.0053) loss 0.8300 (0.8168) grad_norm 9.6007 (8.8634/2.0769) mem 68106MB [2022-12-20 11:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1180/1519] eta 0:05:41 lr 0.000007 time 0.9269 (1.0064) model_time 0.9268 (1.0054) loss 0.8248 (0.8163) grad_norm 9.7932 (8.8627/2.0792) mem 68106MB [2022-12-20 11:24:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1190/1519] eta 0:05:31 lr 0.000007 time 0.9261 (1.0063) model_time 0.9260 (1.0054) loss 0.8178 (0.8168) grad_norm 8.9136 (8.8928/2.0835) mem 68106MB [2022-12-20 11:25:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1200/1519] eta 0:05:21 lr 0.000007 time 0.9928 (1.0064) model_time 0.9927 (1.0054) loss 0.8030 (0.8165) grad_norm 12.6620 (8.9023/2.1009) mem 68106MB [2022-12-20 11:25:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1210/1519] eta 0:05:10 lr 0.000007 time 0.9231 (1.0064) model_time 0.9230 (1.0054) loss 0.8349 (0.8171) grad_norm 8.5565 (8.9047/2.1014) mem 68106MB [2022-12-20 11:25:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1220/1519] eta 0:05:00 lr 0.000007 time 0.9331 (1.0064) model_time 0.9330 (1.0055) loss 0.7610 (0.8168) grad_norm 8.7483 (8.9325/2.0947) mem 68106MB [2022-12-20 11:25:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1230/1519] eta 0:04:50 lr 0.000007 time 0.9253 (1.0064) model_time 0.9252 (1.0054) loss 0.9734 (0.8168) grad_norm 8.0945 (8.9337/2.1044) mem 68106MB [2022-12-20 11:25:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1240/1519] eta 0:04:40 lr 0.000007 time 0.9209 (1.0064) model_time 0.9208 (1.0054) loss 0.7876 (0.8168) grad_norm 5.7491 (8.8988/2.0837) mem 68106MB [2022-12-20 11:25:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1250/1519] eta 0:04:30 lr 0.000007 time 0.9308 (1.0063) model_time 0.9306 (1.0054) loss 0.8567 (0.8170) grad_norm 7.2049 (8.8876/2.0372) mem 68106MB [2022-12-20 11:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1260/1519] eta 0:04:20 lr 0.000007 time 0.9250 (1.0063) model_time 0.9248 (1.0053) loss 0.6703 (0.8166) grad_norm 6.8091 (8.8684/2.0300) mem 68106MB [2022-12-20 11:26:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1270/1519] eta 0:04:10 lr 0.000007 time 1.0749 (1.0063) model_time 1.0748 (1.0054) loss 0.8745 (0.8160) grad_norm 8.3533 (8.8517/2.0316) mem 68106MB [2022-12-20 11:26:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1280/1519] eta 0:04:00 lr 0.000007 time 0.9183 (1.0062) model_time 0.9181 (1.0053) loss 0.9827 (0.8159) grad_norm 15.2220 (8.8767/2.0744) mem 68106MB [2022-12-20 11:26:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1290/1519] eta 0:03:50 lr 0.000007 time 0.9290 (1.0062) model_time 0.9288 (1.0053) loss 0.7745 (0.8158) grad_norm 8.4572 (8.8763/2.0784) mem 68106MB [2022-12-20 11:26:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1300/1519] eta 0:03:40 lr 0.000007 time 0.9217 (1.0062) model_time 0.9216 (1.0052) loss 0.6915 (0.8155) grad_norm 7.5419 (8.8850/2.0620) mem 68106MB [2022-12-20 11:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1310/1519] eta 0:03:30 lr 0.000007 time 0.9219 (1.0061) model_time 0.9217 (1.0052) loss 1.1171 (0.8161) grad_norm 6.6966 (8.8755/2.0778) mem 68106MB [2022-12-20 11:27:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1320/1519] eta 0:03:20 lr 0.000007 time 0.9255 (1.0060) model_time 0.9253 (1.0051) loss 0.8839 (0.8165) grad_norm 8.5620 (8.8697/2.0769) mem 68106MB [2022-12-20 11:27:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1330/1519] eta 0:03:10 lr 0.000007 time 0.9182 (1.0060) model_time 0.9181 (1.0051) loss 0.6776 (0.8164) grad_norm 9.4537 (8.9085/2.1341) mem 68106MB [2022-12-20 11:27:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1340/1519] eta 0:03:00 lr 0.000007 time 0.9375 (1.0060) model_time 0.9374 (1.0051) loss 0.8138 (0.8166) grad_norm 6.7699 (8.9166/2.1225) mem 68106MB [2022-12-20 11:27:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1350/1519] eta 0:02:50 lr 0.000007 time 0.9318 (1.0059) model_time 0.9315 (1.0050) loss 0.7592 (0.8162) grad_norm 20.6889 (8.9524/2.2287) mem 68106MB [2022-12-20 11:27:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1360/1519] eta 0:02:39 lr 0.000007 time 1.0214 (1.0059) model_time 1.0213 (1.0051) loss 0.6989 (0.8166) grad_norm 7.2846 (8.9562/2.2168) mem 68106MB [2022-12-20 11:27:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1370/1519] eta 0:02:29 lr 0.000007 time 0.9285 (1.0059) model_time 0.9284 (1.0051) loss 0.7561 (0.8168) grad_norm 9.0100 (8.9699/2.2088) mem 68106MB [2022-12-20 11:28:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1380/1519] eta 0:02:19 lr 0.000007 time 0.9231 (1.0062) model_time 0.9230 (1.0053) loss 0.6671 (0.8165) grad_norm 7.0438 (8.9515/2.1952) mem 68106MB [2022-12-20 11:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1390/1519] eta 0:02:09 lr 0.000007 time 0.9355 (1.0062) model_time 0.9353 (1.0053) loss 0.8505 (0.8162) grad_norm 8.2471 (8.9451/2.1957) mem 68106MB [2022-12-20 11:28:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1400/1519] eta 0:01:59 lr 0.000007 time 1.0068 (1.0062) model_time 1.0067 (1.0053) loss 1.0053 (0.8161) grad_norm 8.0185 (8.9395/2.2038) mem 68106MB [2022-12-20 11:28:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1410/1519] eta 0:01:49 lr 0.000007 time 0.9281 (1.0061) model_time 0.9280 (1.0053) loss 0.9444 (0.8160) grad_norm 7.1725 (8.9611/2.2003) mem 68106MB [2022-12-20 11:28:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1420/1519] eta 0:01:39 lr 0.000007 time 0.9294 (1.0061) model_time 0.9291 (1.0052) loss 0.6726 (0.8162) grad_norm 6.5526 (8.9561/2.2066) mem 68106MB [2022-12-20 11:28:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1430/1519] eta 0:01:29 lr 0.000007 time 0.9198 (1.0060) model_time 0.9196 (1.0052) loss 0.9855 (0.8164) grad_norm 7.6264 (8.9532/2.2048) mem 68106MB [2022-12-20 11:29:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1440/1519] eta 0:01:19 lr 0.000007 time 0.9229 (1.0060) model_time 0.9228 (1.0051) loss 0.7589 (0.8162) grad_norm 7.1089 (8.9200/2.2071) mem 68106MB [2022-12-20 11:29:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1450/1519] eta 0:01:09 lr 0.000007 time 0.9820 (1.0060) model_time 0.9819 (1.0052) loss 0.8767 (0.8160) grad_norm 8.5768 (8.9143/2.2104) mem 68106MB [2022-12-20 11:29:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1460/1519] eta 0:00:59 lr 0.000007 time 0.9270 (1.0060) model_time 0.9268 (1.0051) loss 1.0033 (0.8160) grad_norm 8.6427 (8.9084/2.2119) mem 68106MB [2022-12-20 11:29:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1470/1519] eta 0:00:49 lr 0.000007 time 0.9230 (1.0061) model_time 0.9228 (1.0052) loss 0.7352 (0.8162) grad_norm 10.6410 (8.9230/2.2165) mem 68106MB [2022-12-20 11:29:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1480/1519] eta 0:00:39 lr 0.000007 time 0.9367 (1.0060) model_time 0.9366 (1.0052) loss 0.6682 (0.8167) grad_norm 8.7393 (8.9088/2.2124) mem 68106MB [2022-12-20 11:29:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1490/1519] eta 0:00:29 lr 0.000007 time 0.9186 (1.0061) model_time 0.9183 (1.0052) loss 0.8159 (0.8170) grad_norm 9.7304 (8.8887/2.1691) mem 68106MB [2022-12-20 11:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1500/1519] eta 0:00:19 lr 0.000007 time 0.9194 (1.0061) model_time 0.9193 (1.0052) loss 1.1378 (0.8170) grad_norm 8.5353 (8.8999/2.1616) mem 68106MB [2022-12-20 11:30:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [70/100][1510/1519] eta 0:00:09 lr 0.000007 time 0.9044 (1.0060) model_time 0.9043 (1.0052) loss 0.7714 (0.8167) grad_norm 6.0436 (8.8880/2.1721) mem 68106MB [2022-12-20 11:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 70 training takes 0:25:28 [2022-12-20 11:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_70.pth saving...... [2022-12-20 11:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_70.pth saved !!! [2022-12-20 11:30:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.676 (0.676) Loss 0.5229 (0.5229) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 11:30:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.330) Loss 0.5324 (0.5015) Acc@1 92.014 (92.771) Acc@5 98.264 (98.516) Mem 68106MB [2022-12-20 11:30:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4837 (0.4971) Acc@1 91.667 (92.808) Acc@5 98.958 (98.495) Mem 68106MB [2022-12-20 11:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.309) Loss 0.6287 (0.5045) Acc@1 89.236 (92.462) Acc@5 97.917 (98.454) Mem 68106MB [2022-12-20 11:30:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.306) Loss 0.4502 (0.4951) Acc@1 94.097 (92.530) Acc@5 98.958 (98.526) Mem 68106MB [2022-12-20 11:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.304) Loss 0.4824 (0.4924) Acc@1 91.667 (92.572) Acc@5 99.653 (98.591) Mem 68106MB [2022-12-20 11:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.304) Loss 0.5144 (0.4923) Acc@1 90.278 (92.492) Acc@5 98.264 (98.560) Mem 68106MB [2022-12-20 11:31:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5423 (0.4933) Acc@1 93.056 (92.454) Acc@5 97.917 (98.548) Mem 68106MB [2022-12-20 11:31:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4339 (0.4921) Acc@1 93.750 (92.485) Acc@5 98.958 (98.585) Mem 68106MB [2022-12-20 11:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:70] * Acc@1 92.416 Acc@5 98.588 [2022-12-20 11:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 11:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.46% [2022-12-20 11:31:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][0/1519] eta 0:49:27 lr 0.000007 time 1.9538 (1.9538) model_time 1.1539 (1.1539) loss 1.0264 (1.0264) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 11:31:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][10/1519] eta 0:27:22 lr 0.000007 time 0.9300 (1.0886) model_time 0.9299 (1.0156) loss 0.8860 (0.8267) grad_norm 14.9528 (8.9621/3.3247) mem 68106MB [2022-12-20 11:31:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][20/1519] eta 0:26:08 lr 0.000007 time 0.9316 (1.0462) model_time 0.9315 (1.0078) loss 0.6973 (0.8249) grad_norm 7.9577 (8.4291/2.6171) mem 68106MB [2022-12-20 11:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][30/1519] eta 0:25:45 lr 0.000007 time 0.9262 (1.0381) model_time 0.9260 (1.0120) loss 0.7117 (0.8280) grad_norm 5.3205 (8.4020/2.6844) mem 68106MB [2022-12-20 11:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][40/1519] eta 0:25:25 lr 0.000007 time 0.9200 (1.0312) model_time 0.9198 (1.0113) loss 0.7862 (0.8096) grad_norm 8.6094 (8.3356/2.4219) mem 68106MB [2022-12-20 11:32:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][50/1519] eta 0:25:05 lr 0.000007 time 0.9261 (1.0249) model_time 0.9260 (1.0089) loss 0.7169 (0.8027) grad_norm 7.2868 (8.2263/2.3076) mem 68106MB [2022-12-20 11:32:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][60/1519] eta 0:24:51 lr 0.000007 time 0.9239 (1.0222) model_time 0.9237 (1.0087) loss 0.8958 (0.8018) grad_norm 6.6288 (8.3676/2.2270) mem 68106MB [2022-12-20 11:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][70/1519] eta 0:24:36 lr 0.000007 time 0.9273 (1.0191) model_time 0.9272 (1.0075) loss 0.6699 (0.8067) grad_norm 9.2867 (8.3118/2.1162) mem 68106MB [2022-12-20 11:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][80/1519] eta 0:24:22 lr 0.000007 time 0.9251 (1.0165) model_time 0.9250 (1.0064) loss 1.0243 (0.8123) grad_norm 7.8671 (8.2272/1.9952) mem 68106MB [2022-12-20 11:32:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][90/1519] eta 0:24:09 lr 0.000007 time 0.9248 (1.0147) model_time 0.9247 (1.0056) loss 0.6690 (0.8098) grad_norm 6.3760 (8.2180/1.9610) mem 68106MB [2022-12-20 11:32:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][100/1519] eta 0:23:57 lr 0.000007 time 0.9234 (1.0131) model_time 0.9233 (1.0049) loss 0.6873 (0.8021) grad_norm 12.6179 (8.2342/2.0192) mem 68106MB [2022-12-20 11:33:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][110/1519] eta 0:23:46 lr 0.000007 time 0.9255 (1.0127) model_time 0.9254 (1.0052) loss 0.8475 (0.8049) grad_norm 7.5093 (8.3978/2.0354) mem 68106MB [2022-12-20 11:33:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][120/1519] eta 0:23:35 lr 0.000007 time 0.9238 (1.0119) model_time 0.9236 (1.0050) loss 0.8581 (0.8039) grad_norm 7.8700 (8.5322/2.1013) mem 68106MB [2022-12-20 11:33:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][130/1519] eta 0:23:24 lr 0.000007 time 0.9355 (1.0115) model_time 0.9354 (1.0051) loss 0.8237 (0.8000) grad_norm 8.2946 (8.5147/2.0383) mem 68106MB [2022-12-20 11:33:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][140/1519] eta 0:23:13 lr 0.000007 time 0.9219 (1.0106) model_time 0.9216 (1.0046) loss 0.7909 (0.8021) grad_norm 9.7237 (8.5623/2.0772) mem 68106MB [2022-12-20 11:33:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][150/1519] eta 0:23:03 lr 0.000007 time 0.9321 (1.0105) model_time 0.9319 (1.0049) loss 0.7247 (0.7989) grad_norm 11.0577 (8.5762/2.0452) mem 68106MB [2022-12-20 11:33:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][160/1519] eta 0:22:53 lr 0.000007 time 0.9936 (1.0104) model_time 0.9935 (1.0051) loss 1.0196 (0.8033) grad_norm 7.3305 (8.6086/2.0262) mem 68106MB [2022-12-20 11:34:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][170/1519] eta 0:22:43 lr 0.000007 time 0.9279 (1.0111) model_time 0.9278 (1.0061) loss 0.8100 (0.8049) grad_norm 7.0846 (8.6228/2.0058) mem 68106MB [2022-12-20 11:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][180/1519] eta 0:22:36 lr 0.000007 time 1.1983 (1.0127) model_time 1.1982 (1.0080) loss 1.0191 (0.8058) grad_norm 13.3183 (8.6471/2.0157) mem 68106MB [2022-12-20 11:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][190/1519] eta 0:22:24 lr 0.000007 time 0.9218 (1.0120) model_time 0.9217 (1.0075) loss 0.7917 (0.8077) grad_norm 8.5972 (8.6047/1.9768) mem 68106MB [2022-12-20 11:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][200/1519] eta 0:22:13 lr 0.000007 time 0.9190 (1.0113) model_time 0.9189 (1.0070) loss 0.8935 (0.8080) grad_norm 9.3866 (8.5778/1.9523) mem 68106MB [2022-12-20 11:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][210/1519] eta 0:22:02 lr 0.000007 time 0.9179 (1.0106) model_time 0.9178 (1.0065) loss 0.8831 (0.8121) grad_norm 9.6359 (8.5698/1.9125) mem 68106MB [2022-12-20 11:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][220/1519] eta 0:21:52 lr 0.000007 time 0.9236 (1.0100) model_time 0.9234 (1.0061) loss 0.6657 (0.8099) grad_norm 7.4800 (8.5624/1.8774) mem 68106MB [2022-12-20 11:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][230/1519] eta 0:21:41 lr 0.000007 time 0.9319 (1.0095) model_time 0.9317 (1.0057) loss 0.6758 (0.8100) grad_norm 7.2013 (8.5808/1.8656) mem 68106MB [2022-12-20 11:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][240/1519] eta 0:21:31 lr 0.000007 time 0.9140 (1.0094) model_time 0.9139 (1.0057) loss 0.7921 (0.8105) grad_norm 7.1789 (8.5878/1.8688) mem 68106MB [2022-12-20 11:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][250/1519] eta 0:21:20 lr 0.000007 time 0.9251 (1.0091) model_time 0.9250 (1.0056) loss 0.7997 (0.8084) grad_norm 8.0959 (8.6003/1.8762) mem 68106MB [2022-12-20 11:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][260/1519] eta 0:21:10 lr 0.000007 time 0.9271 (1.0088) model_time 0.9270 (1.0054) loss 0.7057 (0.8097) grad_norm 8.2846 (8.6363/1.8959) mem 68106MB [2022-12-20 11:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][270/1519] eta 0:21:00 lr 0.000007 time 0.9292 (1.0089) model_time 0.9291 (1.0056) loss 1.2306 (0.8099) grad_norm 9.2754 (8.6237/1.8721) mem 68106MB [2022-12-20 11:35:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][280/1519] eta 0:20:49 lr 0.000007 time 0.9246 (1.0087) model_time 0.9244 (1.0056) loss 0.8758 (0.8101) grad_norm 10.8637 (8.6031/1.8665) mem 68106MB [2022-12-20 11:36:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][290/1519] eta 0:20:39 lr 0.000007 time 0.9263 (1.0085) model_time 0.9261 (1.0054) loss 1.0946 (0.8125) grad_norm 8.1133 (8.6650/1.8798) mem 68106MB [2022-12-20 11:36:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][300/1519] eta 0:20:29 lr 0.000007 time 0.9471 (1.0082) model_time 0.9470 (1.0052) loss 0.8266 (0.8114) grad_norm 9.7692 (8.6679/1.8595) mem 68106MB [2022-12-20 11:36:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][310/1519] eta 0:20:18 lr 0.000007 time 0.9196 (1.0080) model_time 0.9195 (1.0051) loss 0.9672 (0.8124) grad_norm 5.4583 (8.6503/1.8642) mem 68106MB [2022-12-20 11:36:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][320/1519] eta 0:20:08 lr 0.000007 time 1.0075 (1.0079) model_time 1.0074 (1.0051) loss 0.6736 (0.8124) grad_norm 8.4711 (8.6354/1.8384) mem 68106MB [2022-12-20 11:36:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][330/1519] eta 0:19:58 lr 0.000007 time 0.9294 (1.0077) model_time 0.9292 (1.0049) loss 0.8883 (0.8126) grad_norm 8.4197 (8.6433/1.8408) mem 68106MB [2022-12-20 11:36:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][340/1519] eta 0:19:48 lr 0.000007 time 0.9976 (1.0077) model_time 0.9974 (1.0050) loss 0.7913 (0.8153) grad_norm 7.9867 (8.6646/1.8388) mem 68106MB [2022-12-20 11:37:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][350/1519] eta 0:19:38 lr 0.000007 time 0.9292 (1.0080) model_time 0.9290 (1.0053) loss 0.8265 (0.8153) grad_norm 11.0516 (8.6741/1.8451) mem 68106MB [2022-12-20 11:37:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][360/1519] eta 0:19:27 lr 0.000007 time 0.9491 (1.0077) model_time 0.9489 (1.0052) loss 0.9136 (0.8138) grad_norm 6.7340 (8.6660/1.8253) mem 68106MB [2022-12-20 11:37:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][370/1519] eta 0:19:17 lr 0.000007 time 0.9583 (1.0077) model_time 0.9580 (1.0052) loss 0.7623 (0.8142) grad_norm 10.3697 (8.6506/1.8251) mem 68106MB [2022-12-20 11:37:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][380/1519] eta 0:19:07 lr 0.000007 time 0.9274 (1.0075) model_time 0.9273 (1.0050) loss 0.7440 (0.8134) grad_norm 6.2435 (8.6245/1.8283) mem 68106MB [2022-12-20 11:37:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][390/1519] eta 0:18:57 lr 0.000007 time 0.9343 (1.0073) model_time 0.9342 (1.0049) loss 0.9090 (0.8137) grad_norm 7.3868 (8.6164/1.8397) mem 68106MB [2022-12-20 11:37:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][400/1519] eta 0:18:46 lr 0.000007 time 0.9282 (1.0071) model_time 0.9280 (1.0048) loss 0.9301 (0.8133) grad_norm 8.3807 (8.6035/1.8203) mem 68106MB [2022-12-20 11:38:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][410/1519] eta 0:18:36 lr 0.000007 time 0.9344 (1.0070) model_time 0.9342 (1.0047) loss 1.1759 (0.8144) grad_norm 9.8555 (8.5892/1.8122) mem 68106MB [2022-12-20 11:38:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][420/1519] eta 0:18:26 lr 0.000007 time 0.9231 (1.0070) model_time 0.9230 (1.0048) loss 0.7597 (0.8136) grad_norm 12.1289 (8.6221/1.8141) mem 68106MB [2022-12-20 11:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][430/1519] eta 0:18:16 lr 0.000007 time 0.9629 (1.0070) model_time 0.9627 (1.0048) loss 0.6761 (0.8131) grad_norm 7.0727 (8.6152/1.8026) mem 68106MB [2022-12-20 11:38:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][440/1519] eta 0:18:06 lr 0.000007 time 0.9318 (1.0069) model_time 0.9316 (1.0048) loss 0.8333 (0.8129) grad_norm 10.1755 (8.6226/1.8057) mem 68106MB [2022-12-20 11:38:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][450/1519] eta 0:17:56 lr 0.000007 time 0.9309 (1.0070) model_time 0.9307 (1.0049) loss 0.9146 (0.8147) grad_norm 7.3586 (8.6309/1.8076) mem 68106MB [2022-12-20 11:38:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][460/1519] eta 0:17:46 lr 0.000007 time 0.9124 (1.0071) model_time 0.9123 (1.0050) loss 0.9033 (0.8147) grad_norm 17.2669 (8.6686/1.8834) mem 68106MB [2022-12-20 11:39:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][470/1519] eta 0:17:36 lr 0.000007 time 0.9232 (1.0070) model_time 0.9231 (1.0050) loss 0.9262 (0.8138) grad_norm 6.9552 (8.6598/1.8857) mem 68106MB [2022-12-20 11:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][480/1519] eta 0:17:26 lr 0.000007 time 0.9242 (1.0073) model_time 0.9240 (1.0053) loss 0.6846 (0.8141) grad_norm 11.8376 (8.7329/2.0932) mem 68106MB [2022-12-20 11:39:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][490/1519] eta 0:17:16 lr 0.000007 time 0.9358 (1.0074) model_time 0.9357 (1.0054) loss 0.9484 (0.8160) grad_norm 7.3155 (8.7022/2.0868) mem 68106MB [2022-12-20 11:39:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][500/1519] eta 0:17:06 lr 0.000007 time 1.0235 (1.0076) model_time 1.0234 (1.0057) loss 1.0521 (0.8176) grad_norm 8.5411 (8.7130/2.0904) mem 68106MB [2022-12-20 11:39:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][510/1519] eta 0:16:56 lr 0.000007 time 0.9215 (1.0074) model_time 0.9214 (1.0055) loss 0.7334 (0.8173) grad_norm 8.2340 (8.7092/2.0711) mem 68106MB [2022-12-20 11:39:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][520/1519] eta 0:16:46 lr 0.000007 time 0.9895 (1.0074) model_time 0.9894 (1.0055) loss 0.6758 (0.8174) grad_norm 7.5432 (8.7030/2.0599) mem 68106MB [2022-12-20 11:40:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][530/1519] eta 0:16:36 lr 0.000007 time 0.9286 (1.0073) model_time 0.9284 (1.0054) loss 0.8453 (0.8167) grad_norm 16.5526 (8.7502/2.1059) mem 68106MB [2022-12-20 11:40:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][540/1519] eta 0:16:25 lr 0.000007 time 0.9213 (1.0071) model_time 0.9212 (1.0053) loss 0.7848 (0.8166) grad_norm 8.2599 (8.7474/2.0960) mem 68106MB [2022-12-20 11:40:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][550/1519] eta 0:16:15 lr 0.000007 time 0.9175 (1.0069) model_time 0.9174 (1.0052) loss 0.7186 (0.8160) grad_norm 6.0075 (8.7289/2.0984) mem 68106MB [2022-12-20 11:40:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][560/1519] eta 0:16:05 lr 0.000007 time 0.9271 (1.0070) model_time 0.9269 (1.0053) loss 0.6758 (0.8166) grad_norm 9.0635 (8.7451/2.0997) mem 68106MB [2022-12-20 11:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][570/1519] eta 0:15:55 lr 0.000007 time 0.9214 (1.0069) model_time 0.9212 (1.0052) loss 0.7767 (0.8163) grad_norm 11.8865 (8.7822/2.1643) mem 68106MB [2022-12-20 11:40:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][580/1519] eta 0:15:45 lr 0.000007 time 0.9216 (1.0068) model_time 0.9215 (1.0050) loss 0.6877 (0.8162) grad_norm 9.3197 (8.8010/2.1839) mem 68106MB [2022-12-20 11:41:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][590/1519] eta 0:15:35 lr 0.000007 time 0.9254 (1.0067) model_time 0.9252 (1.0050) loss 0.8705 (0.8157) grad_norm 7.2763 (8.7877/2.1754) mem 68106MB [2022-12-20 11:41:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][600/1519] eta 0:15:24 lr 0.000007 time 0.9207 (1.0065) model_time 0.9206 (1.0048) loss 0.6750 (0.8154) grad_norm 8.4360 (8.7942/2.1624) mem 68106MB [2022-12-20 11:41:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][610/1519] eta 0:15:14 lr 0.000007 time 0.9682 (1.0065) model_time 0.9681 (1.0048) loss 0.7054 (0.8143) grad_norm 12.6894 (8.8501/2.2454) mem 68106MB [2022-12-20 11:41:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][620/1519] eta 0:15:04 lr 0.000007 time 0.9274 (1.0064) model_time 0.9273 (1.0048) loss 0.6886 (0.8137) grad_norm 7.2889 (8.8744/2.2416) mem 68106MB [2022-12-20 11:41:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][630/1519] eta 0:14:54 lr 0.000007 time 0.9298 (1.0063) model_time 0.9296 (1.0047) loss 0.8748 (0.8146) grad_norm 6.4862 (8.8542/2.2238) mem 68106MB [2022-12-20 11:41:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][640/1519] eta 0:14:44 lr 0.000007 time 0.9253 (1.0062) model_time 0.9252 (1.0046) loss 1.0910 (0.8165) grad_norm 10.3429 (8.8484/2.2270) mem 68106MB [2022-12-20 11:42:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][650/1519] eta 0:14:34 lr 0.000007 time 0.9247 (1.0061) model_time 0.9246 (1.0045) loss 0.7309 (0.8173) grad_norm 7.6810 (8.8828/2.2246) mem 68106MB [2022-12-20 11:42:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][660/1519] eta 0:14:24 lr 0.000007 time 0.8907 (1.0064) model_time 0.8905 (1.0048) loss 0.9190 (0.8167) grad_norm 8.1029 (8.8799/2.2198) mem 68106MB [2022-12-20 11:42:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][670/1519] eta 0:14:14 lr 0.000007 time 0.9273 (1.0062) model_time 0.9271 (1.0047) loss 0.8682 (0.8167) grad_norm 9.4237 (8.8886/2.2186) mem 68106MB [2022-12-20 11:42:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][680/1519] eta 0:14:04 lr 0.000007 time 0.9218 (1.0061) model_time 0.9217 (1.0046) loss 1.0772 (0.8176) grad_norm 11.4390 (8.9226/2.2240) mem 68106MB [2022-12-20 11:42:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][690/1519] eta 0:13:54 lr 0.000007 time 0.9245 (1.0061) model_time 0.9244 (1.0046) loss 0.8138 (0.8177) grad_norm 9.2047 (8.9385/2.2222) mem 68106MB [2022-12-20 11:42:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][700/1519] eta 0:13:43 lr 0.000007 time 0.9279 (1.0060) model_time 0.9277 (1.0045) loss 0.8098 (0.8176) grad_norm 10.5228 (8.9688/2.2305) mem 68106MB [2022-12-20 11:43:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][710/1519] eta 0:13:33 lr 0.000007 time 0.9258 (1.0058) model_time 0.9256 (1.0044) loss 0.7233 (0.8168) grad_norm 10.3350 (8.9409/2.2289) mem 68106MB [2022-12-20 11:43:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][720/1519] eta 0:13:23 lr 0.000007 time 0.9236 (1.0058) model_time 0.9235 (1.0043) loss 0.7397 (0.8164) grad_norm 7.6792 (8.9073/2.2122) mem 68106MB [2022-12-20 11:43:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][730/1519] eta 0:13:13 lr 0.000007 time 0.9310 (1.0058) model_time 0.9308 (1.0044) loss 0.9172 (0.8175) grad_norm 10.6893 (8.9146/2.2123) mem 68106MB [2022-12-20 11:43:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][740/1519] eta 0:13:03 lr 0.000007 time 0.9356 (1.0058) model_time 0.9355 (1.0044) loss 0.8169 (0.8178) grad_norm 12.4261 (8.9276/2.2494) mem 68106MB [2022-12-20 11:43:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][750/1519] eta 0:12:53 lr 0.000007 time 0.9239 (1.0057) model_time 0.9238 (1.0043) loss 0.9814 (0.8189) grad_norm 7.8761 (8.9115/2.2506) mem 68106MB [2022-12-20 11:43:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][760/1519] eta 0:12:43 lr 0.000007 time 0.9245 (1.0058) model_time 0.9244 (1.0044) loss 0.7256 (0.8194) grad_norm 6.5701 (8.8882/2.2505) mem 68106MB [2022-12-20 11:44:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][770/1519] eta 0:12:33 lr 0.000007 time 0.9225 (1.0058) model_time 0.9223 (1.0044) loss 0.7070 (0.8197) grad_norm 8.9618 (8.9032/2.2584) mem 68106MB [2022-12-20 11:44:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][780/1519] eta 0:12:23 lr 0.000007 time 0.9359 (1.0058) model_time 0.9357 (1.0044) loss 0.8705 (0.8198) grad_norm 7.0362 (8.8834/2.2465) mem 68106MB [2022-12-20 11:44:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][790/1519] eta 0:12:13 lr 0.000007 time 0.9210 (1.0058) model_time 0.9208 (1.0044) loss 0.8239 (0.8195) grad_norm 7.4969 (8.8902/2.2501) mem 68106MB [2022-12-20 11:44:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][800/1519] eta 0:12:03 lr 0.000007 time 0.9690 (1.0058) model_time 0.9688 (1.0045) loss 0.6800 (0.8202) grad_norm 11.1706 (8.9057/2.2506) mem 68106MB [2022-12-20 11:44:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][810/1519] eta 0:11:53 lr 0.000007 time 0.9260 (1.0058) model_time 0.9258 (1.0045) loss 0.8876 (0.8195) grad_norm 5.5340 (8.8942/2.2700) mem 68106MB [2022-12-20 11:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][820/1519] eta 0:11:43 lr 0.000007 time 0.9354 (1.0059) model_time 0.9353 (1.0046) loss 1.0224 (0.8200) grad_norm 8.5188 (8.9176/2.2832) mem 68106MB [2022-12-20 11:45:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][830/1519] eta 0:11:33 lr 0.000007 time 0.9213 (1.0058) model_time 0.9212 (1.0045) loss 0.9271 (0.8205) grad_norm 7.7508 (8.9144/2.2784) mem 68106MB [2022-12-20 11:45:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][840/1519] eta 0:11:23 lr 0.000007 time 0.9210 (1.0059) model_time 0.9209 (1.0046) loss 0.6820 (0.8203) grad_norm 7.6739 (8.8965/2.2739) mem 68106MB [2022-12-20 11:45:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][850/1519] eta 0:11:12 lr 0.000007 time 0.9278 (1.0059) model_time 0.9276 (1.0046) loss 0.7801 (0.8205) grad_norm 8.4366 (8.8931/2.2616) mem 68106MB [2022-12-20 11:45:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][860/1519] eta 0:11:02 lr 0.000007 time 0.9334 (1.0058) model_time 0.9332 (1.0045) loss 0.8624 (0.8211) grad_norm 7.9314 (8.8647/2.2509) mem 68106MB [2022-12-20 11:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][870/1519] eta 0:10:52 lr 0.000007 time 0.9197 (1.0058) model_time 0.9196 (1.0046) loss 0.7648 (0.8209) grad_norm 8.6605 (8.8858/2.2648) mem 68106MB [2022-12-20 11:45:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][880/1519] eta 0:10:42 lr 0.000007 time 0.9282 (1.0057) model_time 0.9281 (1.0045) loss 1.1541 (0.8213) grad_norm 10.2903 (8.8938/2.2572) mem 68106MB [2022-12-20 11:46:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][890/1519] eta 0:10:32 lr 0.000007 time 0.9312 (1.0057) model_time 0.9310 (1.0044) loss 0.8223 (0.8209) grad_norm 11.7870 (8.8606/2.2575) mem 68106MB [2022-12-20 11:46:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][900/1519] eta 0:10:22 lr 0.000007 time 0.9315 (1.0056) model_time 0.9313 (1.0044) loss 0.8137 (0.8208) grad_norm 7.4079 (8.8697/2.2673) mem 68106MB [2022-12-20 11:46:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][910/1519] eta 0:10:12 lr 0.000007 time 0.9286 (1.0056) model_time 0.9284 (1.0044) loss 0.8544 (0.8206) grad_norm 9.0950 (8.8824/2.2632) mem 68106MB [2022-12-20 11:46:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][920/1519] eta 0:10:02 lr 0.000007 time 0.9231 (1.0055) model_time 0.9229 (1.0043) loss 0.6735 (0.8198) grad_norm 8.9394 (8.8790/2.2732) mem 68106MB [2022-12-20 11:46:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][930/1519] eta 0:09:52 lr 0.000007 time 0.9333 (1.0055) model_time 0.9331 (1.0043) loss 0.7524 (0.8199) grad_norm 11.5028 (8.8897/2.2685) mem 68106MB [2022-12-20 11:46:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][940/1519] eta 0:09:42 lr 0.000007 time 0.9345 (1.0055) model_time 0.9343 (1.0043) loss 0.7632 (0.8195) grad_norm 8.4945 (8.8693/2.2648) mem 68106MB [2022-12-20 11:47:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][950/1519] eta 0:09:32 lr 0.000007 time 0.9212 (1.0054) model_time 0.9211 (1.0043) loss 0.8336 (0.8197) grad_norm 6.3955 (8.8527/2.2565) mem 68106MB [2022-12-20 11:47:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][960/1519] eta 0:09:21 lr 0.000007 time 0.9378 (1.0053) model_time 0.9377 (1.0042) loss 0.7422 (0.8196) grad_norm 9.4766 (8.8534/2.2598) mem 68106MB [2022-12-20 11:47:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][970/1519] eta 0:09:11 lr 0.000007 time 1.0108 (1.0054) model_time 1.0107 (1.0042) loss 0.6643 (0.8191) grad_norm 6.7605 (8.8656/2.2680) mem 68106MB [2022-12-20 11:47:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][980/1519] eta 0:09:01 lr 0.000007 time 0.9220 (1.0054) model_time 0.9219 (1.0042) loss 0.8716 (0.8192) grad_norm 6.4546 (8.8563/2.2651) mem 68106MB [2022-12-20 11:47:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][990/1519] eta 0:08:51 lr 0.000007 time 0.9297 (1.0054) model_time 0.9295 (1.0042) loss 0.8551 (0.8192) grad_norm 9.2552 (8.8403/2.2634) mem 68106MB [2022-12-20 11:47:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1000/1519] eta 0:08:41 lr 0.000007 time 0.9248 (1.0053) model_time 0.9247 (1.0042) loss 0.8802 (0.8193) grad_norm 7.8562 (8.8348/2.2711) mem 68106MB [2022-12-20 11:48:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1010/1519] eta 0:08:31 lr 0.000007 time 0.9272 (1.0053) model_time 0.9270 (1.0042) loss 0.9273 (0.8200) grad_norm 6.2655 (8.8363/2.2702) mem 68106MB [2022-12-20 11:48:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1020/1519] eta 0:08:21 lr 0.000007 time 0.9251 (1.0052) model_time 0.9249 (1.0041) loss 0.8400 (0.8201) grad_norm 8.5872 (8.8067/2.2676) mem 68106MB [2022-12-20 11:48:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1030/1519] eta 0:08:11 lr 0.000007 time 0.9072 (1.0052) model_time 0.9070 (1.0041) loss 1.2133 (0.8204) grad_norm 10.3874 (8.8314/2.2865) mem 68106MB [2022-12-20 11:48:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1040/1519] eta 0:08:01 lr 0.000007 time 0.9242 (1.0052) model_time 0.9241 (1.0041) loss 0.8201 (0.8216) grad_norm 6.9099 (8.8119/2.2811) mem 68106MB [2022-12-20 11:48:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1050/1519] eta 0:07:51 lr 0.000007 time 0.9588 (1.0052) model_time 0.9586 (1.0041) loss 0.8802 (0.8224) grad_norm 10.5519 (8.8013/2.2781) mem 68106MB [2022-12-20 11:48:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1060/1519] eta 0:07:41 lr 0.000007 time 0.9219 (1.0052) model_time 0.9218 (1.0041) loss 1.1672 (0.8228) grad_norm 6.8276 (8.7801/2.2329) mem 68106MB [2022-12-20 11:49:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1070/1519] eta 0:07:31 lr 0.000007 time 0.9208 (1.0051) model_time 0.9207 (1.0040) loss 0.6573 (0.8233) grad_norm 8.0605 (8.7750/2.2243) mem 68106MB [2022-12-20 11:49:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1080/1519] eta 0:07:21 lr 0.000007 time 0.9495 (1.0051) model_time 0.9494 (1.0041) loss 0.7881 (0.8232) grad_norm 9.8025 (8.7037/2.0653) mem 68106MB [2022-12-20 11:49:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1090/1519] eta 0:07:11 lr 0.000007 time 0.9511 (1.0053) model_time 0.9510 (1.0043) loss 0.7174 (0.8229) grad_norm 6.1166 (8.7060/2.0629) mem 68106MB [2022-12-20 11:49:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1100/1519] eta 0:07:01 lr 0.000007 time 1.0271 (1.0054) model_time 1.0270 (1.0043) loss 0.6787 (0.8230) grad_norm 7.4908 (8.6701/2.0559) mem 68106MB [2022-12-20 11:49:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1110/1519] eta 0:06:51 lr 0.000007 time 0.9249 (1.0054) model_time 0.9248 (1.0043) loss 0.6698 (0.8226) grad_norm 7.0559 (8.6616/2.0652) mem 68106MB [2022-12-20 11:49:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1120/1519] eta 0:06:41 lr 0.000007 time 0.9227 (1.0054) model_time 0.9226 (1.0044) loss 0.7941 (0.8222) grad_norm 10.6929 (8.6480/2.0730) mem 68106MB [2022-12-20 11:50:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1130/1519] eta 0:06:31 lr 0.000006 time 0.9295 (1.0056) model_time 0.9293 (1.0045) loss 0.7655 (0.8216) grad_norm 7.8552 (8.6184/2.0644) mem 68106MB [2022-12-20 11:50:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1140/1519] eta 0:06:21 lr 0.000006 time 0.9214 (1.0055) model_time 0.9212 (1.0044) loss 0.8245 (0.8215) grad_norm 10.9864 (8.6186/2.0633) mem 68106MB [2022-12-20 11:50:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1150/1519] eta 0:06:11 lr 0.000006 time 0.9284 (1.0057) model_time 0.9283 (1.0046) loss 0.8545 (0.8212) grad_norm 9.9898 (8.6655/2.0668) mem 68106MB [2022-12-20 11:50:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1160/1519] eta 0:06:01 lr 0.000006 time 0.9241 (1.0057) model_time 0.9240 (1.0047) loss 0.9309 (0.8216) grad_norm 8.9357 (8.6613/2.0501) mem 68106MB [2022-12-20 11:50:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1170/1519] eta 0:05:50 lr 0.000006 time 0.9206 (1.0057) model_time 0.9204 (1.0047) loss 0.8807 (0.8219) grad_norm 7.4121 (8.6166/1.9750) mem 68106MB [2022-12-20 11:50:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1180/1519] eta 0:05:40 lr 0.000006 time 0.9442 (1.0057) model_time 0.9441 (1.0047) loss 0.9441 (0.8216) grad_norm 8.6147 (8.6151/1.9558) mem 68106MB [2022-12-20 11:51:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1190/1519] eta 0:05:30 lr 0.000006 time 0.9189 (1.0057) model_time 0.9188 (1.0046) loss 0.7087 (0.8216) grad_norm 6.7745 (8.6025/1.9561) mem 68106MB [2022-12-20 11:51:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1200/1519] eta 0:05:20 lr 0.000006 time 0.9190 (1.0056) model_time 0.9189 (1.0046) loss 0.7262 (0.8215) grad_norm 7.4361 (8.6073/1.9720) mem 68106MB [2022-12-20 11:51:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1210/1519] eta 0:05:10 lr 0.000006 time 0.9212 (1.0055) model_time 0.9211 (1.0045) loss 0.7793 (0.8213) grad_norm 6.9684 (8.5619/1.8453) mem 68106MB [2022-12-20 11:51:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1220/1519] eta 0:05:00 lr 0.000006 time 0.9213 (1.0055) model_time 0.9212 (1.0045) loss 0.9655 (0.8211) grad_norm 7.9180 (8.5385/1.8362) mem 68106MB [2022-12-20 11:51:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1230/1519] eta 0:04:50 lr 0.000006 time 0.9226 (1.0055) model_time 0.9224 (1.0045) loss 0.8956 (0.8213) grad_norm 8.3179 (8.5632/1.8349) mem 68106MB [2022-12-20 11:51:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1240/1519] eta 0:04:40 lr 0.000006 time 0.9245 (1.0054) model_time 0.9244 (1.0045) loss 0.8914 (0.8216) grad_norm 6.1088 (8.5749/1.8429) mem 68106MB [2022-12-20 11:52:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1250/1519] eta 0:04:30 lr 0.000006 time 0.9228 (1.0054) model_time 0.9226 (1.0044) loss 0.6737 (0.8216) grad_norm 9.0751 (8.5485/1.8280) mem 68106MB [2022-12-20 11:52:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1260/1519] eta 0:04:20 lr 0.000006 time 0.9298 (1.0053) model_time 0.9297 (1.0044) loss 0.6672 (0.8215) grad_norm 11.5940 (8.5586/1.8467) mem 68106MB [2022-12-20 11:52:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1270/1519] eta 0:04:10 lr 0.000006 time 0.9250 (1.0053) model_time 0.9248 (1.0043) loss 0.8909 (0.8209) grad_norm 10.5979 (8.5722/1.8506) mem 68106MB [2022-12-20 11:52:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1280/1519] eta 0:04:00 lr 0.000006 time 0.9893 (1.0054) model_time 0.9892 (1.0044) loss 0.7540 (0.8204) grad_norm 7.7949 (8.5645/1.8642) mem 68106MB [2022-12-20 11:52:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1290/1519] eta 0:03:50 lr 0.000006 time 0.9214 (1.0054) model_time 0.9213 (1.0045) loss 0.8513 (0.8205) grad_norm 6.7881 (8.5979/2.1327) mem 68106MB [2022-12-20 11:53:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1300/1519] eta 0:03:40 lr 0.000006 time 0.9237 (1.0055) model_time 0.9236 (1.0046) loss 0.6865 (0.8203) grad_norm 8.1508 (8.5605/2.1027) mem 68106MB [2022-12-20 11:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1310/1519] eta 0:03:30 lr 0.000006 time 0.9216 (1.0055) model_time 0.9215 (1.0046) loss 0.7136 (0.8197) grad_norm 7.5052 (8.5611/2.1064) mem 68106MB [2022-12-20 11:53:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1320/1519] eta 0:03:20 lr 0.000006 time 0.9202 (1.0054) model_time 0.9201 (1.0045) loss 0.9375 (0.8194) grad_norm 6.9171 (8.5600/2.1064) mem 68106MB [2022-12-20 11:53:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1330/1519] eta 0:03:10 lr 0.000006 time 0.9264 (1.0054) model_time 0.9263 (1.0044) loss 0.6635 (0.8193) grad_norm 7.6914 (8.5795/2.1357) mem 68106MB [2022-12-20 11:53:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1340/1519] eta 0:02:59 lr 0.000006 time 0.9210 (1.0053) model_time 0.9209 (1.0044) loss 0.7231 (0.8193) grad_norm 8.9915 (8.5563/2.0701) mem 68106MB [2022-12-20 11:53:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1350/1519] eta 0:02:49 lr 0.000006 time 0.9341 (1.0053) model_time 0.9339 (1.0044) loss 0.9205 (0.8190) grad_norm 7.7003 (8.5572/2.0657) mem 68106MB [2022-12-20 11:54:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1360/1519] eta 0:02:39 lr 0.000006 time 0.9335 (1.0053) model_time 0.9333 (1.0044) loss 0.8899 (0.8196) grad_norm 9.7962 (8.5725/2.0619) mem 68106MB [2022-12-20 11:54:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1370/1519] eta 0:02:29 lr 0.000006 time 0.9194 (1.0053) model_time 0.9192 (1.0044) loss 0.7231 (0.8193) grad_norm 6.8929 (8.5517/2.0574) mem 68106MB [2022-12-20 11:54:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1380/1519] eta 0:02:19 lr 0.000006 time 0.9289 (1.0053) model_time 0.9288 (1.0044) loss 0.7543 (0.8189) grad_norm 6.8702 (8.5653/2.0601) mem 68106MB [2022-12-20 11:54:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1390/1519] eta 0:02:09 lr 0.000006 time 0.9365 (1.0053) model_time 0.9364 (1.0044) loss 0.8220 (0.8195) grad_norm 11.0403 (8.5983/2.0736) mem 68106MB [2022-12-20 11:54:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1400/1519] eta 0:01:59 lr 0.000006 time 0.9433 (1.0053) model_time 0.9431 (1.0044) loss 0.7263 (0.8195) grad_norm 6.5416 (8.5729/2.0701) mem 68106MB [2022-12-20 11:54:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1410/1519] eta 0:01:49 lr 0.000006 time 0.9215 (1.0053) model_time 0.9213 (1.0044) loss 0.7953 (0.8197) grad_norm 6.7844 (8.5879/2.0629) mem 68106MB [2022-12-20 11:55:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1420/1519] eta 0:01:39 lr 0.000006 time 0.9651 (1.0053) model_time 0.9650 (1.0044) loss 1.1492 (0.8198) grad_norm 8.0023 (8.5619/2.0483) mem 68106MB [2022-12-20 11:55:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1430/1519] eta 0:01:29 lr 0.000006 time 0.9328 (1.0053) model_time 0.9326 (1.0044) loss 0.7614 (0.8198) grad_norm 7.4982 (8.5610/2.0892) mem 68106MB [2022-12-20 11:55:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1440/1519] eta 0:01:19 lr 0.000006 time 0.9228 (1.0053) model_time 0.9227 (1.0044) loss 0.9079 (0.8201) grad_norm 8.5297 (8.5902/2.0926) mem 68106MB [2022-12-20 11:55:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1450/1519] eta 0:01:09 lr 0.000006 time 0.9320 (1.0052) model_time 0.9318 (1.0044) loss 0.6908 (0.8202) grad_norm 7.6163 (8.5838/2.0922) mem 68106MB [2022-12-20 11:55:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1460/1519] eta 0:00:59 lr 0.000006 time 0.9216 (1.0052) model_time 0.9215 (1.0043) loss 0.6806 (0.8199) grad_norm 7.1828 (8.5961/2.0959) mem 68106MB [2022-12-20 11:55:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1470/1519] eta 0:00:49 lr 0.000006 time 0.9350 (1.0053) model_time 0.9348 (1.0045) loss 0.7611 (0.8198) grad_norm 8.1019 (8.5868/2.0821) mem 68106MB [2022-12-20 11:56:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1480/1519] eta 0:00:39 lr 0.000006 time 0.9194 (1.0053) model_time 0.9192 (1.0044) loss 0.8905 (0.8201) grad_norm 7.4765 (8.5739/2.0842) mem 68106MB [2022-12-20 11:56:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1490/1519] eta 0:00:29 lr 0.000006 time 0.9206 (1.0054) model_time 0.9205 (1.0045) loss 0.6871 (0.8205) grad_norm 7.5733 (8.5689/2.0713) mem 68106MB [2022-12-20 11:56:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1500/1519] eta 0:00:19 lr 0.000006 time 0.9290 (1.0054) model_time 0.9288 (1.0045) loss 0.9142 (0.8205) grad_norm 9.0380 (8.5766/2.0703) mem 68106MB [2022-12-20 11:56:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [71/100][1510/1519] eta 0:00:09 lr 0.000006 time 0.9243 (1.0054) model_time 0.9242 (1.0045) loss 0.6774 (0.8201) grad_norm 8.9675 (8.5748/2.0804) mem 68106MB [2022-12-20 11:56:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 71 training takes 0:25:27 [2022-12-20 11:56:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_71.pth saving...... [2022-12-20 11:57:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_71.pth saved !!! [2022-12-20 11:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.604 (0.604) Loss 0.5235 (0.5235) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 11:57:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.325) Loss 0.5310 (0.5042) Acc@1 92.361 (92.961) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 11:57:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.312) Loss 0.4825 (0.5002) Acc@1 91.319 (92.758) Acc@5 98.958 (98.330) Mem 68106MB [2022-12-20 11:57:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.299 (0.307) Loss 0.6291 (0.5067) Acc@1 89.236 (92.440) Acc@5 97.569 (98.320) Mem 68106MB [2022-12-20 11:57:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.306) Loss 0.4437 (0.4975) Acc@1 94.097 (92.539) Acc@5 99.306 (98.433) Mem 68106MB [2022-12-20 11:57:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.305) Loss 0.4855 (0.4949) Acc@1 92.361 (92.633) Acc@5 99.653 (98.509) Mem 68106MB [2022-12-20 11:57:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.5113 (0.4944) Acc@1 89.931 (92.538) Acc@5 98.264 (98.497) Mem 68106MB [2022-12-20 11:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5414 (0.4954) Acc@1 93.056 (92.493) Acc@5 98.264 (98.499) Mem 68106MB [2022-12-20 11:57:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.302 (0.303) Loss 0.4357 (0.4935) Acc@1 93.056 (92.554) Acc@5 98.264 (98.538) Mem 68106MB [2022-12-20 11:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:71] * Acc@1 92.510 Acc@5 98.543 [2022-12-20 11:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 11:57:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 11:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 11:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 11:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][0/1519] eta 0:37:04 lr 0.000006 time 1.4647 (1.4647) model_time 0.9838 (0.9838) loss 0.9218 (0.9218) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 11:58:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][10/1519] eta 0:26:19 lr 0.000006 time 0.9347 (1.0470) model_time 0.9346 (1.0029) loss 0.8435 (0.8352) grad_norm 12.0292 (9.2089/1.5666) mem 68106MB [2022-12-20 11:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][20/1519] eta 0:25:43 lr 0.000006 time 0.9251 (1.0296) model_time 0.9250 (1.0064) loss 0.6868 (0.8520) grad_norm 8.1769 (8.8958/1.6554) mem 68106MB [2022-12-20 11:58:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][30/1519] eta 0:25:19 lr 0.000006 time 0.9299 (1.0203) model_time 0.9297 (1.0045) loss 0.7101 (0.8614) grad_norm 7.0666 (8.6202/1.4914) mem 68106MB [2022-12-20 11:58:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][40/1519] eta 0:25:04 lr 0.000006 time 0.9873 (1.0172) model_time 0.9872 (1.0052) loss 0.9437 (0.8413) grad_norm 8.8993 (8.8766/1.6767) mem 68106MB [2022-12-20 11:58:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][50/1519] eta 0:24:48 lr 0.000006 time 0.9225 (1.0134) model_time 0.9224 (1.0037) loss 0.7392 (0.8370) grad_norm 8.2853 (8.7758/1.6220) mem 68106MB [2022-12-20 11:58:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][60/1519] eta 0:24:36 lr 0.000006 time 0.9219 (1.0121) model_time 0.9218 (1.0039) loss 0.9160 (0.8527) grad_norm 7.7149 (8.4982/1.6129) mem 68106MB [2022-12-20 11:59:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][70/1519] eta 0:24:23 lr 0.000006 time 0.9252 (1.0103) model_time 0.9251 (1.0032) loss 0.7843 (0.8446) grad_norm 6.4866 (8.4778/1.6263) mem 68106MB [2022-12-20 11:59:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][80/1519] eta 0:24:15 lr 0.000006 time 0.9232 (1.0113) model_time 0.9231 (1.0051) loss 0.6736 (0.8363) grad_norm 7.4074 (8.5081/1.5935) mem 68106MB [2022-12-20 11:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][90/1519] eta 0:24:06 lr 0.000006 time 1.0085 (1.0123) model_time 1.0083 (1.0067) loss 0.8764 (0.8332) grad_norm 7.2870 (8.5817/1.5507) mem 68106MB [2022-12-20 11:59:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][100/1519] eta 0:23:55 lr 0.000006 time 0.9253 (1.0117) model_time 0.9251 (1.0067) loss 0.9759 (0.8350) grad_norm 7.8118 (8.5452/1.5599) mem 68106MB [2022-12-20 11:59:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][110/1519] eta 0:23:43 lr 0.000006 time 0.9272 (1.0106) model_time 0.9271 (1.0060) loss 0.6624 (0.8264) grad_norm 9.4976 (8.5433/1.5252) mem 68106MB [2022-12-20 11:59:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][120/1519] eta 0:23:33 lr 0.000006 time 0.9250 (1.0101) model_time 0.9249 (1.0058) loss 0.7312 (0.8273) grad_norm 7.2408 (8.5895/1.5095) mem 68106MB [2022-12-20 12:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][130/1519] eta 0:23:21 lr 0.000006 time 0.9248 (1.0091) model_time 0.9246 (1.0051) loss 0.7080 (0.8222) grad_norm 7.0379 (8.5901/1.6858) mem 68106MB [2022-12-20 12:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][140/1519] eta 0:23:10 lr 0.000006 time 0.9253 (1.0086) model_time 0.9250 (1.0049) loss 1.0379 (0.8189) grad_norm 11.3126 (8.6109/1.6779) mem 68106MB [2022-12-20 12:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][150/1519] eta 0:23:01 lr 0.000006 time 0.9660 (1.0092) model_time 0.9658 (1.0057) loss 0.8322 (0.8216) grad_norm 6.3015 (8.5537/1.7005) mem 68106MB [2022-12-20 12:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][160/1519] eta 0:22:50 lr 0.000006 time 0.9669 (1.0087) model_time 0.9668 (1.0054) loss 0.9030 (0.8178) grad_norm 7.3637 (8.5378/1.6947) mem 68106MB [2022-12-20 12:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][170/1519] eta 0:22:39 lr 0.000006 time 0.9230 (1.0080) model_time 0.9229 (1.0049) loss 0.6866 (0.8162) grad_norm 9.7819 (8.5454/1.7003) mem 68106MB [2022-12-20 12:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][180/1519] eta 0:22:29 lr 0.000006 time 0.9287 (1.0077) model_time 0.9286 (1.0048) loss 0.6675 (0.8203) grad_norm 12.0088 (8.5536/1.7051) mem 68106MB [2022-12-20 12:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][190/1519] eta 0:22:18 lr 0.000006 time 0.9359 (1.0074) model_time 0.9357 (1.0046) loss 0.6950 (0.8243) grad_norm 9.9191 (8.5928/1.7446) mem 68106MB [2022-12-20 12:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][200/1519] eta 0:22:08 lr 0.000006 time 0.9241 (1.0072) model_time 0.9239 (1.0045) loss 1.1292 (0.8271) grad_norm 11.8286 (8.6533/1.7489) mem 68106MB [2022-12-20 12:01:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][210/1519] eta 0:21:58 lr 0.000006 time 0.9279 (1.0074) model_time 0.9277 (1.0048) loss 0.8918 (0.8271) grad_norm 8.3933 (8.6151/1.7399) mem 68106MB [2022-12-20 12:01:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][220/1519] eta 0:21:48 lr 0.000006 time 0.9282 (1.0070) model_time 0.9280 (1.0045) loss 0.8315 (0.8269) grad_norm 7.9649 (8.6132/1.7081) mem 68106MB [2022-12-20 12:01:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][230/1519] eta 0:21:37 lr 0.000006 time 0.9203 (1.0066) model_time 0.9202 (1.0042) loss 1.0441 (0.8275) grad_norm 12.9471 (8.6539/1.7514) mem 68106MB [2022-12-20 12:01:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][240/1519] eta 0:21:27 lr 0.000006 time 0.9260 (1.0063) model_time 0.9259 (1.0040) loss 0.9261 (0.8270) grad_norm 10.5344 (8.7188/1.7954) mem 68106MB [2022-12-20 12:02:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][250/1519] eta 0:21:16 lr 0.000006 time 0.9311 (1.0060) model_time 0.9309 (1.0038) loss 0.6579 (0.8282) grad_norm 7.8850 (8.7002/1.8232) mem 68106MB [2022-12-20 12:02:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][260/1519] eta 0:21:06 lr 0.000006 time 0.9241 (1.0060) model_time 0.9240 (1.0038) loss 0.8956 (0.8293) grad_norm 7.3279 (8.6718/1.8307) mem 68106MB [2022-12-20 12:02:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][270/1519] eta 0:20:56 lr 0.000006 time 0.9860 (1.0060) model_time 0.9858 (1.0039) loss 0.6692 (0.8259) grad_norm 9.5071 (8.6715/1.8049) mem 68106MB [2022-12-20 12:02:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][280/1519] eta 0:20:46 lr 0.000006 time 0.9281 (1.0057) model_time 0.9279 (1.0037) loss 0.6546 (0.8257) grad_norm 7.3244 (8.6378/1.8151) mem 68106MB [2022-12-20 12:02:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][290/1519] eta 0:20:35 lr 0.000006 time 0.9269 (1.0054) model_time 0.9268 (1.0035) loss 0.8192 (0.8246) grad_norm 6.9240 (8.6394/1.8052) mem 68106MB [2022-12-20 12:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][300/1519] eta 0:20:25 lr 0.000006 time 0.9255 (1.0052) model_time 0.9254 (1.0033) loss 0.8465 (0.8269) grad_norm 12.9578 (8.7215/1.8789) mem 68106MB [2022-12-20 12:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][310/1519] eta 0:20:14 lr 0.000006 time 0.9429 (1.0050) model_time 0.9427 (1.0031) loss 0.8486 (0.8260) grad_norm 9.1961 (8.7175/1.8543) mem 68106MB [2022-12-20 12:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][320/1519] eta 0:20:04 lr 0.000006 time 0.9343 (1.0048) model_time 0.9342 (1.0030) loss 0.8637 (0.8273) grad_norm 10.6581 (8.7360/1.8613) mem 68106MB [2022-12-20 12:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][330/1519] eta 0:19:54 lr 0.000006 time 0.9769 (1.0050) model_time 0.9768 (1.0032) loss 0.8880 (0.8276) grad_norm 9.6698 (8.7386/1.8455) mem 68106MB [2022-12-20 12:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][340/1519] eta 0:19:44 lr 0.000006 time 0.9199 (1.0051) model_time 0.9198 (1.0033) loss 0.9431 (0.8266) grad_norm 12.6793 (8.8197/1.9451) mem 68106MB [2022-12-20 12:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][350/1519] eta 0:19:34 lr 0.000006 time 0.9253 (1.0049) model_time 0.9247 (1.0032) loss 0.7058 (0.8277) grad_norm 7.7486 (8.8062/1.9452) mem 68106MB [2022-12-20 12:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][360/1519] eta 0:19:24 lr 0.000006 time 0.9219 (1.0050) model_time 0.9218 (1.0033) loss 0.6961 (0.8281) grad_norm 8.0645 (8.7982/1.9331) mem 68106MB [2022-12-20 12:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][370/1519] eta 0:19:15 lr 0.000006 time 1.1526 (1.0054) model_time 1.1524 (1.0038) loss 0.8047 (0.8271) grad_norm 9.9841 (8.7850/1.9171) mem 68106MB [2022-12-20 12:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][380/1519] eta 0:19:05 lr 0.000006 time 0.9260 (1.0053) model_time 0.9258 (1.0037) loss 0.7362 (0.8271) grad_norm 14.4207 (8.8604/2.0316) mem 68106MB [2022-12-20 12:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][390/1519] eta 0:18:55 lr 0.000006 time 0.9188 (1.0057) model_time 0.9186 (1.0042) loss 0.9093 (0.8285) grad_norm 8.9888 (8.8718/2.0444) mem 68106MB [2022-12-20 12:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][400/1519] eta 0:18:45 lr 0.000006 time 0.9235 (1.0057) model_time 0.9234 (1.0042) loss 0.8296 (0.8275) grad_norm 8.3432 (8.8633/2.0353) mem 68106MB [2022-12-20 12:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][410/1519] eta 0:18:35 lr 0.000006 time 0.9400 (1.0060) model_time 0.9399 (1.0045) loss 0.8106 (0.8259) grad_norm 4.8167 (8.8223/2.0409) mem 68106MB [2022-12-20 12:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][420/1519] eta 0:18:25 lr 0.000006 time 0.9316 (1.0058) model_time 0.9314 (1.0043) loss 0.8409 (0.8259) grad_norm 10.5024 (8.8216/2.0298) mem 68106MB [2022-12-20 12:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][430/1519] eta 0:18:15 lr 0.000006 time 0.9287 (1.0057) model_time 0.9286 (1.0043) loss 0.6635 (0.8274) grad_norm 8.6209 (8.7847/2.0245) mem 68106MB [2022-12-20 12:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][440/1519] eta 0:18:04 lr 0.000006 time 0.9239 (1.0056) model_time 0.9238 (1.0041) loss 0.6638 (0.8266) grad_norm 6.4796 (8.7558/2.0170) mem 68106MB [2022-12-20 12:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][450/1519] eta 0:17:54 lr 0.000006 time 0.9254 (1.0054) model_time 0.9252 (1.0040) loss 0.7255 (0.8265) grad_norm 11.2633 (8.8173/2.0980) mem 68106MB [2022-12-20 12:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][460/1519] eta 0:17:44 lr 0.000006 time 0.9284 (1.0053) model_time 0.9283 (1.0040) loss 0.8411 (0.8250) grad_norm 7.0092 (8.7806/2.0914) mem 68106MB [2022-12-20 12:05:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][470/1519] eta 0:17:34 lr 0.000006 time 0.9270 (1.0055) model_time 0.9269 (1.0041) loss 1.1353 (0.8253) grad_norm 7.4053 (8.7652/2.0761) mem 68106MB [2022-12-20 12:05:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][480/1519] eta 0:17:24 lr 0.000006 time 0.9268 (1.0054) model_time 0.9267 (1.0041) loss 0.7398 (0.8249) grad_norm 7.3024 (8.7409/2.0642) mem 68106MB [2022-12-20 12:06:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][490/1519] eta 0:17:14 lr 0.000006 time 0.9253 (1.0054) model_time 0.9251 (1.0041) loss 0.6706 (0.8224) grad_norm 8.1781 (8.7801/2.1217) mem 68106MB [2022-12-20 12:06:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][500/1519] eta 0:17:04 lr 0.000006 time 0.9280 (1.0053) model_time 0.9279 (1.0040) loss 0.7995 (0.8221) grad_norm 7.8131 (8.7775/2.1088) mem 68106MB [2022-12-20 12:06:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][510/1519] eta 0:16:54 lr 0.000006 time 0.9279 (1.0053) model_time 0.9278 (1.0040) loss 0.6586 (0.8204) grad_norm 10.1279 (8.7893/2.1040) mem 68106MB [2022-12-20 12:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][520/1519] eta 0:16:44 lr 0.000006 time 0.9306 (1.0056) model_time 0.9305 (1.0044) loss 0.7888 (0.8197) grad_norm 7.6419 (8.7675/2.0910) mem 68106MB [2022-12-20 12:06:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][530/1519] eta 0:16:34 lr 0.000006 time 0.9225 (1.0055) model_time 0.9223 (1.0043) loss 0.8199 (0.8205) grad_norm 8.4858 (8.7532/2.0772) mem 68106MB [2022-12-20 12:06:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][540/1519] eta 0:16:24 lr 0.000006 time 0.9240 (1.0054) model_time 0.9238 (1.0042) loss 0.8180 (0.8191) grad_norm 6.8111 (8.7309/2.0790) mem 68106MB [2022-12-20 12:07:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][550/1519] eta 0:16:14 lr 0.000006 time 1.0619 (1.0055) model_time 1.0618 (1.0043) loss 0.6907 (0.8176) grad_norm 7.4428 (8.7348/2.0734) mem 68106MB [2022-12-20 12:07:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][560/1519] eta 0:16:04 lr 0.000006 time 0.9297 (1.0054) model_time 0.9296 (1.0042) loss 0.6970 (0.8172) grad_norm 8.8412 (8.7143/2.0630) mem 68106MB [2022-12-20 12:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][570/1519] eta 0:15:54 lr 0.000006 time 0.9238 (1.0056) model_time 0.9236 (1.0044) loss 0.6703 (0.8175) grad_norm 7.4406 (8.7282/2.0751) mem 68106MB [2022-12-20 12:07:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][580/1519] eta 0:15:44 lr 0.000006 time 0.9351 (1.0056) model_time 0.9349 (1.0044) loss 1.1808 (0.8166) grad_norm 11.7310 (8.7360/2.0658) mem 68106MB [2022-12-20 12:07:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][590/1519] eta 0:15:34 lr 0.000006 time 0.9339 (1.0057) model_time 0.9338 (1.0046) loss 0.9779 (0.8182) grad_norm 9.4045 (8.7315/2.0625) mem 68106MB [2022-12-20 12:07:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][600/1519] eta 0:15:24 lr 0.000006 time 0.9262 (1.0056) model_time 0.9260 (1.0045) loss 0.6967 (0.8187) grad_norm 7.9147 (8.7189/2.0507) mem 68106MB [2022-12-20 12:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][610/1519] eta 0:15:14 lr 0.000006 time 0.9351 (1.0056) model_time 0.9349 (1.0044) loss 0.6688 (0.8184) grad_norm 7.7499 (8.7107/2.0910) mem 68106MB [2022-12-20 12:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][620/1519] eta 0:15:03 lr 0.000006 time 0.9289 (1.0054) model_time 0.9288 (1.0043) loss 0.7531 (0.8186) grad_norm 7.6140 (8.7135/2.0846) mem 68106MB [2022-12-20 12:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][630/1519] eta 0:14:53 lr 0.000006 time 0.9357 (1.0053) model_time 0.9355 (1.0042) loss 0.6667 (0.8188) grad_norm 7.9338 (8.7131/2.0858) mem 68106MB [2022-12-20 12:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][640/1519] eta 0:14:43 lr 0.000006 time 0.9287 (1.0052) model_time 0.9285 (1.0041) loss 0.7232 (0.8179) grad_norm 8.0234 (8.6941/2.0754) mem 68106MB [2022-12-20 12:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][650/1519] eta 0:14:33 lr 0.000006 time 0.8875 (1.0054) model_time 0.8874 (1.0044) loss 0.8359 (0.8182) grad_norm 6.8262 (8.7168/2.0942) mem 68106MB [2022-12-20 12:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][660/1519] eta 0:14:23 lr 0.000006 time 0.9743 (1.0054) model_time 0.9741 (1.0044) loss 0.6815 (0.8169) grad_norm 8.8254 (8.7461/2.1107) mem 68106MB [2022-12-20 12:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][670/1519] eta 0:14:13 lr 0.000006 time 0.8940 (1.0054) model_time 0.8939 (1.0044) loss 0.9687 (0.8160) grad_norm 7.1063 (8.7149/2.1200) mem 68106MB [2022-12-20 12:09:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][680/1519] eta 0:14:03 lr 0.000006 time 0.9270 (1.0053) model_time 0.9269 (1.0043) loss 0.8948 (0.8158) grad_norm 10.6622 (8.7087/2.1250) mem 68106MB [2022-12-20 12:09:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][690/1519] eta 0:13:53 lr 0.000006 time 0.9281 (1.0052) model_time 0.9279 (1.0042) loss 0.6843 (0.8161) grad_norm 8.0906 (8.7061/2.1874) mem 68106MB [2022-12-20 12:09:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][700/1519] eta 0:13:43 lr 0.000006 time 1.0064 (1.0053) model_time 1.0063 (1.0042) loss 0.7041 (0.8152) grad_norm 8.7314 (8.6986/2.1858) mem 68106MB [2022-12-20 12:09:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][710/1519] eta 0:13:33 lr 0.000006 time 0.9271 (1.0052) model_time 0.9270 (1.0042) loss 0.9138 (0.8155) grad_norm 6.1408 (8.6912/2.1908) mem 68106MB [2022-12-20 12:09:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][720/1519] eta 0:13:23 lr 0.000006 time 0.9210 (1.0052) model_time 0.9208 (1.0042) loss 0.8974 (0.8151) grad_norm 10.3195 (8.6863/2.1937) mem 68106MB [2022-12-20 12:10:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][730/1519] eta 0:13:12 lr 0.000006 time 0.9272 (1.0051) model_time 0.9271 (1.0041) loss 0.8163 (0.8146) grad_norm 6.3845 (8.6716/2.1632) mem 68106MB [2022-12-20 12:10:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][740/1519] eta 0:13:02 lr 0.000006 time 0.9292 (1.0051) model_time 0.9290 (1.0041) loss 0.7101 (0.8142) grad_norm 12.6987 (8.6903/2.1729) mem 68106MB [2022-12-20 12:10:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][750/1519] eta 0:12:52 lr 0.000006 time 0.9336 (1.0052) model_time 0.9334 (1.0042) loss 0.7037 (0.8134) grad_norm 8.1865 (8.7047/2.1668) mem 68106MB [2022-12-20 12:10:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][760/1519] eta 0:12:42 lr 0.000006 time 0.9332 (1.0051) model_time 0.9330 (1.0041) loss 1.1232 (0.8137) grad_norm 8.7661 (8.7094/2.1664) mem 68106MB [2022-12-20 12:10:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][770/1519] eta 0:12:32 lr 0.000006 time 0.9296 (1.0050) model_time 0.9295 (1.0041) loss 0.7160 (0.8132) grad_norm 9.1552 (8.7105/2.1608) mem 68106MB [2022-12-20 12:10:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][780/1519] eta 0:12:22 lr 0.000006 time 0.9251 (1.0051) model_time 0.9249 (1.0041) loss 0.7089 (0.8121) grad_norm 7.1739 (8.7007/2.1544) mem 68106MB [2022-12-20 12:11:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][790/1519] eta 0:12:12 lr 0.000006 time 0.9210 (1.0051) model_time 0.9208 (1.0041) loss 0.7016 (0.8112) grad_norm 8.9966 (8.6960/2.1420) mem 68106MB [2022-12-20 12:11:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][800/1519] eta 0:12:02 lr 0.000006 time 0.9219 (1.0051) model_time 0.9218 (1.0041) loss 0.7371 (0.8113) grad_norm 6.9575 (8.7046/2.2445) mem 68106MB [2022-12-20 12:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][810/1519] eta 0:11:52 lr 0.000006 time 0.9217 (1.0050) model_time 0.9216 (1.0041) loss 0.6935 (0.8108) grad_norm 11.6883 (8.7336/2.2506) mem 68106MB [2022-12-20 12:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][820/1519] eta 0:11:42 lr 0.000006 time 0.9891 (1.0050) model_time 0.9890 (1.0041) loss 1.3089 (0.8114) grad_norm 8.0695 (8.7430/2.2751) mem 68106MB [2022-12-20 12:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][830/1519] eta 0:11:32 lr 0.000006 time 1.2151 (1.0054) model_time 1.2149 (1.0045) loss 1.2497 (0.8115) grad_norm 6.0273 (8.7024/2.2654) mem 68106MB [2022-12-20 12:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][840/1519] eta 0:11:22 lr 0.000006 time 0.9861 (1.0053) model_time 0.9859 (1.0044) loss 0.9452 (0.8116) grad_norm 8.9228 (8.6673/2.2482) mem 68106MB [2022-12-20 12:12:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][850/1519] eta 0:11:12 lr 0.000006 time 0.9335 (1.0053) model_time 0.9333 (1.0044) loss 1.0683 (0.8122) grad_norm 8.4296 (8.6617/2.2340) mem 68106MB [2022-12-20 12:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][860/1519] eta 0:11:02 lr 0.000006 time 0.9274 (1.0052) model_time 0.9273 (1.0043) loss 0.6624 (0.8128) grad_norm 7.8299 (8.6713/2.2395) mem 68106MB [2022-12-20 12:12:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][870/1519] eta 0:10:52 lr 0.000006 time 0.9322 (1.0052) model_time 0.9321 (1.0043) loss 0.6845 (0.8127) grad_norm 7.3316 (8.6596/2.2413) mem 68106MB [2022-12-20 12:12:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][880/1519] eta 0:10:42 lr 0.000006 time 1.0304 (1.0054) model_time 1.0302 (1.0046) loss 0.8499 (0.8122) grad_norm 11.6004 (8.7112/2.2444) mem 68106MB [2022-12-20 12:12:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][890/1519] eta 0:10:32 lr 0.000006 time 0.9341 (1.0054) model_time 0.9339 (1.0045) loss 0.7713 (0.8125) grad_norm 7.6478 (8.6964/2.2408) mem 68106MB [2022-12-20 12:13:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][900/1519] eta 0:10:22 lr 0.000006 time 0.9385 (1.0055) model_time 0.9383 (1.0046) loss 0.6983 (0.8127) grad_norm 9.1196 (8.6465/2.2072) mem 68106MB [2022-12-20 12:13:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][910/1519] eta 0:10:12 lr 0.000006 time 0.9275 (1.0054) model_time 0.9274 (1.0046) loss 0.8157 (0.8124) grad_norm 5.7230 (8.6525/2.2424) mem 68106MB [2022-12-20 12:13:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][920/1519] eta 0:10:02 lr 0.000006 time 0.9280 (1.0054) model_time 0.9279 (1.0046) loss 0.8309 (0.8120) grad_norm 5.9664 (8.6112/2.2411) mem 68106MB [2022-12-20 12:13:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][930/1519] eta 0:09:52 lr 0.000006 time 0.9351 (1.0054) model_time 0.9346 (1.0045) loss 0.8551 (0.8127) grad_norm 10.0800 (8.5947/2.2426) mem 68106MB [2022-12-20 12:13:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][940/1519] eta 0:09:42 lr 0.000006 time 0.9325 (1.0053) model_time 0.9324 (1.0044) loss 0.8522 (0.8124) grad_norm 9.3429 (8.5286/2.1888) mem 68106MB [2022-12-20 12:13:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][950/1519] eta 0:09:32 lr 0.000006 time 0.9395 (1.0053) model_time 0.9394 (1.0045) loss 1.1569 (0.8122) grad_norm 11.3239 (8.5256/2.1869) mem 68106MB [2022-12-20 12:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][960/1519] eta 0:09:21 lr 0.000006 time 0.9336 (1.0053) model_time 0.9334 (1.0045) loss 0.8341 (0.8120) grad_norm 6.1845 (8.5063/2.1865) mem 68106MB [2022-12-20 12:14:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][970/1519] eta 0:09:11 lr 0.000006 time 0.9397 (1.0053) model_time 0.9396 (1.0045) loss 1.1163 (0.8124) grad_norm 8.4612 (8.4889/2.1928) mem 68106MB [2022-12-20 12:14:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][980/1519] eta 0:09:01 lr 0.000006 time 0.9383 (1.0054) model_time 0.9381 (1.0045) loss 1.1380 (0.8131) grad_norm 9.7552 (8.4531/2.1085) mem 68106MB [2022-12-20 12:14:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][990/1519] eta 0:08:51 lr 0.000006 time 0.9289 (1.0053) model_time 0.9288 (1.0045) loss 0.9999 (0.8132) grad_norm 8.2334 (8.4364/2.0907) mem 68106MB [2022-12-20 12:14:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1000/1519] eta 0:08:41 lr 0.000006 time 0.9389 (1.0053) model_time 0.9388 (1.0045) loss 0.6699 (0.8130) grad_norm 9.0636 (8.4176/2.0915) mem 68106MB [2022-12-20 12:14:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1010/1519] eta 0:08:31 lr 0.000006 time 0.9301 (1.0056) model_time 0.9300 (1.0047) loss 0.7242 (0.8130) grad_norm 8.1607 (8.4313/2.0807) mem 68106MB [2022-12-20 12:15:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1020/1519] eta 0:08:21 lr 0.000006 time 0.9360 (1.0056) model_time 0.9359 (1.0048) loss 0.9086 (0.8141) grad_norm 9.8726 (8.4283/2.0782) mem 68106MB [2022-12-20 12:15:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1030/1519] eta 0:08:11 lr 0.000006 time 0.9397 (1.0056) model_time 0.9396 (1.0048) loss 0.9948 (0.8143) grad_norm 8.3404 (8.4283/2.0778) mem 68106MB [2022-12-20 12:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1040/1519] eta 0:08:01 lr 0.000006 time 0.9312 (1.0055) model_time 0.9310 (1.0047) loss 0.7039 (0.8142) grad_norm 7.4879 (8.4718/2.1169) mem 68106MB [2022-12-20 12:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1050/1519] eta 0:07:51 lr 0.000006 time 0.9308 (1.0055) model_time 0.9306 (1.0047) loss 0.7079 (0.8146) grad_norm 9.3162 (8.4224/2.0412) mem 68106MB [2022-12-20 12:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1060/1519] eta 0:07:41 lr 0.000006 time 0.9292 (1.0054) model_time 0.9290 (1.0046) loss 1.0291 (0.8146) grad_norm 9.7107 (8.4508/2.0460) mem 68106MB [2022-12-20 12:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1070/1519] eta 0:07:31 lr 0.000006 time 0.9491 (1.0054) model_time 0.9489 (1.0046) loss 0.8553 (0.8140) grad_norm 10.4647 (8.4692/2.0476) mem 68106MB [2022-12-20 12:16:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1080/1519] eta 0:07:21 lr 0.000006 time 0.9245 (1.0053) model_time 0.9244 (1.0046) loss 0.8108 (0.8146) grad_norm 10.3582 (8.4859/2.0503) mem 68106MB [2022-12-20 12:16:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1090/1519] eta 0:07:11 lr 0.000006 time 0.9313 (1.0054) model_time 0.9310 (1.0046) loss 0.7782 (0.8142) grad_norm 9.8617 (8.4744/1.9999) mem 68106MB [2022-12-20 12:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1100/1519] eta 0:07:01 lr 0.000006 time 0.9326 (1.0054) model_time 0.9324 (1.0047) loss 0.8434 (0.8145) grad_norm 8.6839 (8.4557/2.0001) mem 68106MB [2022-12-20 12:16:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1110/1519] eta 0:06:51 lr 0.000006 time 0.9314 (1.0054) model_time 0.9313 (1.0046) loss 0.6741 (0.8140) grad_norm 6.6089 (8.4221/1.9919) mem 68106MB [2022-12-20 12:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1120/1519] eta 0:06:41 lr 0.000006 time 0.9211 (1.0053) model_time 0.9209 (1.0046) loss 0.9705 (0.8142) grad_norm 6.9503 (8.4241/1.9944) mem 68106MB [2022-12-20 12:16:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1130/1519] eta 0:06:31 lr 0.000006 time 0.9251 (1.0053) model_time 0.9249 (1.0045) loss 0.6623 (0.8138) grad_norm 6.2937 (8.4368/2.0145) mem 68106MB [2022-12-20 12:17:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1140/1519] eta 0:06:21 lr 0.000006 time 0.9298 (1.0053) model_time 0.9297 (1.0046) loss 0.7411 (0.8130) grad_norm 9.7428 (8.4648/2.0050) mem 68106MB [2022-12-20 12:17:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1150/1519] eta 0:06:10 lr 0.000006 time 0.9277 (1.0053) model_time 0.9275 (1.0046) loss 0.6666 (0.8128) grad_norm 8.8228 (8.4538/1.9942) mem 68106MB [2022-12-20 12:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1160/1519] eta 0:06:00 lr 0.000006 time 0.9236 (1.0054) model_time 0.9234 (1.0046) loss 0.6764 (0.8134) grad_norm 6.4638 (8.4582/1.9936) mem 68106MB [2022-12-20 12:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1170/1519] eta 0:05:50 lr 0.000006 time 0.9327 (1.0053) model_time 0.9326 (1.0045) loss 0.7185 (0.8132) grad_norm 8.5312 (8.4519/1.9836) mem 68106MB [2022-12-20 12:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1180/1519] eta 0:05:40 lr 0.000006 time 0.9280 (1.0053) model_time 0.9278 (1.0046) loss 1.0448 (0.8133) grad_norm 10.0044 (8.4307/1.9806) mem 68106MB [2022-12-20 12:17:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1190/1519] eta 0:05:30 lr 0.000006 time 0.9553 (1.0053) model_time 0.9552 (1.0046) loss 0.6662 (0.8133) grad_norm 6.2225 (8.4110/1.9754) mem 68106MB [2022-12-20 12:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1200/1519] eta 0:05:20 lr 0.000006 time 0.9381 (1.0056) model_time 0.9376 (1.0049) loss 0.6607 (0.8126) grad_norm 10.9090 (8.4189/1.9813) mem 68106MB [2022-12-20 12:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1210/1519] eta 0:05:10 lr 0.000006 time 0.9602 (1.0058) model_time 0.9600 (1.0051) loss 0.7903 (0.8125) grad_norm 7.7504 (8.4248/1.9783) mem 68106MB [2022-12-20 12:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1220/1519] eta 0:05:00 lr 0.000006 time 0.9305 (1.0058) model_time 0.9303 (1.0050) loss 0.6705 (0.8123) grad_norm 8.6516 (8.4147/1.9746) mem 68106MB [2022-12-20 12:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1230/1519] eta 0:04:50 lr 0.000006 time 0.9141 (1.0058) model_time 0.9139 (1.0051) loss 0.7818 (0.8124) grad_norm 5.9576 (8.4173/1.9829) mem 68106MB [2022-12-20 12:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1240/1519] eta 0:04:40 lr 0.000006 time 0.9280 (1.0058) model_time 0.9278 (1.0051) loss 0.6885 (0.8120) grad_norm 11.1308 (8.4477/2.0001) mem 68106MB [2022-12-20 12:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1250/1519] eta 0:04:30 lr 0.000006 time 0.9281 (1.0058) model_time 0.9280 (1.0050) loss 0.6704 (0.8120) grad_norm 8.0737 (8.4042/1.9792) mem 68106MB [2022-12-20 12:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1260/1519] eta 0:04:20 lr 0.000006 time 0.9088 (1.0058) model_time 0.9086 (1.0050) loss 0.8106 (0.8114) grad_norm 8.2073 (8.3912/1.9567) mem 68106MB [2022-12-20 12:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1270/1519] eta 0:04:10 lr 0.000006 time 0.9591 (1.0058) model_time 0.9590 (1.0051) loss 0.7417 (0.8116) grad_norm 8.7483 (8.4049/1.9486) mem 68106MB [2022-12-20 12:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1280/1519] eta 0:04:00 lr 0.000006 time 0.9330 (1.0058) model_time 0.9328 (1.0051) loss 0.7819 (0.8116) grad_norm 6.4430 (8.4071/1.9511) mem 68106MB [2022-12-20 12:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1290/1519] eta 0:03:50 lr 0.000006 time 0.9063 (1.0057) model_time 0.9062 (1.0050) loss 0.6900 (0.8114) grad_norm inf (8.4125/1.8881) mem 68106MB [2022-12-20 12:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1300/1519] eta 0:03:40 lr 0.000006 time 0.9277 (1.0057) model_time 0.9275 (1.0050) loss 0.6713 (0.8114) grad_norm 8.7636 (8.4301/1.8833) mem 68106MB [2022-12-20 12:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1310/1519] eta 0:03:30 lr 0.000006 time 0.9283 (1.0056) model_time 0.9281 (1.0049) loss 0.9827 (0.8112) grad_norm 8.3325 (8.4501/1.9497) mem 68106MB [2022-12-20 12:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1320/1519] eta 0:03:20 lr 0.000006 time 0.9333 (1.0056) model_time 0.9332 (1.0049) loss 0.8385 (0.8111) grad_norm 8.1622 (8.4449/1.9472) mem 68106MB [2022-12-20 12:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1330/1519] eta 0:03:10 lr 0.000006 time 0.9276 (1.0056) model_time 0.9275 (1.0049) loss 0.9915 (0.8109) grad_norm 8.4204 (8.4510/1.9508) mem 68106MB [2022-12-20 12:20:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1340/1519] eta 0:02:59 lr 0.000006 time 0.9442 (1.0055) model_time 0.9441 (1.0048) loss 0.8432 (0.8111) grad_norm 11.1669 (8.5109/2.0769) mem 68106MB [2022-12-20 12:20:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1350/1519] eta 0:02:49 lr 0.000006 time 0.9283 (1.0055) model_time 0.9282 (1.0048) loss 0.6799 (0.8114) grad_norm 7.6778 (8.4882/2.0540) mem 68106MB [2022-12-20 12:20:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1360/1519] eta 0:02:39 lr 0.000006 time 0.9374 (1.0055) model_time 0.9372 (1.0048) loss 0.9426 (0.8112) grad_norm 6.7874 (8.5030/2.0602) mem 68106MB [2022-12-20 12:20:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1370/1519] eta 0:02:29 lr 0.000006 time 0.9359 (1.0054) model_time 0.9358 (1.0047) loss 0.6770 (0.8110) grad_norm 7.5247 (8.4956/2.0572) mem 68106MB [2022-12-20 12:21:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1380/1519] eta 0:02:19 lr 0.000006 time 0.9322 (1.0055) model_time 0.9320 (1.0048) loss 0.9605 (0.8107) grad_norm 6.5209 (8.4732/2.0618) mem 68106MB [2022-12-20 12:21:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1390/1519] eta 0:02:09 lr 0.000006 time 0.9332 (1.0054) model_time 0.9330 (1.0047) loss 0.6750 (0.8104) grad_norm 11.5558 (8.4871/2.0730) mem 68106MB [2022-12-20 12:21:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1400/1519] eta 0:01:59 lr 0.000006 time 0.9339 (1.0054) model_time 0.9337 (1.0047) loss 0.6623 (0.8101) grad_norm 6.5900 (8.4490/1.9503) mem 68106MB [2022-12-20 12:21:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1410/1519] eta 0:01:49 lr 0.000006 time 0.9338 (1.0054) model_time 0.9337 (1.0047) loss 0.9108 (0.8101) grad_norm 7.5692 (8.4346/1.9409) mem 68106MB [2022-12-20 12:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1420/1519] eta 0:01:39 lr 0.000006 time 0.9380 (1.0053) model_time 0.9379 (1.0047) loss 0.6801 (0.8107) grad_norm 7.0361 (8.3973/1.9026) mem 68106MB [2022-12-20 12:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1430/1519] eta 0:01:29 lr 0.000006 time 0.9357 (1.0053) model_time 0.9356 (1.0046) loss 0.9509 (0.8111) grad_norm 9.8584 (8.4387/1.9465) mem 68106MB [2022-12-20 12:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1440/1519] eta 0:01:19 lr 0.000006 time 0.9352 (1.0053) model_time 0.9350 (1.0046) loss 0.6554 (0.8117) grad_norm 15.7831 (8.4745/1.9819) mem 68106MB [2022-12-20 12:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1450/1519] eta 0:01:09 lr 0.000006 time 0.9268 (1.0053) model_time 0.9266 (1.0046) loss 1.2006 (0.8117) grad_norm 7.2248 (8.4926/1.9925) mem 68106MB [2022-12-20 12:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1460/1519] eta 0:00:59 lr 0.000006 time 0.9330 (1.0054) model_time 0.9329 (1.0047) loss 1.0456 (0.8118) grad_norm 8.4412 (8.4885/1.9758) mem 68106MB [2022-12-20 12:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1470/1519] eta 0:00:49 lr 0.000006 time 0.9245 (1.0054) model_time 0.9244 (1.0047) loss 0.7431 (0.8117) grad_norm 25.0047 (8.5781/2.2452) mem 68106MB [2022-12-20 12:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1480/1519] eta 0:00:39 lr 0.000006 time 0.9376 (1.0054) model_time 0.9374 (1.0048) loss 0.6893 (0.8116) grad_norm 8.1091 (8.5430/2.2319) mem 68106MB [2022-12-20 12:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1490/1519] eta 0:00:29 lr 0.000006 time 0.9310 (1.0056) model_time 0.9308 (1.0049) loss 0.7652 (0.8120) grad_norm 9.0611 (8.5417/2.2346) mem 68106MB [2022-12-20 12:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1500/1519] eta 0:00:19 lr 0.000006 time 0.9295 (1.0055) model_time 0.9293 (1.0048) loss 0.9616 (0.8119) grad_norm 7.8391 (8.5509/2.2314) mem 68106MB [2022-12-20 12:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [72/100][1510/1519] eta 0:00:09 lr 0.000006 time 0.9291 (1.0056) model_time 0.9290 (1.0049) loss 1.0879 (0.8118) grad_norm 8.1364 (8.5233/2.2024) mem 68106MB [2022-12-20 12:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 72 training takes 0:25:27 [2022-12-20 12:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_72.pth saving...... [2022-12-20 12:23:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_72.pth saved !!! [2022-12-20 12:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.698 (0.698) Loss 0.5255 (0.5255) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 12:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.293 (0.334) Loss 0.5339 (0.5011) Acc@1 92.014 (92.393) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 12:23:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.320) Loss 0.4838 (0.4975) Acc@1 92.014 (92.576) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 12:23:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.303 (0.313) Loss 0.6271 (0.5040) Acc@1 90.278 (92.283) Acc@5 97.917 (98.410) Mem 68106MB [2022-12-20 12:23:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.310) Loss 0.4446 (0.4943) Acc@1 93.750 (92.395) Acc@5 99.306 (98.501) Mem 68106MB [2022-12-20 12:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.308) Loss 0.4773 (0.4915) Acc@1 92.014 (92.490) Acc@5 99.653 (98.543) Mem 68106MB [2022-12-20 12:24:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.306) Loss 0.5164 (0.4916) Acc@1 90.625 (92.458) Acc@5 97.917 (98.526) Mem 68106MB [2022-12-20 12:24:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.306) Loss 0.5392 (0.4928) Acc@1 92.361 (92.390) Acc@5 98.264 (98.543) Mem 68106MB [2022-12-20 12:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.304 (0.304) Loss 0.4197 (0.4910) Acc@1 93.056 (92.425) Acc@5 98.958 (98.590) Mem 68106MB [2022-12-20 12:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:72] * Acc@1 92.391 Acc@5 98.592 [2022-12-20 12:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 12:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 12:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][0/1519] eta 0:45:54 lr 0.000006 time 1.8132 (1.8132) model_time 1.0863 (1.0863) loss 0.8661 (0.8661) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 12:24:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][10/1519] eta 0:27:16 lr 0.000006 time 0.9319 (1.0843) model_time 0.9318 (1.0177) loss 1.0047 (0.8597) grad_norm 7.2299 (7.5207/1.0479) mem 68106MB [2022-12-20 12:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][20/1519] eta 0:26:05 lr 0.000006 time 0.9352 (1.0442) model_time 0.9350 (1.0092) loss 0.6520 (0.8253) grad_norm 8.8313 (7.9881/1.4499) mem 68106MB [2022-12-20 12:24:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][30/1519] eta 0:25:34 lr 0.000006 time 0.9330 (1.0303) model_time 0.9329 (1.0064) loss 0.6803 (0.8207) grad_norm 6.7607 (8.2763/1.5320) mem 68106MB [2022-12-20 12:24:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][40/1519] eta 0:25:13 lr 0.000006 time 0.9451 (1.0231) model_time 0.9450 (1.0049) loss 0.9211 (0.8164) grad_norm 7.9761 (8.5061/1.7149) mem 68106MB [2022-12-20 12:25:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][50/1519] eta 0:24:56 lr 0.000006 time 0.9399 (1.0190) model_time 0.9398 (1.0044) loss 0.7434 (0.8235) grad_norm 9.0343 (8.8955/2.0296) mem 68106MB [2022-12-20 12:25:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][60/1519] eta 0:24:45 lr 0.000006 time 1.0242 (1.0180) model_time 1.0240 (1.0057) loss 1.2084 (0.8275) grad_norm 8.8342 (8.9267/2.0045) mem 68106MB [2022-12-20 12:25:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][70/1519] eta 0:24:32 lr 0.000006 time 0.9349 (1.0160) model_time 0.9348 (1.0054) loss 0.7515 (0.8302) grad_norm 7.7728 (8.9225/1.9176) mem 68106MB [2022-12-20 12:25:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][80/1519] eta 0:24:20 lr 0.000006 time 0.9323 (1.0149) model_time 0.9321 (1.0055) loss 0.9193 (0.8340) grad_norm 7.9858 (8.7899/1.8530) mem 68106MB [2022-12-20 12:25:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][90/1519] eta 0:24:09 lr 0.000006 time 0.9312 (1.0142) model_time 0.9310 (1.0058) loss 0.6718 (0.8293) grad_norm 6.6307 (8.6513/1.8532) mem 68106MB [2022-12-20 12:25:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][100/1519] eta 0:23:58 lr 0.000006 time 0.9328 (1.0135) model_time 0.9327 (1.0059) loss 0.6806 (0.8260) grad_norm 9.4625 (8.5556/1.8644) mem 68106MB [2022-12-20 12:26:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][110/1519] eta 0:23:46 lr 0.000006 time 0.9340 (1.0127) model_time 0.9339 (1.0058) loss 0.7826 (0.8254) grad_norm 7.8565 (8.4706/1.8440) mem 68106MB [2022-12-20 12:26:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][120/1519] eta 0:23:35 lr 0.000006 time 0.9334 (1.0119) model_time 0.9333 (1.0056) loss 0.7253 (0.8207) grad_norm 8.5959 (8.5477/1.7963) mem 68106MB [2022-12-20 12:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][130/1519] eta 0:23:24 lr 0.000006 time 0.9288 (1.0110) model_time 0.9287 (1.0051) loss 0.7435 (0.8181) grad_norm 11.1178 (8.6036/1.8370) mem 68106MB [2022-12-20 12:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][140/1519] eta 0:23:12 lr 0.000006 time 0.9357 (1.0101) model_time 0.9355 (1.0046) loss 0.7112 (0.8193) grad_norm 8.3042 (8.6549/1.8242) mem 68106MB [2022-12-20 12:26:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][150/1519] eta 0:23:02 lr 0.000006 time 0.9911 (1.0097) model_time 0.9910 (1.0046) loss 0.6768 (0.8173) grad_norm 11.7281 (8.6291/1.8394) mem 68106MB [2022-12-20 12:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][160/1519] eta 0:22:51 lr 0.000006 time 0.9326 (1.0092) model_time 0.9324 (1.0043) loss 0.8536 (0.8122) grad_norm 8.2930 (8.5667/1.8470) mem 68106MB [2022-12-20 12:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][170/1519] eta 0:22:40 lr 0.000006 time 0.9271 (1.0084) model_time 0.9269 (1.0038) loss 0.8239 (0.8134) grad_norm 9.9292 (8.5515/1.8158) mem 68106MB [2022-12-20 12:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][180/1519] eta 0:22:30 lr 0.000006 time 0.9788 (1.0086) model_time 0.9787 (1.0043) loss 0.7089 (0.8128) grad_norm 5.7503 (8.4474/1.8277) mem 68106MB [2022-12-20 12:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][190/1519] eta 0:22:20 lr 0.000006 time 0.9274 (1.0083) model_time 0.9273 (1.0041) loss 0.6926 (0.8110) grad_norm 12.2423 (8.4837/1.8268) mem 68106MB [2022-12-20 12:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][200/1519] eta 0:22:09 lr 0.000006 time 0.9393 (1.0080) model_time 0.9392 (1.0041) loss 0.6858 (0.8103) grad_norm 9.0277 (8.5312/1.8137) mem 68106MB [2022-12-20 12:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][210/1519] eta 0:21:59 lr 0.000006 time 0.9386 (1.0079) model_time 0.9385 (1.0041) loss 0.7729 (0.8089) grad_norm 7.6166 (8.4875/1.7908) mem 68106MB [2022-12-20 12:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][220/1519] eta 0:21:48 lr 0.000006 time 0.9328 (1.0075) model_time 0.9327 (1.0038) loss 0.9292 (0.8108) grad_norm 8.6278 (8.5267/1.7954) mem 68106MB [2022-12-20 12:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][230/1519] eta 0:21:38 lr 0.000006 time 0.9192 (1.0072) model_time 0.9190 (1.0037) loss 0.7554 (0.8120) grad_norm 9.0738 (8.4754/1.7832) mem 68106MB [2022-12-20 12:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][240/1519] eta 0:21:28 lr 0.000006 time 0.9978 (1.0074) model_time 0.9977 (1.0041) loss 0.8237 (0.8133) grad_norm 11.8784 (8.5354/1.8007) mem 68106MB [2022-12-20 12:28:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][250/1519] eta 0:21:18 lr 0.000006 time 0.9293 (1.0073) model_time 0.9292 (1.0041) loss 0.7340 (0.8115) grad_norm 8.1646 (8.5023/1.7881) mem 68106MB [2022-12-20 12:28:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][260/1519] eta 0:21:08 lr 0.000006 time 0.9391 (1.0074) model_time 0.9389 (1.0042) loss 0.8659 (0.8149) grad_norm 9.0374 (8.4960/1.7619) mem 68106MB [2022-12-20 12:28:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][270/1519] eta 0:20:59 lr 0.000006 time 0.9334 (1.0082) model_time 0.9332 (1.0052) loss 0.6996 (0.8128) grad_norm 6.5256 (8.4987/1.7464) mem 68106MB [2022-12-20 12:28:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][280/1519] eta 0:20:48 lr 0.000006 time 0.9345 (1.0080) model_time 0.9344 (1.0050) loss 0.6638 (0.8099) grad_norm 5.9491 (8.4455/1.7523) mem 68106MB [2022-12-20 12:29:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][290/1519] eta 0:20:38 lr 0.000006 time 0.9362 (1.0081) model_time 0.9361 (1.0052) loss 0.7116 (0.8083) grad_norm 8.4788 (8.4149/1.7489) mem 68106MB [2022-12-20 12:29:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][300/1519] eta 0:20:29 lr 0.000006 time 0.9288 (1.0082) model_time 0.9287 (1.0055) loss 0.7803 (0.8062) grad_norm 8.1076 (8.3876/1.7287) mem 68106MB [2022-12-20 12:29:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][310/1519] eta 0:20:18 lr 0.000006 time 0.9356 (1.0081) model_time 0.9355 (1.0054) loss 0.7459 (0.8050) grad_norm 8.2880 (8.4038/1.7195) mem 68106MB [2022-12-20 12:29:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][320/1519] eta 0:20:08 lr 0.000006 time 0.9280 (1.0082) model_time 0.9279 (1.0055) loss 0.6805 (0.8043) grad_norm 7.9469 (8.3907/1.6983) mem 68106MB [2022-12-20 12:29:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][330/1519] eta 0:19:58 lr 0.000006 time 0.9414 (1.0080) model_time 0.9412 (1.0054) loss 1.1123 (0.8047) grad_norm 8.9164 (8.4223/1.7241) mem 68106MB [2022-12-20 12:29:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][340/1519] eta 0:19:48 lr 0.000006 time 0.9367 (1.0078) model_time 0.9364 (1.0053) loss 0.7328 (0.8057) grad_norm 8.4033 (8.4228/1.7093) mem 68106MB [2022-12-20 12:30:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][350/1519] eta 0:19:37 lr 0.000006 time 0.9364 (1.0076) model_time 0.9363 (1.0052) loss 0.8689 (0.8062) grad_norm 7.3234 (8.4019/1.6934) mem 68106MB [2022-12-20 12:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][360/1519] eta 0:19:27 lr 0.000006 time 0.9270 (1.0074) model_time 0.9267 (1.0050) loss 0.7644 (0.8056) grad_norm 7.6532 (8.3959/1.6766) mem 68106MB [2022-12-20 12:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][370/1519] eta 0:19:17 lr 0.000006 time 0.9287 (1.0072) model_time 0.9285 (1.0048) loss 1.0316 (0.8052) grad_norm 11.5076 (8.4116/1.6720) mem 68106MB [2022-12-20 12:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][380/1519] eta 0:19:07 lr 0.000006 time 0.9389 (1.0072) model_time 0.9388 (1.0049) loss 0.8005 (0.8060) grad_norm 9.0231 (8.4371/1.6644) mem 68106MB [2022-12-20 12:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][390/1519] eta 0:18:56 lr 0.000006 time 0.9319 (1.0071) model_time 0.9317 (1.0048) loss 0.6599 (0.8055) grad_norm 7.2855 (8.4042/1.6683) mem 68106MB [2022-12-20 12:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][400/1519] eta 0:18:46 lr 0.000006 time 0.9318 (1.0069) model_time 0.9317 (1.0047) loss 1.1575 (0.8080) grad_norm 8.2839 (8.3973/1.6783) mem 68106MB [2022-12-20 12:31:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][410/1519] eta 0:18:36 lr 0.000006 time 0.9335 (1.0069) model_time 0.9334 (1.0048) loss 1.0701 (0.8096) grad_norm 9.4801 (8.3961/1.6739) mem 68106MB [2022-12-20 12:31:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][420/1519] eta 0:18:26 lr 0.000006 time 1.0269 (1.0070) model_time 1.0267 (1.0049) loss 0.9118 (0.8100) grad_norm 12.9600 (8.4438/1.7381) mem 68106MB [2022-12-20 12:31:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][430/1519] eta 0:18:16 lr 0.000006 time 0.9313 (1.0069) model_time 0.9311 (1.0048) loss 0.6704 (0.8086) grad_norm 6.4758 (8.4894/1.8106) mem 68106MB [2022-12-20 12:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][440/1519] eta 0:18:06 lr 0.000006 time 0.9307 (1.0067) model_time 0.9305 (1.0046) loss 1.0258 (0.8104) grad_norm 5.9297 (8.4761/1.8016) mem 68106MB [2022-12-20 12:31:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][450/1519] eta 0:17:55 lr 0.000006 time 0.9298 (1.0065) model_time 0.9297 (1.0045) loss 0.8954 (0.8116) grad_norm 12.2580 (8.4937/1.8324) mem 68106MB [2022-12-20 12:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][460/1519] eta 0:17:45 lr 0.000006 time 0.9354 (1.0063) model_time 0.9352 (1.0044) loss 1.0829 (0.8114) grad_norm 8.7299 (8.4784/1.8170) mem 68106MB [2022-12-20 12:32:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][470/1519] eta 0:17:35 lr 0.000006 time 0.9393 (1.0063) model_time 0.9391 (1.0044) loss 0.6829 (0.8101) grad_norm 9.9502 (8.4566/1.8153) mem 68106MB [2022-12-20 12:32:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][480/1519] eta 0:17:25 lr 0.000006 time 0.9243 (1.0063) model_time 0.9241 (1.0044) loss 0.8910 (0.8103) grad_norm 8.4652 (8.4836/1.8340) mem 68106MB [2022-12-20 12:32:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][490/1519] eta 0:17:15 lr 0.000006 time 0.9253 (1.0064) model_time 0.9251 (1.0045) loss 0.6735 (0.8093) grad_norm 6.8167 (8.4556/1.8349) mem 68106MB [2022-12-20 12:32:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][500/1519] eta 0:17:05 lr 0.000006 time 0.9598 (1.0069) model_time 0.9596 (1.0050) loss 0.8340 (0.8090) grad_norm 8.9406 (8.4524/1.8211) mem 68106MB [2022-12-20 12:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][510/1519] eta 0:16:55 lr 0.000006 time 0.9323 (1.0067) model_time 0.9322 (1.0049) loss 0.7009 (0.8086) grad_norm 8.0629 (8.4573/1.8211) mem 68106MB [2022-12-20 12:32:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][520/1519] eta 0:16:45 lr 0.000006 time 0.9363 (1.0067) model_time 0.9362 (1.0049) loss 0.9098 (0.8084) grad_norm 10.4040 (8.4622/1.8163) mem 68106MB [2022-12-20 12:33:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][530/1519] eta 0:16:35 lr 0.000006 time 0.9366 (1.0065) model_time 0.9364 (1.0048) loss 0.9085 (0.8074) grad_norm 8.1400 (8.4656/1.8013) mem 68106MB [2022-12-20 12:33:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][540/1519] eta 0:16:25 lr 0.000006 time 0.9341 (1.0064) model_time 0.9340 (1.0047) loss 0.8806 (0.8077) grad_norm 7.1270 (8.4592/1.8014) mem 68106MB [2022-12-20 12:33:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][550/1519] eta 0:16:15 lr 0.000006 time 0.9342 (1.0063) model_time 0.9341 (1.0046) loss 0.7999 (0.8083) grad_norm 8.9699 (8.4788/1.7945) mem 68106MB [2022-12-20 12:33:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][560/1519] eta 0:16:05 lr 0.000006 time 0.9323 (1.0066) model_time 0.9321 (1.0049) loss 0.6982 (0.8100) grad_norm 6.8696 (8.4844/1.8151) mem 68106MB [2022-12-20 12:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][570/1519] eta 0:15:55 lr 0.000006 time 1.1212 (1.0069) model_time 1.1211 (1.0053) loss 0.8337 (0.8103) grad_norm 8.1450 (8.5030/1.8101) mem 68106MB [2022-12-20 12:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][580/1519] eta 0:15:45 lr 0.000006 time 0.9277 (1.0069) model_time 0.9275 (1.0053) loss 0.8110 (0.8094) grad_norm 6.5130 (8.4838/1.8116) mem 68106MB [2022-12-20 12:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][590/1519] eta 0:15:35 lr 0.000006 time 0.9355 (1.0069) model_time 0.9354 (1.0053) loss 0.7445 (0.8079) grad_norm 6.5169 (8.4940/1.8294) mem 68106MB [2022-12-20 12:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][600/1519] eta 0:15:25 lr 0.000006 time 0.9355 (1.0070) model_time 0.9354 (1.0054) loss 0.6585 (0.8073) grad_norm 5.8464 (8.4907/1.8492) mem 68106MB [2022-12-20 12:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][610/1519] eta 0:15:15 lr 0.000006 time 0.9316 (1.0069) model_time 0.9315 (1.0054) loss 0.7745 (0.8068) grad_norm 8.1209 (8.4907/1.8457) mem 68106MB [2022-12-20 12:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][620/1519] eta 0:15:05 lr 0.000006 time 0.9387 (1.0068) model_time 0.9386 (1.0053) loss 0.6614 (0.8062) grad_norm 9.2042 (8.5015/1.8538) mem 68106MB [2022-12-20 12:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][630/1519] eta 0:14:55 lr 0.000006 time 0.9325 (1.0068) model_time 0.9324 (1.0053) loss 0.8842 (0.8055) grad_norm 6.6074 (8.4766/1.8510) mem 68106MB [2022-12-20 12:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][640/1519] eta 0:14:44 lr 0.000006 time 0.9286 (1.0067) model_time 0.9285 (1.0052) loss 1.0618 (0.8060) grad_norm 5.8789 (8.4414/1.8411) mem 68106MB [2022-12-20 12:35:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][650/1519] eta 0:14:34 lr 0.000006 time 0.9361 (1.0065) model_time 0.9360 (1.0050) loss 0.7290 (0.8050) grad_norm 7.8718 (8.4109/1.7979) mem 68106MB [2022-12-20 12:35:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][660/1519] eta 0:14:24 lr 0.000006 time 0.9471 (1.0065) model_time 0.9470 (1.0050) loss 1.0680 (0.8055) grad_norm 11.1647 (8.4135/1.7969) mem 68106MB [2022-12-20 12:35:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][670/1519] eta 0:14:14 lr 0.000006 time 0.9374 (1.0063) model_time 0.9373 (1.0049) loss 1.1528 (0.8055) grad_norm 7.5380 (8.4071/1.7923) mem 68106MB [2022-12-20 12:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][680/1519] eta 0:14:04 lr 0.000006 time 0.9643 (1.0064) model_time 0.9641 (1.0049) loss 1.0836 (0.8058) grad_norm 6.8260 (8.4193/1.8144) mem 68106MB [2022-12-20 12:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][690/1519] eta 0:13:54 lr 0.000006 time 0.9258 (1.0063) model_time 0.9257 (1.0048) loss 0.6588 (0.8050) grad_norm 8.6703 (8.4331/1.8124) mem 68106MB [2022-12-20 12:35:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][700/1519] eta 0:13:44 lr 0.000006 time 0.9370 (1.0062) model_time 0.9369 (1.0048) loss 0.7654 (0.8050) grad_norm 6.5511 (8.4133/1.8141) mem 68106MB [2022-12-20 12:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][710/1519] eta 0:13:33 lr 0.000006 time 0.9280 (1.0061) model_time 0.9278 (1.0047) loss 0.7549 (0.8059) grad_norm 7.5875 (8.4314/1.8285) mem 68106MB [2022-12-20 12:36:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][720/1519] eta 0:13:23 lr 0.000006 time 1.0098 (1.0061) model_time 1.0097 (1.0047) loss 0.6731 (0.8051) grad_norm 5.8827 (8.4028/1.8340) mem 68106MB [2022-12-20 12:36:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][730/1519] eta 0:13:13 lr 0.000006 time 0.9398 (1.0060) model_time 0.9397 (1.0047) loss 0.7182 (0.8058) grad_norm 5.2631 (8.3770/1.8302) mem 68106MB [2022-12-20 12:36:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][740/1519] eta 0:13:03 lr 0.000006 time 0.9373 (1.0062) model_time 0.9372 (1.0048) loss 0.7094 (0.8058) grad_norm 7.2031 (8.3560/1.8281) mem 68106MB [2022-12-20 12:36:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][750/1519] eta 0:12:53 lr 0.000006 time 0.9290 (1.0061) model_time 0.9288 (1.0048) loss 0.7131 (0.8066) grad_norm 9.4140 (8.3672/1.8174) mem 68106MB [2022-12-20 12:36:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][760/1519] eta 0:12:43 lr 0.000006 time 0.9312 (1.0060) model_time 0.9310 (1.0047) loss 0.7158 (0.8060) grad_norm 6.5433 (8.3722/1.8073) mem 68106MB [2022-12-20 12:37:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][770/1519] eta 0:12:33 lr 0.000006 time 0.9314 (1.0060) model_time 0.9312 (1.0047) loss 0.7064 (0.8058) grad_norm 6.9104 (8.3536/1.8075) mem 68106MB [2022-12-20 12:37:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][780/1519] eta 0:12:23 lr 0.000006 time 0.9304 (1.0060) model_time 0.9302 (1.0047) loss 0.6902 (0.8055) grad_norm 8.5101 (8.3853/1.7980) mem 68106MB [2022-12-20 12:37:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][790/1519] eta 0:12:13 lr 0.000006 time 0.9270 (1.0059) model_time 0.9268 (1.0046) loss 0.6806 (0.8073) grad_norm 11.0476 (8.3839/1.7947) mem 68106MB [2022-12-20 12:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][800/1519] eta 0:12:03 lr 0.000006 time 0.9298 (1.0061) model_time 0.9297 (1.0048) loss 0.9618 (0.8069) grad_norm 7.9622 (8.3490/1.7879) mem 68106MB [2022-12-20 12:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][810/1519] eta 0:11:53 lr 0.000006 time 0.9198 (1.0061) model_time 0.9197 (1.0048) loss 0.6725 (0.8070) grad_norm 9.1648 (8.3608/1.7887) mem 68106MB [2022-12-20 12:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][820/1519] eta 0:11:43 lr 0.000006 time 0.9390 (1.0060) model_time 0.9388 (1.0047) loss 0.9231 (0.8075) grad_norm 8.0465 (8.3716/1.8044) mem 68106MB [2022-12-20 12:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][830/1519] eta 0:11:33 lr 0.000006 time 0.9418 (1.0059) model_time 0.9417 (1.0047) loss 1.1801 (0.8085) grad_norm 8.5085 (8.3799/1.8012) mem 68106MB [2022-12-20 12:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][840/1519] eta 0:11:22 lr 0.000006 time 0.9197 (1.0059) model_time 0.9193 (1.0046) loss 0.9263 (0.8093) grad_norm 8.5860 (8.3941/1.8423) mem 68106MB [2022-12-20 12:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][850/1519] eta 0:11:12 lr 0.000006 time 0.9293 (1.0059) model_time 0.9291 (1.0046) loss 1.0495 (0.8092) grad_norm 9.6500 (8.4126/1.8356) mem 68106MB [2022-12-20 12:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][860/1519] eta 0:11:02 lr 0.000006 time 0.9329 (1.0058) model_time 0.9327 (1.0046) loss 0.8213 (0.8095) grad_norm 5.5988 (8.4211/1.8734) mem 68106MB [2022-12-20 12:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][870/1519] eta 0:10:52 lr 0.000006 time 0.9233 (1.0058) model_time 0.9231 (1.0046) loss 1.4441 (0.8111) grad_norm 7.6312 (8.4035/1.8773) mem 68106MB [2022-12-20 12:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][880/1519] eta 0:10:42 lr 0.000006 time 0.9302 (1.0058) model_time 0.9301 (1.0046) loss 0.8235 (0.8104) grad_norm 6.9057 (8.4248/1.8652) mem 68106MB [2022-12-20 12:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][890/1519] eta 0:10:32 lr 0.000006 time 0.9167 (1.0058) model_time 0.9166 (1.0046) loss 0.6856 (0.8111) grad_norm 8.1809 (8.4513/1.8640) mem 68106MB [2022-12-20 12:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][900/1519] eta 0:10:22 lr 0.000006 time 0.9172 (1.0060) model_time 0.9170 (1.0048) loss 0.7928 (0.8112) grad_norm 7.3454 (8.4910/1.9361) mem 68106MB [2022-12-20 12:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][910/1519] eta 0:10:12 lr 0.000006 time 1.2274 (1.0063) model_time 1.2273 (1.0051) loss 0.8043 (0.8107) grad_norm 15.7171 (8.5168/1.9862) mem 68106MB [2022-12-20 12:39:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][920/1519] eta 0:10:02 lr 0.000006 time 0.9689 (1.0064) model_time 0.9687 (1.0052) loss 0.7345 (0.8100) grad_norm 9.0515 (8.5291/1.9988) mem 68106MB [2022-12-20 12:39:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][930/1519] eta 0:09:52 lr 0.000006 time 0.9274 (1.0063) model_time 0.9273 (1.0051) loss 1.0833 (0.8108) grad_norm 7.9701 (8.5060/1.9770) mem 68106MB [2022-12-20 12:40:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][940/1519] eta 0:09:42 lr 0.000006 time 0.9904 (1.0063) model_time 0.9900 (1.0051) loss 0.7773 (0.8104) grad_norm 7.1944 (8.4964/1.9928) mem 68106MB [2022-12-20 12:40:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][950/1519] eta 0:09:32 lr 0.000006 time 0.9221 (1.0062) model_time 0.9220 (1.0050) loss 0.6986 (0.8099) grad_norm 7.1775 (8.5191/2.0033) mem 68106MB [2022-12-20 12:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][960/1519] eta 0:09:22 lr 0.000006 time 0.9268 (1.0061) model_time 0.9265 (1.0050) loss 0.9033 (0.8099) grad_norm 9.5254 (8.5342/2.0102) mem 68106MB [2022-12-20 12:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][970/1519] eta 0:09:12 lr 0.000006 time 0.9263 (1.0061) model_time 0.9262 (1.0049) loss 0.9482 (0.8106) grad_norm 9.3526 (8.5248/2.0040) mem 68106MB [2022-12-20 12:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][980/1519] eta 0:09:02 lr 0.000006 time 0.9388 (1.0060) model_time 0.9387 (1.0049) loss 0.8560 (0.8104) grad_norm 8.3872 (8.5155/2.0002) mem 68106MB [2022-12-20 12:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][990/1519] eta 0:08:52 lr 0.000006 time 0.9484 (1.0062) model_time 0.9483 (1.0051) loss 0.7460 (0.8104) grad_norm 6.6693 (8.5135/1.9963) mem 68106MB [2022-12-20 12:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1000/1519] eta 0:08:42 lr 0.000006 time 0.9172 (1.0061) model_time 0.9171 (1.0050) loss 0.6735 (0.8102) grad_norm 7.3233 (8.5378/2.0191) mem 68106MB [2022-12-20 12:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1010/1519] eta 0:08:32 lr 0.000006 time 0.9209 (1.0061) model_time 0.9208 (1.0050) loss 0.9544 (0.8102) grad_norm 6.6285 (8.5418/2.0171) mem 68106MB [2022-12-20 12:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1020/1519] eta 0:08:22 lr 0.000006 time 0.9296 (1.0061) model_time 0.9295 (1.0050) loss 0.7160 (0.8101) grad_norm 8.5105 (8.5207/1.9778) mem 68106MB [2022-12-20 12:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1030/1519] eta 0:08:11 lr 0.000006 time 0.9215 (1.0060) model_time 0.9213 (1.0049) loss 0.9225 (0.8098) grad_norm 7.4141 (8.4817/1.9231) mem 68106MB [2022-12-20 12:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1040/1519] eta 0:08:01 lr 0.000006 time 0.9294 (1.0060) model_time 0.9293 (1.0050) loss 0.7886 (0.8096) grad_norm 8.0926 (8.5100/1.9304) mem 68106MB [2022-12-20 12:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1050/1519] eta 0:07:51 lr 0.000006 time 0.9264 (1.0061) model_time 0.9263 (1.0050) loss 1.0187 (0.8092) grad_norm 6.7296 (8.5041/1.9230) mem 68106MB [2022-12-20 12:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1060/1519] eta 0:07:41 lr 0.000006 time 0.9336 (1.0060) model_time 0.9334 (1.0049) loss 0.6799 (0.8089) grad_norm 12.0596 (8.5247/1.9383) mem 68106MB [2022-12-20 12:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1070/1519] eta 0:07:31 lr 0.000006 time 0.9278 (1.0060) model_time 0.9277 (1.0050) loss 0.9284 (0.8091) grad_norm 7.3172 (8.5273/1.9353) mem 68106MB [2022-12-20 12:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1080/1519] eta 0:07:21 lr 0.000006 time 0.9233 (1.0060) model_time 0.9232 (1.0049) loss 0.7057 (0.8088) grad_norm 6.7242 (8.4911/1.9214) mem 68106MB [2022-12-20 12:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1090/1519] eta 0:07:11 lr 0.000006 time 1.0235 (1.0061) model_time 1.0233 (1.0050) loss 0.7384 (0.8089) grad_norm 7.6247 (8.5079/1.9143) mem 68106MB [2022-12-20 12:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1100/1519] eta 0:07:01 lr 0.000006 time 0.9297 (1.0060) model_time 0.9294 (1.0050) loss 0.6726 (0.8088) grad_norm 10.5969 (8.5315/1.9200) mem 68106MB [2022-12-20 12:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1110/1519] eta 0:06:51 lr 0.000006 time 1.0295 (1.0061) model_time 1.0293 (1.0050) loss 0.7786 (0.8088) grad_norm 8.5898 (8.5648/1.9368) mem 68106MB [2022-12-20 12:43:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1120/1519] eta 0:06:41 lr 0.000006 time 0.9892 (1.0062) model_time 0.9890 (1.0051) loss 1.0272 (0.8083) grad_norm 7.7126 (8.5468/1.9312) mem 68106MB [2022-12-20 12:43:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1130/1519] eta 0:06:31 lr 0.000006 time 0.9273 (1.0061) model_time 0.9272 (1.0051) loss 0.6715 (0.8085) grad_norm 8.3566 (8.5540/1.9335) mem 68106MB [2022-12-20 12:43:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1140/1519] eta 0:06:21 lr 0.000006 time 0.9761 (1.0061) model_time 0.9759 (1.0051) loss 0.9723 (0.8086) grad_norm 10.9377 (8.5729/1.9279) mem 68106MB [2022-12-20 12:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1150/1519] eta 0:06:11 lr 0.000006 time 0.9266 (1.0060) model_time 0.9265 (1.0050) loss 0.8463 (0.8084) grad_norm 8.6302 (8.5381/1.9295) mem 68106MB [2022-12-20 12:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1160/1519] eta 0:06:01 lr 0.000006 time 0.9361 (1.0060) model_time 0.9360 (1.0050) loss 0.9232 (0.8079) grad_norm 7.6465 (8.5274/1.9080) mem 68106MB [2022-12-20 12:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1170/1519] eta 0:05:51 lr 0.000006 time 0.9763 (1.0060) model_time 0.9761 (1.0050) loss 0.8399 (0.8078) grad_norm 12.3959 (8.5159/1.9212) mem 68106MB [2022-12-20 12:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1180/1519] eta 0:05:41 lr 0.000006 time 0.9231 (1.0060) model_time 0.9230 (1.0050) loss 0.8838 (0.8082) grad_norm 8.1932 (8.5500/1.9235) mem 68106MB [2022-12-20 12:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1190/1519] eta 0:05:30 lr 0.000006 time 0.9423 (1.0060) model_time 0.9422 (1.0050) loss 0.6789 (0.8079) grad_norm 6.8491 (8.5259/1.8985) mem 68106MB [2022-12-20 12:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1200/1519] eta 0:05:20 lr 0.000006 time 0.9729 (1.0060) model_time 0.9727 (1.0050) loss 0.7193 (0.8082) grad_norm 8.8087 (8.5454/1.8799) mem 68106MB [2022-12-20 12:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1210/1519] eta 0:05:10 lr 0.000006 time 0.9311 (1.0060) model_time 0.9309 (1.0050) loss 0.7168 (0.8075) grad_norm 7.6154 (8.5743/1.8839) mem 68106MB [2022-12-20 12:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1220/1519] eta 0:05:00 lr 0.000006 time 0.9325 (1.0060) model_time 0.9323 (1.0050) loss 0.7001 (0.8074) grad_norm 6.5170 (8.5630/1.8871) mem 68106MB [2022-12-20 12:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1230/1519] eta 0:04:50 lr 0.000006 time 0.9277 (1.0060) model_time 0.9273 (1.0050) loss 0.8962 (0.8074) grad_norm 8.0931 (8.5751/1.8941) mem 68106MB [2022-12-20 12:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1240/1519] eta 0:04:40 lr 0.000006 time 0.9290 (1.0060) model_time 0.9289 (1.0050) loss 0.6479 (0.8069) grad_norm 7.7821 (8.6389/1.9602) mem 68106MB [2022-12-20 12:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1250/1519] eta 0:04:30 lr 0.000006 time 0.9227 (1.0059) model_time 0.9226 (1.0049) loss 0.8366 (0.8070) grad_norm 7.8569 (8.6344/1.9618) mem 68106MB [2022-12-20 12:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1260/1519] eta 0:04:20 lr 0.000006 time 0.9383 (1.0058) model_time 0.9381 (1.0049) loss 0.8129 (0.8073) grad_norm 6.6674 (8.6088/1.9617) mem 68106MB [2022-12-20 12:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1270/1519] eta 0:04:10 lr 0.000006 time 0.9355 (1.0058) model_time 0.9353 (1.0048) loss 0.8887 (0.8073) grad_norm 8.0597 (8.6167/1.9724) mem 68106MB [2022-12-20 12:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1280/1519] eta 0:04:00 lr 0.000006 time 0.9257 (1.0058) model_time 0.9255 (1.0048) loss 0.6890 (0.8070) grad_norm 6.9655 (8.5867/1.9622) mem 68106MB [2022-12-20 12:45:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1290/1519] eta 0:03:50 lr 0.000006 time 1.0981 (1.0058) model_time 1.0976 (1.0049) loss 0.6960 (0.8073) grad_norm 13.5707 (8.6171/1.9940) mem 68106MB [2022-12-20 12:46:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1300/1519] eta 0:03:40 lr 0.000006 time 0.9937 (1.0059) model_time 0.9935 (1.0049) loss 0.8815 (0.8082) grad_norm 10.7728 (8.6954/2.0816) mem 68106MB [2022-12-20 12:46:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1310/1519] eta 0:03:30 lr 0.000006 time 0.9319 (1.0058) model_time 0.9317 (1.0049) loss 0.7428 (0.8077) grad_norm 6.9857 (8.6871/2.0630) mem 68106MB [2022-12-20 12:46:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1320/1519] eta 0:03:20 lr 0.000006 time 0.9826 (1.0059) model_time 0.9825 (1.0049) loss 0.7252 (0.8080) grad_norm 6.5090 (8.6991/2.0635) mem 68106MB [2022-12-20 12:46:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1330/1519] eta 0:03:10 lr 0.000006 time 0.9277 (1.0058) model_time 0.9275 (1.0049) loss 0.8800 (0.8081) grad_norm 7.8082 (8.7088/2.0476) mem 68106MB [2022-12-20 12:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1340/1519] eta 0:03:00 lr 0.000006 time 0.9368 (1.0058) model_time 0.9367 (1.0048) loss 0.7095 (0.8076) grad_norm 12.0656 (8.7683/2.0892) mem 68106MB [2022-12-20 12:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1350/1519] eta 0:02:49 lr 0.000006 time 1.0182 (1.0058) model_time 1.0180 (1.0048) loss 0.9138 (0.8074) grad_norm 6.1144 (8.7369/2.0986) mem 68106MB [2022-12-20 12:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1360/1519] eta 0:02:39 lr 0.000006 time 0.9690 (1.0057) model_time 0.9688 (1.0048) loss 0.6718 (0.8068) grad_norm 6.8704 (8.7360/2.1011) mem 68106MB [2022-12-20 12:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1370/1519] eta 0:02:29 lr 0.000006 time 0.9310 (1.0057) model_time 0.9308 (1.0048) loss 0.7694 (0.8066) grad_norm 7.5645 (8.7507/2.0945) mem 68106MB [2022-12-20 12:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1380/1519] eta 0:02:19 lr 0.000006 time 0.9918 (1.0057) model_time 0.9916 (1.0048) loss 0.8068 (0.8069) grad_norm 6.1365 (8.7513/2.1034) mem 68106MB [2022-12-20 12:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1390/1519] eta 0:02:09 lr 0.000006 time 0.9362 (1.0059) model_time 0.9361 (1.0050) loss 0.6737 (0.8064) grad_norm 8.9491 (8.7653/2.1113) mem 68106MB [2022-12-20 12:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1400/1519] eta 0:01:59 lr 0.000006 time 0.9355 (1.0058) model_time 0.9354 (1.0049) loss 0.6992 (0.8065) grad_norm 9.2073 (8.7627/2.1187) mem 68106MB [2022-12-20 12:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1410/1519] eta 0:01:49 lr 0.000006 time 0.9322 (1.0059) model_time 0.9321 (1.0050) loss 0.7657 (0.8064) grad_norm 7.7784 (8.7475/2.1230) mem 68106MB [2022-12-20 12:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1420/1519] eta 0:01:39 lr 0.000006 time 0.9324 (1.0059) model_time 0.9323 (1.0050) loss 0.9618 (0.8066) grad_norm 7.9112 (8.7545/2.2231) mem 68106MB [2022-12-20 12:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1430/1519] eta 0:01:29 lr 0.000006 time 0.9767 (1.0059) model_time 0.9765 (1.0050) loss 0.9498 (0.8074) grad_norm 7.8075 (8.7856/2.2413) mem 68106MB [2022-12-20 12:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1440/1519] eta 0:01:19 lr 0.000006 time 0.9333 (1.0059) model_time 0.9332 (1.0050) loss 0.6970 (0.8071) grad_norm 6.6145 (8.7475/2.2067) mem 68106MB [2022-12-20 12:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1450/1519] eta 0:01:09 lr 0.000006 time 0.9287 (1.0058) model_time 0.9285 (1.0049) loss 0.6741 (0.8068) grad_norm 7.1531 (8.7481/2.2132) mem 68106MB [2022-12-20 12:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1460/1519] eta 0:00:59 lr 0.000006 time 0.9267 (1.0058) model_time 0.9264 (1.0049) loss 0.6713 (0.8066) grad_norm 9.3495 (8.7575/2.1923) mem 68106MB [2022-12-20 12:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1470/1519] eta 0:00:49 lr 0.000006 time 0.9297 (1.0057) model_time 0.9296 (1.0048) loss 0.6845 (0.8060) grad_norm 9.0405 (8.7989/2.1993) mem 68106MB [2022-12-20 12:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1480/1519] eta 0:00:39 lr 0.000006 time 0.9402 (1.0057) model_time 0.9401 (1.0048) loss 0.6609 (0.8058) grad_norm 11.2065 (8.8086/2.2032) mem 68106MB [2022-12-20 12:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1490/1519] eta 0:00:29 lr 0.000006 time 1.0375 (1.0057) model_time 1.0374 (1.0049) loss 0.7784 (0.8057) grad_norm 7.7049 (8.8156/2.2117) mem 68106MB [2022-12-20 12:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1500/1519] eta 0:00:19 lr 0.000006 time 0.9301 (1.0057) model_time 0.9299 (1.0048) loss 0.7483 (0.8057) grad_norm 6.5506 (8.7848/2.1546) mem 68106MB [2022-12-20 12:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [73/100][1510/1519] eta 0:00:09 lr 0.000006 time 0.9225 (1.0057) model_time 0.9224 (1.0048) loss 0.7949 (0.8059) grad_norm 9.4705 (8.7519/2.1108) mem 68106MB [2022-12-20 12:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 73 training takes 0:25:27 [2022-12-20 12:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_73.pth saving...... [2022-12-20 12:50:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_73.pth saved !!! [2022-12-20 12:50:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.637 (0.637) Loss 0.5283 (0.5283) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 12:50:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.293 (0.328) Loss 0.5350 (0.5025) Acc@1 91.667 (92.740) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 12:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.314) Loss 0.4919 (0.4990) Acc@1 90.278 (92.659) Acc@5 98.958 (98.446) Mem 68106MB [2022-12-20 12:50:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.308) Loss 0.6319 (0.5048) Acc@1 89.583 (92.451) Acc@5 98.264 (98.421) Mem 68106MB [2022-12-20 12:50:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.306) Loss 0.4481 (0.4949) Acc@1 94.097 (92.573) Acc@5 99.306 (98.509) Mem 68106MB [2022-12-20 12:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.4812 (0.4925) Acc@1 92.361 (92.613) Acc@5 99.306 (98.550) Mem 68106MB [2022-12-20 12:50:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.307 (0.304) Loss 0.5161 (0.4921) Acc@1 90.625 (92.543) Acc@5 98.264 (98.543) Mem 68106MB [2022-12-20 12:50:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5423 (0.4932) Acc@1 92.708 (92.483) Acc@5 98.264 (98.552) Mem 68106MB [2022-12-20 12:50:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.302) Loss 0.4317 (0.4918) Acc@1 93.403 (92.524) Acc@5 98.264 (98.577) Mem 68106MB [2022-12-20 12:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:73] * Acc@1 92.469 Acc@5 98.580 [2022-12-20 12:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 12:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 12:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][0/1519] eta 0:49:26 lr 0.000006 time 1.9532 (1.9532) model_time 1.1771 (1.1771) loss 0.8749 (0.8749) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 12:50:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][10/1519] eta 0:27:25 lr 0.000006 time 0.9301 (1.0902) model_time 0.9299 (1.0192) loss 0.8140 (0.7531) grad_norm 10.2718 (8.8623/1.2736) mem 68106MB [2022-12-20 12:50:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][20/1519] eta 0:26:14 lr 0.000006 time 0.9298 (1.0507) model_time 0.9296 (1.0133) loss 0.7025 (0.7665) grad_norm 8.0921 (7.9854/1.5041) mem 68106MB [2022-12-20 12:51:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][30/1519] eta 0:25:40 lr 0.000006 time 0.9342 (1.0346) model_time 0.9340 (1.0092) loss 0.9098 (0.7677) grad_norm 7.4238 (7.7903/1.3292) mem 68106MB [2022-12-20 12:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][40/1519] eta 0:25:20 lr 0.000006 time 0.9322 (1.0284) model_time 0.9321 (1.0090) loss 0.8866 (0.7660) grad_norm 6.4679 (7.6036/1.2704) mem 68106MB [2022-12-20 12:51:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][50/1519] eta 0:25:02 lr 0.000006 time 0.9274 (1.0227) model_time 0.9272 (1.0070) loss 1.0527 (0.7937) grad_norm 10.3339 (7.8645/1.3862) mem 68106MB [2022-12-20 12:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][60/1519] eta 0:24:46 lr 0.000006 time 0.9280 (1.0188) model_time 0.9278 (1.0056) loss 1.0838 (0.8004) grad_norm 13.1430 (7.9807/1.6314) mem 68106MB [2022-12-20 12:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][70/1519] eta 0:24:35 lr 0.000006 time 0.9286 (1.0181) model_time 0.9284 (1.0067) loss 0.8914 (0.7979) grad_norm 9.9293 (8.1499/1.7504) mem 68106MB [2022-12-20 12:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][80/1519] eta 0:24:22 lr 0.000006 time 0.9401 (1.0163) model_time 0.9400 (1.0063) loss 0.6677 (0.7945) grad_norm 8.7680 (8.1666/1.6644) mem 68106MB [2022-12-20 12:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][90/1519] eta 0:24:11 lr 0.000006 time 0.9325 (1.0156) model_time 0.9323 (1.0066) loss 0.6693 (0.8014) grad_norm 9.1549 (8.2042/1.6100) mem 68106MB [2022-12-20 12:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][100/1519] eta 0:24:00 lr 0.000006 time 0.9306 (1.0153) model_time 0.9305 (1.0072) loss 0.6977 (0.8001) grad_norm 6.0220 (8.0328/1.6386) mem 68106MB [2022-12-20 12:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][110/1519] eta 0:23:49 lr 0.000006 time 0.9814 (1.0142) model_time 0.9812 (1.0068) loss 0.6845 (0.7988) grad_norm 7.3918 (7.9858/1.6028) mem 68106MB [2022-12-20 12:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][120/1519] eta 0:23:40 lr 0.000006 time 0.9315 (1.0154) model_time 0.9313 (1.0085) loss 0.6926 (0.7955) grad_norm 6.7998 (8.1059/1.7891) mem 68106MB [2022-12-20 12:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][130/1519] eta 0:23:29 lr 0.000006 time 0.9961 (1.0147) model_time 0.9960 (1.0083) loss 0.8684 (0.7927) grad_norm 6.3007 (8.0916/1.7574) mem 68106MB [2022-12-20 12:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][140/1519] eta 0:23:18 lr 0.000006 time 0.9308 (1.0142) model_time 0.9304 (1.0083) loss 0.7423 (0.7953) grad_norm 6.6866 (8.0443/1.7402) mem 68106MB [2022-12-20 12:53:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][150/1519] eta 0:23:07 lr 0.000006 time 0.9382 (1.0136) model_time 0.9380 (1.0080) loss 0.6723 (0.7903) grad_norm 6.2351 (7.9684/1.7526) mem 68106MB [2022-12-20 12:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][160/1519] eta 0:22:56 lr 0.000006 time 0.9307 (1.0131) model_time 0.9306 (1.0078) loss 0.7098 (0.7947) grad_norm 9.5312 (7.9486/1.7330) mem 68106MB [2022-12-20 12:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][170/1519] eta 0:22:46 lr 0.000006 time 0.9316 (1.0130) model_time 0.9314 (1.0081) loss 0.6711 (0.7949) grad_norm 7.0486 (7.9439/1.6935) mem 68106MB [2022-12-20 12:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][180/1519] eta 0:22:35 lr 0.000006 time 0.9243 (1.0126) model_time 0.9241 (1.0079) loss 0.7144 (0.7967) grad_norm 8.9545 (8.0668/1.7765) mem 68106MB [2022-12-20 12:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][190/1519] eta 0:22:25 lr 0.000006 time 0.9282 (1.0123) model_time 0.9281 (1.0078) loss 0.7772 (0.7990) grad_norm 7.9938 (8.0027/1.7677) mem 68106MB [2022-12-20 12:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][200/1519] eta 0:22:14 lr 0.000006 time 0.9389 (1.0117) model_time 0.9388 (1.0075) loss 0.8506 (0.7978) grad_norm 7.4986 (7.9761/1.7301) mem 68106MB [2022-12-20 12:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][210/1519] eta 0:22:03 lr 0.000006 time 0.9256 (1.0114) model_time 0.9255 (1.0073) loss 0.7372 (0.7979) grad_norm 7.1725 (8.0657/1.7897) mem 68106MB [2022-12-20 12:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][220/1519] eta 0:21:52 lr 0.000006 time 0.9302 (1.0107) model_time 0.9301 (1.0068) loss 0.8931 (0.7984) grad_norm 6.7659 (8.1001/1.8217) mem 68106MB [2022-12-20 12:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][230/1519] eta 0:21:42 lr 0.000006 time 0.9337 (1.0104) model_time 0.9336 (1.0066) loss 0.7208 (0.7983) grad_norm 8.8097 (8.0897/1.7963) mem 68106MB [2022-12-20 12:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][240/1519] eta 0:21:31 lr 0.000006 time 0.9325 (1.0100) model_time 0.9324 (1.0064) loss 0.6784 (0.7954) grad_norm 6.6246 (8.0781/1.7666) mem 68106MB [2022-12-20 12:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][250/1519] eta 0:21:21 lr 0.000006 time 0.9383 (1.0098) model_time 0.9382 (1.0063) loss 0.8591 (0.7962) grad_norm 11.7098 (8.1563/1.7829) mem 68106MB [2022-12-20 12:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][260/1519] eta 0:21:10 lr 0.000006 time 0.9291 (1.0093) model_time 0.9289 (1.0060) loss 0.8730 (0.7960) grad_norm 8.5832 (8.2172/1.8446) mem 68106MB [2022-12-20 12:55:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][270/1519] eta 0:21:01 lr 0.000006 time 0.9339 (1.0098) model_time 0.9338 (1.0065) loss 0.7460 (0.7959) grad_norm 8.7765 (8.2047/1.8365) mem 68106MB [2022-12-20 12:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][280/1519] eta 0:20:50 lr 0.000006 time 0.9342 (1.0095) model_time 0.9341 (1.0063) loss 0.7646 (0.7977) grad_norm 17.6409 (8.3011/1.9983) mem 68106MB [2022-12-20 12:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][290/1519] eta 0:20:40 lr 0.000006 time 0.9326 (1.0093) model_time 0.9325 (1.0062) loss 0.7107 (0.8011) grad_norm 9.3073 (8.3303/1.9741) mem 68106MB [2022-12-20 12:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][300/1519] eta 0:20:30 lr 0.000006 time 0.9386 (1.0092) model_time 0.9384 (1.0062) loss 0.7042 (0.8015) grad_norm 7.8988 (8.3198/1.9548) mem 68106MB [2022-12-20 12:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][310/1519] eta 0:20:19 lr 0.000006 time 0.9431 (1.0089) model_time 0.9426 (1.0060) loss 0.9361 (0.8038) grad_norm 7.7963 (8.3119/1.9477) mem 68106MB [2022-12-20 12:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][320/1519] eta 0:20:09 lr 0.000006 time 0.9280 (1.0086) model_time 0.9279 (1.0058) loss 0.7008 (0.8026) grad_norm 7.8783 (8.3047/1.9320) mem 68106MB [2022-12-20 12:56:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][330/1519] eta 0:19:59 lr 0.000006 time 0.9946 (1.0087) model_time 0.9945 (1.0060) loss 0.8948 (0.8013) grad_norm 7.2398 (8.2624/1.9260) mem 68106MB [2022-12-20 12:56:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][340/1519] eta 0:19:48 lr 0.000006 time 0.9379 (1.0084) model_time 0.9378 (1.0058) loss 0.9025 (0.8002) grad_norm 13.6842 (8.2884/1.9507) mem 68106MB [2022-12-20 12:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][350/1519] eta 0:19:38 lr 0.000006 time 0.9317 (1.0084) model_time 0.9316 (1.0059) loss 1.0420 (0.8014) grad_norm 6.9839 (8.3006/1.9497) mem 68106MB [2022-12-20 12:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][360/1519] eta 0:19:28 lr 0.000006 time 0.9295 (1.0083) model_time 0.9294 (1.0058) loss 0.7843 (0.8024) grad_norm 10.5821 (8.3546/1.9878) mem 68106MB [2022-12-20 12:56:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][370/1519] eta 0:19:18 lr 0.000006 time 0.9256 (1.0081) model_time 0.9255 (1.0057) loss 0.6957 (0.8041) grad_norm 7.4793 (8.3607/1.9832) mem 68106MB [2022-12-20 12:56:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][380/1519] eta 0:19:08 lr 0.000006 time 0.9320 (1.0079) model_time 0.9318 (1.0055) loss 0.6972 (0.8055) grad_norm 7.1413 (8.3601/1.9670) mem 68106MB [2022-12-20 12:57:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][390/1519] eta 0:18:57 lr 0.000006 time 0.9375 (1.0078) model_time 0.9373 (1.0055) loss 0.8446 (0.8062) grad_norm 10.8402 (8.3901/1.9573) mem 68106MB [2022-12-20 12:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][400/1519] eta 0:18:47 lr 0.000006 time 0.9325 (1.0076) model_time 0.9323 (1.0052) loss 0.7878 (0.8070) grad_norm 7.0013 (8.3788/1.9404) mem 68106MB [2022-12-20 12:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][410/1519] eta 0:18:37 lr 0.000006 time 0.9125 (1.0075) model_time 0.9123 (1.0053) loss 0.9374 (0.8060) grad_norm 7.9085 (8.3849/1.9202) mem 68106MB [2022-12-20 12:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][420/1519] eta 0:18:27 lr 0.000006 time 0.9294 (1.0073) model_time 0.9292 (1.0051) loss 0.6736 (0.8050) grad_norm 6.0241 (8.3817/1.9231) mem 68106MB [2022-12-20 12:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][430/1519] eta 0:18:17 lr 0.000005 time 0.9232 (1.0074) model_time 0.9231 (1.0052) loss 0.7211 (0.8056) grad_norm 6.3552 (8.3759/1.9332) mem 68106MB [2022-12-20 12:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][440/1519] eta 0:18:06 lr 0.000005 time 0.9331 (1.0072) model_time 0.9329 (1.0051) loss 0.8568 (0.8062) grad_norm 10.6524 (8.3871/1.9258) mem 68106MB [2022-12-20 12:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][450/1519] eta 0:17:56 lr 0.000005 time 0.9273 (1.0073) model_time 0.9271 (1.0052) loss 0.8484 (0.8062) grad_norm 8.0562 (8.3955/1.9265) mem 68106MB [2022-12-20 12:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][460/1519] eta 0:17:46 lr 0.000005 time 0.9370 (1.0072) model_time 0.9368 (1.0051) loss 0.6581 (0.8059) grad_norm 10.3701 (8.3832/1.9179) mem 68106MB [2022-12-20 12:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][470/1519] eta 0:17:36 lr 0.000005 time 0.9833 (1.0072) model_time 0.9831 (1.0052) loss 0.7534 (0.8053) grad_norm 8.3756 (8.3728/1.9052) mem 68106MB [2022-12-20 12:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][480/1519] eta 0:17:26 lr 0.000005 time 0.9335 (1.0070) model_time 0.9334 (1.0050) loss 0.6781 (0.8046) grad_norm 8.0598 (8.3726/1.9026) mem 68106MB [2022-12-20 12:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][490/1519] eta 0:17:16 lr 0.000005 time 0.9322 (1.0068) model_time 0.9320 (1.0049) loss 0.9217 (0.8045) grad_norm 12.1284 (8.3957/1.8995) mem 68106MB [2022-12-20 12:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][500/1519] eta 0:17:05 lr 0.000005 time 0.9182 (1.0069) model_time 0.9181 (1.0049) loss 0.6551 (0.8038) grad_norm 6.4156 (8.3999/1.9219) mem 68106MB [2022-12-20 12:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][510/1519] eta 0:16:55 lr 0.000005 time 0.9306 (1.0067) model_time 0.9304 (1.0048) loss 0.9149 (0.8040) grad_norm 8.9961 (8.4167/1.9653) mem 68106MB [2022-12-20 12:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][520/1519] eta 0:16:45 lr 0.000005 time 0.9317 (1.0066) model_time 0.9316 (1.0047) loss 0.7856 (0.8045) grad_norm 8.4437 (8.4078/1.9497) mem 68106MB [2022-12-20 12:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][530/1519] eta 0:16:35 lr 0.000005 time 0.9318 (1.0066) model_time 0.9317 (1.0048) loss 0.8276 (0.8044) grad_norm 11.7078 (8.4359/1.9725) mem 68106MB [2022-12-20 12:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][540/1519] eta 0:16:25 lr 0.000005 time 0.9303 (1.0065) model_time 0.9302 (1.0047) loss 0.6819 (0.8034) grad_norm 8.0193 (8.4634/1.9717) mem 68106MB [2022-12-20 12:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][550/1519] eta 0:16:15 lr 0.000005 time 0.9380 (1.0064) model_time 0.9379 (1.0046) loss 0.8011 (0.8033) grad_norm 6.1657 (8.4455/1.9610) mem 68106MB [2022-12-20 12:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][560/1519] eta 0:16:05 lr 0.000005 time 0.9305 (1.0065) model_time 0.9303 (1.0048) loss 0.7593 (0.8027) grad_norm 7.5597 (8.4536/1.9673) mem 68106MB [2022-12-20 13:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][570/1519] eta 0:15:55 lr 0.000005 time 0.9306 (1.0064) model_time 0.9305 (1.0047) loss 0.9894 (0.8021) grad_norm 7.4834 (8.4706/1.9741) mem 68106MB [2022-12-20 13:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][580/1519] eta 0:15:45 lr 0.000005 time 0.9349 (1.0067) model_time 0.9348 (1.0050) loss 0.6689 (0.8031) grad_norm 11.3832 (8.4960/1.9764) mem 68106MB [2022-12-20 13:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][590/1519] eta 0:15:35 lr 0.000005 time 0.9364 (1.0067) model_time 0.9361 (1.0050) loss 0.9616 (0.8027) grad_norm 7.6369 (8.4753/1.9684) mem 68106MB [2022-12-20 13:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][600/1519] eta 0:15:25 lr 0.000005 time 0.9128 (1.0066) model_time 0.9126 (1.0049) loss 0.7456 (0.8030) grad_norm 9.5806 (8.4991/1.9747) mem 68106MB [2022-12-20 13:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][610/1519] eta 0:15:14 lr 0.000005 time 0.9298 (1.0066) model_time 0.9296 (1.0049) loss 0.7736 (0.8035) grad_norm 8.3453 (8.4849/1.9726) mem 68106MB [2022-12-20 13:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][620/1519] eta 0:15:04 lr 0.000005 time 0.9286 (1.0064) model_time 0.9285 (1.0048) loss 1.1368 (0.8044) grad_norm 8.9766 (8.5018/1.9647) mem 68106MB [2022-12-20 13:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][630/1519] eta 0:14:54 lr 0.000005 time 0.9369 (1.0064) model_time 0.9368 (1.0048) loss 0.7067 (0.8047) grad_norm 6.6020 (8.5293/1.9652) mem 68106MB [2022-12-20 13:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][640/1519] eta 0:14:44 lr 0.000005 time 0.9293 (1.0064) model_time 0.9292 (1.0048) loss 0.7426 (0.8052) grad_norm 6.3117 (8.5833/2.0350) mem 68106MB [2022-12-20 13:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][650/1519] eta 0:14:34 lr 0.000005 time 0.9878 (1.0065) model_time 0.9877 (1.0049) loss 0.9298 (0.8051) grad_norm 7.9437 (8.5730/2.0491) mem 68106MB [2022-12-20 13:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][660/1519] eta 0:14:24 lr 0.000005 time 0.9372 (1.0065) model_time 0.9370 (1.0049) loss 0.8480 (0.8045) grad_norm 8.5079 (8.5600/2.0324) mem 68106MB [2022-12-20 13:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][670/1519] eta 0:14:14 lr 0.000005 time 0.9304 (1.0066) model_time 0.9303 (1.0050) loss 0.7418 (0.8033) grad_norm 7.5465 (8.5379/2.0192) mem 68106MB [2022-12-20 13:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][680/1519] eta 0:14:04 lr 0.000005 time 0.9356 (1.0065) model_time 0.9354 (1.0050) loss 1.2026 (0.8040) grad_norm 9.8623 (8.5450/2.0210) mem 68106MB [2022-12-20 13:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][690/1519] eta 0:13:54 lr 0.000005 time 0.9353 (1.0064) model_time 0.9351 (1.0049) loss 1.0736 (0.8046) grad_norm 8.3435 (8.5469/2.0208) mem 68106MB [2022-12-20 13:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][700/1519] eta 0:13:44 lr 0.000005 time 0.9172 (1.0063) model_time 0.9171 (1.0048) loss 1.0132 (0.8054) grad_norm 8.4793 (8.6047/2.0213) mem 68106MB [2022-12-20 13:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][710/1519] eta 0:13:34 lr 0.000005 time 0.9316 (1.0062) model_time 0.9315 (1.0047) loss 0.8046 (0.8051) grad_norm 6.7275 (8.6776/2.2334) mem 68106MB [2022-12-20 13:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][720/1519] eta 0:13:23 lr 0.000005 time 0.9310 (1.0061) model_time 0.9307 (1.0047) loss 0.7524 (0.8046) grad_norm 9.0712 (8.6815/2.2164) mem 68106MB [2022-12-20 13:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][730/1519] eta 0:13:13 lr 0.000005 time 0.9299 (1.0061) model_time 0.9298 (1.0046) loss 0.8000 (0.8051) grad_norm 6.8419 (8.6759/2.2145) mem 68106MB [2022-12-20 13:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][740/1519] eta 0:13:03 lr 0.000005 time 0.9323 (1.0062) model_time 0.9322 (1.0047) loss 0.9939 (0.8045) grad_norm 7.9200 (8.6763/2.2125) mem 68106MB [2022-12-20 13:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][750/1519] eta 0:12:53 lr 0.000005 time 0.9380 (1.0061) model_time 0.9379 (1.0047) loss 0.6700 (0.8042) grad_norm 8.0178 (8.7087/2.2341) mem 68106MB [2022-12-20 13:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][760/1519] eta 0:12:43 lr 0.000005 time 0.9284 (1.0062) model_time 0.9282 (1.0048) loss 0.7800 (0.8033) grad_norm 6.2332 (8.7077/2.2302) mem 68106MB [2022-12-20 13:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][770/1519] eta 0:12:33 lr 0.000005 time 0.9302 (1.0061) model_time 0.9300 (1.0047) loss 0.6803 (0.8031) grad_norm 11.7682 (8.7177/2.2389) mem 68106MB [2022-12-20 13:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][780/1519] eta 0:12:23 lr 0.000005 time 0.9454 (1.0061) model_time 0.9453 (1.0047) loss 0.7125 (0.8034) grad_norm 10.8487 (8.7098/2.2319) mem 68106MB [2022-12-20 13:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][790/1519] eta 0:12:13 lr 0.000005 time 0.9342 (1.0060) model_time 0.9341 (1.0047) loss 0.8251 (0.8036) grad_norm 7.7000 (8.7374/2.2254) mem 68106MB [2022-12-20 13:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][800/1519] eta 0:12:03 lr 0.000005 time 0.9339 (1.0060) model_time 0.9337 (1.0046) loss 0.8580 (0.8043) grad_norm 9.2685 (8.7798/2.2304) mem 68106MB [2022-12-20 13:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][810/1519] eta 0:11:53 lr 0.000005 time 0.9297 (1.0059) model_time 0.9296 (1.0046) loss 0.7851 (0.8042) grad_norm 8.6618 (8.7708/2.2339) mem 68106MB [2022-12-20 13:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][820/1519] eta 0:11:43 lr 0.000005 time 0.9277 (1.0059) model_time 0.9276 (1.0045) loss 0.6896 (0.8045) grad_norm 6.4498 (8.7513/2.2246) mem 68106MB [2022-12-20 13:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][830/1519] eta 0:11:33 lr 0.000005 time 0.9345 (1.0059) model_time 0.9343 (1.0045) loss 0.6908 (0.8041) grad_norm 8.4000 (8.7593/2.2203) mem 68106MB [2022-12-20 13:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][840/1519] eta 0:11:23 lr 0.000005 time 0.9356 (1.0059) model_time 0.9354 (1.0046) loss 0.7025 (0.8043) grad_norm 9.2223 (8.7740/2.2292) mem 68106MB [2022-12-20 13:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][850/1519] eta 0:11:12 lr 0.000005 time 0.9304 (1.0058) model_time 0.9303 (1.0045) loss 0.8072 (0.8035) grad_norm 5.5392 (8.7448/2.2309) mem 68106MB [2022-12-20 13:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][860/1519] eta 0:11:02 lr 0.000005 time 0.9322 (1.0057) model_time 0.9320 (1.0045) loss 0.7069 (0.8041) grad_norm 7.9393 (8.7164/2.2079) mem 68106MB [2022-12-20 13:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][870/1519] eta 0:10:52 lr 0.000005 time 0.9292 (1.0057) model_time 0.9290 (1.0044) loss 1.3483 (0.8044) grad_norm 9.0456 (8.7244/2.1986) mem 68106MB [2022-12-20 13:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][880/1519] eta 0:10:42 lr 0.000005 time 0.9348 (1.0056) model_time 0.9347 (1.0044) loss 0.8275 (0.8041) grad_norm 6.7064 (8.6627/2.1459) mem 68106MB [2022-12-20 13:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][890/1519] eta 0:10:32 lr 0.000005 time 1.0316 (1.0056) model_time 1.0315 (1.0044) loss 0.8644 (0.8058) grad_norm 8.6985 (8.6544/2.1435) mem 68106MB [2022-12-20 13:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][900/1519] eta 0:10:22 lr 0.000005 time 0.9296 (1.0057) model_time 0.9295 (1.0044) loss 0.6645 (0.8058) grad_norm 7.6657 (8.6693/2.1424) mem 68106MB [2022-12-20 13:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][910/1519] eta 0:10:12 lr 0.000005 time 0.9358 (1.0056) model_time 0.9357 (1.0044) loss 0.9136 (0.8061) grad_norm 7.6819 (8.6779/2.1336) mem 68106MB [2022-12-20 13:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][920/1519] eta 0:10:02 lr 0.000005 time 0.9596 (1.0055) model_time 0.9594 (1.0043) loss 0.7657 (0.8066) grad_norm 9.4204 (8.6955/2.1409) mem 68106MB [2022-12-20 13:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][930/1519] eta 0:09:52 lr 0.000005 time 0.9343 (1.0055) model_time 0.9341 (1.0043) loss 0.9410 (0.8073) grad_norm 11.1200 (8.7389/2.1360) mem 68106MB [2022-12-20 13:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][940/1519] eta 0:09:42 lr 0.000005 time 0.9302 (1.0055) model_time 0.9299 (1.0043) loss 0.6576 (0.8068) grad_norm 7.5235 (8.7110/2.1191) mem 68106MB [2022-12-20 13:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][950/1519] eta 0:09:32 lr 0.000005 time 0.9808 (1.0055) model_time 0.9806 (1.0043) loss 0.8026 (0.8063) grad_norm 7.3984 (8.6978/2.1138) mem 68106MB [2022-12-20 13:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][960/1519] eta 0:09:22 lr 0.000005 time 0.9309 (1.0055) model_time 0.9307 (1.0043) loss 0.7618 (0.8059) grad_norm 8.9393 (8.6586/2.0862) mem 68106MB [2022-12-20 13:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][970/1519] eta 0:09:12 lr 0.000005 time 0.9393 (1.0055) model_time 0.9391 (1.0043) loss 1.0361 (0.8061) grad_norm 9.0188 (8.6588/2.0745) mem 68106MB [2022-12-20 13:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][980/1519] eta 0:09:02 lr 0.000005 time 0.9664 (1.0057) model_time 0.9662 (1.0046) loss 0.8559 (0.8057) grad_norm 7.0686 (8.6511/2.0780) mem 68106MB [2022-12-20 13:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][990/1519] eta 0:08:52 lr 0.000005 time 0.9328 (1.0057) model_time 0.9326 (1.0045) loss 1.0773 (0.8061) grad_norm 8.0265 (8.6369/2.0739) mem 68106MB [2022-12-20 13:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1000/1519] eta 0:08:41 lr 0.000005 time 0.9359 (1.0056) model_time 0.9358 (1.0045) loss 0.6620 (0.8057) grad_norm 8.3376 (8.6445/2.0859) mem 68106MB [2022-12-20 13:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1010/1519] eta 0:08:31 lr 0.000005 time 0.9305 (1.0056) model_time 0.9304 (1.0045) loss 1.1904 (0.8054) grad_norm 8.2060 (8.6696/2.1189) mem 68106MB [2022-12-20 13:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1020/1519] eta 0:08:21 lr 0.000005 time 0.9346 (1.0056) model_time 0.9344 (1.0044) loss 0.6808 (0.8057) grad_norm 6.5629 (8.6803/2.1090) mem 68106MB [2022-12-20 13:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1030/1519] eta 0:08:11 lr 0.000005 time 0.9285 (1.0055) model_time 0.9283 (1.0044) loss 0.7129 (0.8055) grad_norm 9.3664 (8.6801/2.0958) mem 68106MB [2022-12-20 13:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1040/1519] eta 0:08:01 lr 0.000005 time 0.9290 (1.0055) model_time 0.9289 (1.0044) loss 0.8838 (0.8059) grad_norm 8.4158 (8.6829/2.0934) mem 68106MB [2022-12-20 13:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1050/1519] eta 0:07:51 lr 0.000005 time 1.0045 (1.0056) model_time 1.0043 (1.0044) loss 0.7290 (0.8059) grad_norm 7.9201 (8.6698/2.0837) mem 68106MB [2022-12-20 13:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1060/1519] eta 0:07:41 lr 0.000005 time 0.9312 (1.0055) model_time 0.9310 (1.0044) loss 0.6782 (0.8055) grad_norm 7.4298 (8.6869/2.0819) mem 68106MB [2022-12-20 13:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1070/1519] eta 0:07:31 lr 0.000005 time 0.9307 (1.0055) model_time 0.9306 (1.0044) loss 0.6530 (0.8052) grad_norm 6.5524 (8.6783/2.0913) mem 68106MB [2022-12-20 13:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1080/1519] eta 0:07:21 lr 0.000005 time 0.9280 (1.0055) model_time 0.9279 (1.0044) loss 0.6847 (0.8058) grad_norm 8.4467 (8.6615/2.0884) mem 68106MB [2022-12-20 13:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1090/1519] eta 0:07:11 lr 0.000005 time 0.9115 (1.0055) model_time 0.9114 (1.0044) loss 0.8723 (0.8058) grad_norm 9.0101 (8.6447/2.0916) mem 68106MB [2022-12-20 13:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1100/1519] eta 0:07:01 lr 0.000005 time 0.9364 (1.0055) model_time 0.9362 (1.0044) loss 1.0060 (0.8055) grad_norm 9.6005 (8.6547/2.0670) mem 68106MB [2022-12-20 13:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1110/1519] eta 0:06:51 lr 0.000005 time 0.9318 (1.0054) model_time 0.9317 (1.0043) loss 0.6934 (0.8052) grad_norm 9.1958 (8.6661/2.0577) mem 68106MB [2022-12-20 13:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1120/1519] eta 0:06:41 lr 0.000005 time 0.9272 (1.0053) model_time 0.9271 (1.0043) loss 0.9788 (0.8054) grad_norm 6.2504 (8.6692/2.0599) mem 68106MB [2022-12-20 13:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1130/1519] eta 0:06:31 lr 0.000005 time 0.9417 (1.0054) model_time 0.9415 (1.0043) loss 0.9027 (0.8062) grad_norm 8.7834 (8.6611/2.0444) mem 68106MB [2022-12-20 13:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1140/1519] eta 0:06:21 lr 0.000005 time 0.9397 (1.0054) model_time 0.9394 (1.0043) loss 0.7869 (0.8060) grad_norm 5.9382 (8.6369/2.0487) mem 68106MB [2022-12-20 13:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1150/1519] eta 0:06:10 lr 0.000005 time 0.9325 (1.0054) model_time 0.9324 (1.0043) loss 0.8611 (0.8057) grad_norm 7.3642 (8.6386/2.0464) mem 68106MB [2022-12-20 13:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1160/1519] eta 0:06:00 lr 0.000005 time 0.9276 (1.0053) model_time 0.9274 (1.0043) loss 0.6577 (0.8051) grad_norm 6.7804 (8.6229/2.0330) mem 68106MB [2022-12-20 13:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1170/1519] eta 0:05:50 lr 0.000005 time 0.9313 (1.0053) model_time 0.9311 (1.0042) loss 0.8285 (0.8056) grad_norm 6.6154 (8.6155/2.0247) mem 68106MB [2022-12-20 13:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1180/1519] eta 0:05:40 lr 0.000005 time 0.9875 (1.0053) model_time 0.9874 (1.0043) loss 0.6941 (0.8057) grad_norm 12.9571 (8.5985/2.0312) mem 68106MB [2022-12-20 13:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1190/1519] eta 0:05:30 lr 0.000005 time 0.9476 (1.0053) model_time 0.9475 (1.0042) loss 0.6900 (0.8052) grad_norm 7.6962 (8.6165/2.0234) mem 68106MB [2022-12-20 13:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1200/1519] eta 0:05:20 lr 0.000005 time 0.9316 (1.0052) model_time 0.9315 (1.0042) loss 0.7064 (0.8051) grad_norm 17.6787 (8.6304/2.0780) mem 68106MB [2022-12-20 13:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1210/1519] eta 0:05:10 lr 0.000005 time 0.9820 (1.0053) model_time 0.9818 (1.0042) loss 0.7635 (0.8045) grad_norm 12.9403 (8.6418/2.0931) mem 68106MB [2022-12-20 13:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1220/1519] eta 0:05:00 lr 0.000005 time 0.9343 (1.0053) model_time 0.9341 (1.0043) loss 0.7550 (0.8043) grad_norm 7.4714 (8.6620/2.1026) mem 68106MB [2022-12-20 13:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1230/1519] eta 0:04:50 lr 0.000005 time 1.0219 (1.0054) model_time 1.0217 (1.0044) loss 0.8180 (0.8045) grad_norm 8.8082 (8.6551/2.0982) mem 68106MB [2022-12-20 13:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1240/1519] eta 0:04:40 lr 0.000005 time 0.9291 (1.0054) model_time 0.9290 (1.0044) loss 0.9683 (0.8049) grad_norm 8.3031 (8.6410/2.0319) mem 68106MB [2022-12-20 13:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1250/1519] eta 0:04:30 lr 0.000005 time 0.9809 (1.0054) model_time 0.9808 (1.0044) loss 1.0011 (0.8051) grad_norm 7.4689 (8.6429/2.0146) mem 68106MB [2022-12-20 13:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1260/1519] eta 0:04:20 lr 0.000005 time 0.9330 (1.0054) model_time 0.9328 (1.0044) loss 0.9710 (0.8057) grad_norm 13.7794 (8.6951/2.0444) mem 68106MB [2022-12-20 13:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1270/1519] eta 0:04:10 lr 0.000005 time 0.9324 (1.0054) model_time 0.9322 (1.0044) loss 0.6844 (0.8059) grad_norm 8.2856 (8.7080/2.0426) mem 68106MB [2022-12-20 13:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1280/1519] eta 0:04:00 lr 0.000005 time 0.9384 (1.0054) model_time 0.9382 (1.0044) loss 0.9414 (0.8065) grad_norm 7.7391 (8.7054/2.0538) mem 68106MB [2022-12-20 13:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1290/1519] eta 0:03:50 lr 0.000005 time 0.9430 (1.0054) model_time 0.9429 (1.0044) loss 0.9774 (0.8068) grad_norm 11.5030 (8.7175/2.0629) mem 68106MB [2022-12-20 13:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1300/1519] eta 0:03:40 lr 0.000005 time 0.9372 (1.0054) model_time 0.9370 (1.0044) loss 0.8082 (0.8066) grad_norm 8.1786 (8.6832/2.0535) mem 68106MB [2022-12-20 13:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1310/1519] eta 0:03:30 lr 0.000005 time 0.9323 (1.0054) model_time 0.9322 (1.0044) loss 0.6735 (0.8063) grad_norm 7.5885 (8.6222/1.8146) mem 68106MB [2022-12-20 13:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1320/1519] eta 0:03:20 lr 0.000005 time 0.9300 (1.0054) model_time 0.9298 (1.0044) loss 1.5121 (0.8068) grad_norm 6.6107 (8.5676/1.8186) mem 68106MB [2022-12-20 13:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1330/1519] eta 0:03:10 lr 0.000005 time 0.9387 (1.0054) model_time 0.9385 (1.0044) loss 0.7375 (0.8072) grad_norm 7.4479 (8.5757/1.8226) mem 68106MB [2022-12-20 13:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1340/1519] eta 0:02:59 lr 0.000005 time 0.9408 (1.0053) model_time 0.9407 (1.0044) loss 0.9358 (0.8069) grad_norm 10.3881 (8.5935/1.8202) mem 68106MB [2022-12-20 13:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1350/1519] eta 0:02:49 lr 0.000005 time 0.9299 (1.0053) model_time 0.9298 (1.0043) loss 0.6975 (0.8071) grad_norm 11.5119 (8.5946/1.7846) mem 68106MB [2022-12-20 13:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1360/1519] eta 0:02:39 lr 0.000005 time 1.0002 (1.0053) model_time 1.0001 (1.0043) loss 0.7446 (0.8071) grad_norm 9.4663 (8.6116/1.7827) mem 68106MB [2022-12-20 13:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1370/1519] eta 0:02:29 lr 0.000005 time 0.9311 (1.0052) model_time 0.9310 (1.0043) loss 0.8146 (0.8069) grad_norm 9.2821 (8.6078/1.7874) mem 68106MB [2022-12-20 13:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1380/1519] eta 0:02:19 lr 0.000005 time 0.9309 (1.0052) model_time 0.9308 (1.0042) loss 0.9416 (0.8070) grad_norm 7.2068 (8.5799/1.7749) mem 68106MB [2022-12-20 13:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1390/1519] eta 0:02:09 lr 0.000005 time 0.9806 (1.0053) model_time 0.9805 (1.0043) loss 0.6922 (0.8070) grad_norm 7.6007 (8.5677/1.7702) mem 68106MB [2022-12-20 13:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1400/1519] eta 0:01:59 lr 0.000005 time 0.9319 (1.0052) model_time 0.9317 (1.0043) loss 0.9115 (0.8072) grad_norm 7.5571 (8.5575/1.7677) mem 68106MB [2022-12-20 13:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1410/1519] eta 0:01:49 lr 0.000005 time 0.9280 (1.0052) model_time 0.9279 (1.0043) loss 0.7202 (0.8074) grad_norm 7.4791 (8.5264/1.7480) mem 68106MB [2022-12-20 13:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1420/1519] eta 0:01:39 lr 0.000005 time 0.9341 (1.0052) model_time 0.9339 (1.0043) loss 1.0441 (0.8077) grad_norm 8.3717 (8.5405/1.7547) mem 68106MB [2022-12-20 13:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1430/1519] eta 0:01:29 lr 0.000005 time 0.9394 (1.0052) model_time 0.9393 (1.0043) loss 1.1067 (0.8078) grad_norm 9.6136 (8.5607/1.7772) mem 68106MB [2022-12-20 13:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1440/1519] eta 0:01:19 lr 0.000005 time 0.9376 (1.0052) model_time 0.9375 (1.0043) loss 0.7627 (0.8074) grad_norm 6.2899 (8.5592/1.7684) mem 68106MB [2022-12-20 13:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1450/1519] eta 0:01:09 lr 0.000005 time 0.9326 (1.0052) model_time 0.9324 (1.0043) loss 0.8216 (0.8074) grad_norm 8.0220 (8.5578/1.7598) mem 68106MB [2022-12-20 13:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1460/1519] eta 0:00:59 lr 0.000005 time 0.9267 (1.0051) model_time 0.9265 (1.0042) loss 0.7929 (0.8076) grad_norm 9.9879 (8.5783/1.7932) mem 68106MB [2022-12-20 13:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1470/1519] eta 0:00:49 lr 0.000005 time 0.9389 (1.0052) model_time 0.9387 (1.0043) loss 0.6594 (0.8077) grad_norm 9.3918 (8.6096/1.8127) mem 68106MB [2022-12-20 13:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1480/1519] eta 0:00:39 lr 0.000005 time 0.9306 (1.0052) model_time 0.9305 (1.0043) loss 0.7608 (0.8075) grad_norm 7.6278 (8.6331/1.7947) mem 68106MB [2022-12-20 13:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1490/1519] eta 0:00:29 lr 0.000005 time 0.9335 (1.0052) model_time 0.9333 (1.0043) loss 0.7562 (0.8075) grad_norm 11.6075 (8.6756/1.8656) mem 68106MB [2022-12-20 13:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1500/1519] eta 0:00:19 lr 0.000005 time 0.9300 (1.0051) model_time 0.9299 (1.0042) loss 0.6865 (0.8071) grad_norm 5.8507 (8.6517/1.8714) mem 68106MB [2022-12-20 13:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [74/100][1510/1519] eta 0:00:09 lr 0.000005 time 0.9123 (1.0051) model_time 0.9122 (1.0042) loss 0.8759 (0.8070) grad_norm 8.0642 (8.6563/1.8695) mem 68106MB [2022-12-20 13:15:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 74 training takes 0:25:26 [2022-12-20 13:15:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_74.pth saving...... [2022-12-20 13:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_74.pth saved !!! [2022-12-20 13:16:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.641 (0.641) Loss 0.5242 (0.5242) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 13:16:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.301 (0.329) Loss 0.5248 (0.4983) Acc@1 92.014 (92.866) Acc@5 98.264 (98.390) Mem 68106MB [2022-12-20 13:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4825 (0.4956) Acc@1 90.972 (92.725) Acc@5 99.306 (98.380) Mem 68106MB [2022-12-20 13:16:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.308) Loss 0.6268 (0.5025) Acc@1 90.278 (92.406) Acc@5 97.917 (98.387) Mem 68106MB [2022-12-20 13:16:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.304 (0.306) Loss 0.4469 (0.4925) Acc@1 93.056 (92.463) Acc@5 99.306 (98.493) Mem 68106MB [2022-12-20 13:16:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.295 (0.305) Loss 0.4828 (0.4900) Acc@1 91.319 (92.504) Acc@5 99.653 (98.536) Mem 68106MB [2022-12-20 13:16:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.5179 (0.4901) Acc@1 90.972 (92.475) Acc@5 98.264 (98.531) Mem 68106MB [2022-12-20 13:16:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.303) Loss 0.5478 (0.4910) Acc@1 91.319 (92.430) Acc@5 98.611 (98.538) Mem 68106MB [2022-12-20 13:16:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.302) Loss 0.4279 (0.4894) Acc@1 93.403 (92.494) Acc@5 98.264 (98.555) Mem 68106MB [2022-12-20 13:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:74] * Acc@1 92.461 Acc@5 98.563 [2022-12-20 13:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 13:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 13:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][0/1519] eta 0:45:01 lr 0.000005 time 1.7787 (1.7787) model_time 1.0365 (1.0365) loss 0.6988 (0.6988) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 13:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][10/1519] eta 0:27:19 lr 0.000005 time 0.9230 (1.0866) model_time 0.9228 (1.0187) loss 0.6799 (0.8367) grad_norm 12.9973 (8.8007/2.7888) mem 68106MB [2022-12-20 13:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][20/1519] eta 0:26:06 lr 0.000005 time 0.9286 (1.0448) model_time 0.9285 (1.0090) loss 0.6879 (0.8264) grad_norm 7.0947 (8.0442/2.1245) mem 68106MB [2022-12-20 13:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][30/1519] eta 0:25:39 lr 0.000005 time 0.9330 (1.0337) model_time 0.9328 (1.0093) loss 0.7899 (0.8228) grad_norm 6.7354 (7.9396/1.8328) mem 68106MB [2022-12-20 13:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][40/1519] eta 0:25:17 lr 0.000005 time 0.9283 (1.0260) model_time 0.9281 (1.0074) loss 0.7365 (0.8036) grad_norm 11.7471 (7.9715/1.8454) mem 68106MB [2022-12-20 13:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][50/1519] eta 0:25:04 lr 0.000005 time 0.9273 (1.0241) model_time 0.9272 (1.0091) loss 0.6953 (0.8088) grad_norm 7.8768 (7.8174/1.7054) mem 68106MB [2022-12-20 13:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][60/1519] eta 0:24:48 lr 0.000005 time 0.9319 (1.0202) model_time 0.9317 (1.0076) loss 0.8686 (0.8151) grad_norm 6.9455 (8.0418/1.7833) mem 68106MB [2022-12-20 13:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][70/1519] eta 0:24:34 lr 0.000005 time 0.9052 (1.0178) model_time 0.9051 (1.0069) loss 0.7484 (0.8113) grad_norm 11.3674 (8.1426/1.8869) mem 68106MB [2022-12-20 13:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][80/1519] eta 0:24:21 lr 0.000005 time 0.9248 (1.0157) model_time 0.9247 (1.0061) loss 0.9275 (0.8165) grad_norm 8.6515 (8.0497/1.8341) mem 68106MB [2022-12-20 13:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][90/1519] eta 0:24:09 lr 0.000005 time 0.9250 (1.0145) model_time 0.9248 (1.0059) loss 0.9279 (0.8211) grad_norm 8.5557 (8.0873/1.8315) mem 68106MB [2022-12-20 13:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][100/1519] eta 0:24:01 lr 0.000005 time 0.9206 (1.0161) model_time 0.9204 (1.0083) loss 1.1204 (0.8235) grad_norm 9.1247 (8.1639/1.7683) mem 68106MB [2022-12-20 13:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][110/1519] eta 0:23:49 lr 0.000005 time 0.9326 (1.0146) model_time 0.9324 (1.0074) loss 0.6868 (0.8191) grad_norm 8.0862 (8.2119/1.7693) mem 68106MB [2022-12-20 13:18:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][120/1519] eta 0:23:38 lr 0.000005 time 0.9377 (1.0138) model_time 0.9376 (1.0073) loss 0.8770 (0.8251) grad_norm 7.8053 (8.2665/1.7728) mem 68106MB [2022-12-20 13:19:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][130/1519] eta 0:23:26 lr 0.000005 time 0.9321 (1.0126) model_time 0.9320 (1.0065) loss 0.9381 (0.8229) grad_norm 7.2960 (8.2409/1.7196) mem 68106MB [2022-12-20 13:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][140/1519] eta 0:23:15 lr 0.000005 time 0.9319 (1.0116) model_time 0.9316 (1.0060) loss 0.7430 (0.8217) grad_norm 6.2542 (8.2493/1.7299) mem 68106MB [2022-12-20 13:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][150/1519] eta 0:23:03 lr 0.000005 time 0.9437 (1.0108) model_time 0.9436 (1.0054) loss 1.0617 (0.8229) grad_norm 9.3299 (8.2212/1.6996) mem 68106MB [2022-12-20 13:19:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][160/1519] eta 0:22:53 lr 0.000005 time 0.9330 (1.0107) model_time 0.9329 (1.0057) loss 0.7210 (0.8201) grad_norm 6.5328 (8.1857/1.6805) mem 68106MB [2022-12-20 13:19:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][170/1519] eta 0:22:42 lr 0.000005 time 0.9472 (1.0101) model_time 0.9470 (1.0054) loss 0.8422 (0.8204) grad_norm 9.3976 (8.2165/1.6564) mem 68106MB [2022-12-20 13:19:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][180/1519] eta 0:22:33 lr 0.000005 time 0.9264 (1.0109) model_time 0.9262 (1.0064) loss 0.6890 (0.8198) grad_norm 8.5203 (8.2187/1.6241) mem 68106MB [2022-12-20 13:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][190/1519] eta 0:22:23 lr 0.000005 time 0.9304 (1.0108) model_time 0.9303 (1.0065) loss 0.6753 (0.8236) grad_norm 9.2236 (8.1947/1.6039) mem 68106MB [2022-12-20 13:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][200/1519] eta 0:22:12 lr 0.000005 time 0.9325 (1.0105) model_time 0.9322 (1.0064) loss 0.7679 (0.8259) grad_norm 8.0882 (8.2724/1.6173) mem 68106MB [2022-12-20 13:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][210/1519] eta 0:22:02 lr 0.000005 time 0.9372 (1.0107) model_time 0.9370 (1.0067) loss 0.8371 (0.8255) grad_norm 8.5525 (8.2448/1.5893) mem 68106MB [2022-12-20 13:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][220/1519] eta 0:21:52 lr 0.000005 time 0.9189 (1.0108) model_time 0.9188 (1.0070) loss 0.7850 (0.8240) grad_norm 7.1039 (8.3459/1.8031) mem 68106MB [2022-12-20 13:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][230/1519] eta 0:21:42 lr 0.000005 time 0.9326 (1.0108) model_time 0.9324 (1.0072) loss 0.7128 (0.8259) grad_norm 8.2452 (8.3245/1.7715) mem 68106MB [2022-12-20 13:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][240/1519] eta 0:21:32 lr 0.000005 time 0.9340 (1.0105) model_time 0.9338 (1.0071) loss 0.9332 (0.8224) grad_norm 7.9517 (8.2688/1.7590) mem 68106MB [2022-12-20 13:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][250/1519] eta 0:21:21 lr 0.000005 time 0.9295 (1.0102) model_time 0.9294 (1.0069) loss 0.7931 (0.8204) grad_norm 7.9066 (8.2532/1.7336) mem 68106MB [2022-12-20 13:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][260/1519] eta 0:21:12 lr 0.000005 time 1.1854 (1.0108) model_time 1.1852 (1.0075) loss 1.2379 (0.8205) grad_norm 9.2056 (8.2603/1.7964) mem 68106MB [2022-12-20 13:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][270/1519] eta 0:21:02 lr 0.000005 time 0.9270 (1.0105) model_time 0.9269 (1.0073) loss 0.7729 (0.8203) grad_norm 9.6087 (8.3009/1.7884) mem 68106MB [2022-12-20 13:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][280/1519] eta 0:20:52 lr 0.000005 time 0.9325 (1.0105) model_time 0.9323 (1.0075) loss 0.6704 (0.8220) grad_norm 9.8993 (8.2703/1.7798) mem 68106MB [2022-12-20 13:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][290/1519] eta 0:20:41 lr 0.000005 time 0.9418 (1.0102) model_time 0.9415 (1.0072) loss 0.8153 (0.8232) grad_norm 8.4093 (8.2596/1.7609) mem 68106MB [2022-12-20 13:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][300/1519] eta 0:20:31 lr 0.000005 time 0.9307 (1.0099) model_time 0.9305 (1.0070) loss 0.7142 (0.8203) grad_norm 7.4472 (8.2896/1.7579) mem 68106MB [2022-12-20 13:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][310/1519] eta 0:20:20 lr 0.000005 time 0.9371 (1.0095) model_time 0.9369 (1.0067) loss 0.7741 (0.8200) grad_norm 10.4329 (8.3167/1.7631) mem 68106MB [2022-12-20 13:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][320/1519] eta 0:20:10 lr 0.000005 time 0.9366 (1.0094) model_time 0.9364 (1.0067) loss 0.9289 (0.8211) grad_norm 7.9481 (8.3212/1.7647) mem 68106MB [2022-12-20 13:22:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][330/1519] eta 0:19:59 lr 0.000005 time 0.9265 (1.0092) model_time 0.9263 (1.0065) loss 0.6811 (0.8193) grad_norm 9.5140 (8.3199/1.7437) mem 68106MB [2022-12-20 13:22:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][340/1519] eta 0:19:49 lr 0.000005 time 0.9291 (1.0090) model_time 0.9289 (1.0064) loss 0.8155 (0.8174) grad_norm 9.0699 (8.3384/1.7269) mem 68106MB [2022-12-20 13:22:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][350/1519] eta 0:19:39 lr 0.000005 time 0.9303 (1.0089) model_time 0.9301 (1.0064) loss 0.7229 (0.8146) grad_norm 9.6234 (8.3462/1.7112) mem 68106MB [2022-12-20 13:22:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][360/1519] eta 0:19:29 lr 0.000005 time 0.9277 (1.0090) model_time 0.9276 (1.0065) loss 0.7309 (0.8148) grad_norm 8.7744 (8.3315/1.7237) mem 68106MB [2022-12-20 13:23:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][370/1519] eta 0:19:19 lr 0.000005 time 0.9314 (1.0088) model_time 0.9312 (1.0064) loss 0.7181 (0.8128) grad_norm 8.0528 (8.3703/1.7278) mem 68106MB [2022-12-20 13:23:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][380/1519] eta 0:19:08 lr 0.000005 time 0.9269 (1.0085) model_time 0.9267 (1.0062) loss 0.8462 (0.8119) grad_norm 8.0105 (8.4159/1.7646) mem 68106MB [2022-12-20 13:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][390/1519] eta 0:18:58 lr 0.000005 time 0.9296 (1.0085) model_time 0.9294 (1.0062) loss 0.7931 (0.8127) grad_norm 6.5584 (8.4021/1.7603) mem 68106MB [2022-12-20 13:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][400/1519] eta 0:18:48 lr 0.000005 time 0.9334 (1.0082) model_time 0.9331 (1.0060) loss 1.0110 (0.8125) grad_norm 7.9776 (8.4078/1.7471) mem 68106MB [2022-12-20 13:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][410/1519] eta 0:18:37 lr 0.000005 time 0.9298 (1.0080) model_time 0.9297 (1.0058) loss 0.8442 (0.8122) grad_norm 10.2062 (8.4001/1.7366) mem 68106MB [2022-12-20 13:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][420/1519] eta 0:18:28 lr 0.000005 time 0.9362 (1.0084) model_time 0.9360 (1.0063) loss 0.8335 (0.8133) grad_norm 13.5373 (8.4209/1.7606) mem 68106MB [2022-12-20 13:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][430/1519] eta 0:18:17 lr 0.000005 time 0.9322 (1.0082) model_time 0.9320 (1.0061) loss 0.7306 (0.8122) grad_norm 6.6431 (8.3869/1.7565) mem 68106MB [2022-12-20 13:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][440/1519] eta 0:18:07 lr 0.000005 time 1.0199 (1.0082) model_time 1.0197 (1.0062) loss 0.8458 (0.8112) grad_norm 6.5764 (8.3912/1.7479) mem 68106MB [2022-12-20 13:24:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][450/1519] eta 0:17:57 lr 0.000005 time 0.9295 (1.0082) model_time 0.9293 (1.0061) loss 0.6789 (0.8105) grad_norm 9.5435 (8.3847/1.7462) mem 68106MB [2022-12-20 13:24:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][460/1519] eta 0:17:47 lr 0.000005 time 0.9310 (1.0079) model_time 0.9308 (1.0059) loss 1.0650 (0.8111) grad_norm 7.1757 (8.3852/1.7400) mem 68106MB [2022-12-20 13:24:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][470/1519] eta 0:17:37 lr 0.000005 time 0.9310 (1.0078) model_time 0.9308 (1.0058) loss 1.4264 (0.8113) grad_norm 12.3219 (8.4227/1.7537) mem 68106MB [2022-12-20 13:24:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][480/1519] eta 0:17:26 lr 0.000005 time 0.9376 (1.0077) model_time 0.9374 (1.0057) loss 0.9016 (0.8115) grad_norm 8.9695 (8.4144/1.7453) mem 68106MB [2022-12-20 13:25:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][490/1519] eta 0:17:16 lr 0.000005 time 1.0323 (1.0077) model_time 1.0321 (1.0058) loss 0.6825 (0.8107) grad_norm 10.5560 (8.4181/1.7395) mem 68106MB [2022-12-20 13:25:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][500/1519] eta 0:17:06 lr 0.000005 time 0.9331 (1.0075) model_time 0.9329 (1.0057) loss 0.7650 (0.8103) grad_norm 8.6644 (8.4352/1.7479) mem 68106MB [2022-12-20 13:25:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][510/1519] eta 0:16:56 lr 0.000005 time 0.9289 (1.0075) model_time 0.9287 (1.0056) loss 0.6982 (0.8088) grad_norm 6.8981 (8.4213/1.7492) mem 68106MB [2022-12-20 13:25:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][520/1519] eta 0:16:46 lr 0.000005 time 0.9343 (1.0075) model_time 0.9341 (1.0057) loss 0.8126 (0.8095) grad_norm 11.7251 (8.5119/2.2011) mem 68106MB [2022-12-20 13:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][530/1519] eta 0:16:36 lr 0.000005 time 0.9297 (1.0074) model_time 0.9295 (1.0057) loss 0.6785 (0.8093) grad_norm 10.0015 (8.5043/2.1872) mem 68106MB [2022-12-20 13:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][540/1519] eta 0:16:26 lr 0.000005 time 0.9409 (1.0074) model_time 0.9407 (1.0057) loss 0.8628 (0.8109) grad_norm 8.5150 (8.5011/2.1720) mem 68106MB [2022-12-20 13:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][550/1519] eta 0:16:16 lr 0.000005 time 0.9330 (1.0073) model_time 0.9329 (1.0055) loss 0.8466 (0.8117) grad_norm 7.8351 (8.5735/2.3821) mem 68106MB [2022-12-20 13:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][560/1519] eta 0:16:05 lr 0.000005 time 0.9319 (1.0072) model_time 0.9317 (1.0055) loss 0.6623 (0.8117) grad_norm 9.3044 (8.5742/2.3676) mem 68106MB [2022-12-20 13:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][570/1519] eta 0:15:55 lr 0.000005 time 0.9385 (1.0071) model_time 0.9383 (1.0054) loss 0.7393 (0.8132) grad_norm 8.2771 (8.5694/2.3496) mem 68106MB [2022-12-20 13:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][580/1519] eta 0:15:45 lr 0.000005 time 0.9343 (1.0070) model_time 0.9341 (1.0053) loss 0.8397 (0.8134) grad_norm 6.1710 (8.5397/2.3429) mem 68106MB [2022-12-20 13:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][590/1519] eta 0:15:35 lr 0.000005 time 0.9402 (1.0073) model_time 0.9401 (1.0056) loss 0.6649 (0.8125) grad_norm 8.5443 (8.5251/2.3282) mem 68106MB [2022-12-20 13:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][600/1519] eta 0:15:25 lr 0.000005 time 0.9372 (1.0073) model_time 0.9370 (1.0057) loss 0.9029 (0.8121) grad_norm 11.8512 (8.5363/2.3228) mem 68106MB [2022-12-20 13:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][610/1519] eta 0:15:15 lr 0.000005 time 0.9328 (1.0072) model_time 0.9327 (1.0056) loss 0.7187 (0.8115) grad_norm 10.3350 (8.5287/2.3013) mem 68106MB [2022-12-20 13:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][620/1519] eta 0:15:05 lr 0.000005 time 0.9302 (1.0070) model_time 0.9301 (1.0055) loss 0.7429 (0.8109) grad_norm 8.4687 (8.5352/2.3109) mem 68106MB [2022-12-20 13:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][630/1519] eta 0:14:55 lr 0.000005 time 1.0504 (1.0071) model_time 1.0503 (1.0055) loss 0.6986 (0.8112) grad_norm 9.2115 (8.5529/2.3096) mem 68106MB [2022-12-20 13:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][640/1519] eta 0:14:45 lr 0.000005 time 0.9275 (1.0069) model_time 0.9274 (1.0054) loss 1.1831 (0.8116) grad_norm 8.5954 (8.5622/2.3030) mem 68106MB [2022-12-20 13:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][650/1519] eta 0:14:34 lr 0.000005 time 0.9278 (1.0069) model_time 0.9274 (1.0054) loss 0.7669 (0.8130) grad_norm 7.7723 (8.5657/2.3001) mem 68106MB [2022-12-20 13:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][660/1519] eta 0:14:24 lr 0.000005 time 0.9647 (1.0068) model_time 0.9646 (1.0053) loss 0.7724 (0.8127) grad_norm 7.2921 (8.5459/2.2905) mem 68106MB [2022-12-20 13:28:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][670/1519] eta 0:14:15 lr 0.000005 time 0.9287 (1.0072) model_time 0.9286 (1.0057) loss 0.6854 (0.8131) grad_norm 6.9137 (8.5442/2.2865) mem 68106MB [2022-12-20 13:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][680/1519] eta 0:14:04 lr 0.000005 time 0.9324 (1.0071) model_time 0.9323 (1.0057) loss 0.9184 (0.8130) grad_norm 8.0361 (8.5606/2.2866) mem 68106MB [2022-12-20 13:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][690/1519] eta 0:13:54 lr 0.000005 time 0.9332 (1.0070) model_time 0.9331 (1.0055) loss 0.8505 (0.8133) grad_norm 8.8974 (8.5641/2.2771) mem 68106MB [2022-12-20 13:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][700/1519] eta 0:13:44 lr 0.000005 time 0.9132 (1.0070) model_time 0.9131 (1.0056) loss 0.9029 (0.8137) grad_norm 13.6303 (8.5627/2.3024) mem 68106MB [2022-12-20 13:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][710/1519] eta 0:13:34 lr 0.000005 time 0.9293 (1.0072) model_time 0.9292 (1.0057) loss 0.8203 (0.8127) grad_norm 6.5523 (8.5613/2.3255) mem 68106MB [2022-12-20 13:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][720/1519] eta 0:13:24 lr 0.000005 time 0.9340 (1.0070) model_time 0.9337 (1.0056) loss 0.7954 (0.8120) grad_norm 4.8726 (8.5455/2.3300) mem 68106MB [2022-12-20 13:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][730/1519] eta 0:13:14 lr 0.000005 time 0.9128 (1.0071) model_time 0.9126 (1.0057) loss 0.7527 (0.8118) grad_norm 9.7948 (8.5668/2.3395) mem 68106MB [2022-12-20 13:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][740/1519] eta 0:13:04 lr 0.000005 time 0.9244 (1.0069) model_time 0.9242 (1.0055) loss 0.6914 (0.8121) grad_norm 8.2003 (8.5636/2.3278) mem 68106MB [2022-12-20 13:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][750/1519] eta 0:12:54 lr 0.000005 time 0.9331 (1.0069) model_time 0.9330 (1.0055) loss 0.9214 (0.8126) grad_norm 7.1061 (8.5969/2.4221) mem 68106MB [2022-12-20 13:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][760/1519] eta 0:12:44 lr 0.000005 time 0.9145 (1.0069) model_time 0.9144 (1.0056) loss 0.8804 (0.8132) grad_norm 9.2735 (8.6161/2.4171) mem 68106MB [2022-12-20 13:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][770/1519] eta 0:12:34 lr 0.000005 time 0.9313 (1.0068) model_time 0.9311 (1.0055) loss 0.8359 (0.8135) grad_norm 7.1756 (8.6112/2.4164) mem 68106MB [2022-12-20 13:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][780/1519] eta 0:12:23 lr 0.000005 time 0.9289 (1.0067) model_time 0.9288 (1.0054) loss 0.6876 (0.8133) grad_norm 12.6189 (8.6064/2.4351) mem 68106MB [2022-12-20 13:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][790/1519] eta 0:12:13 lr 0.000005 time 0.9264 (1.0066) model_time 0.9262 (1.0053) loss 0.6913 (0.8135) grad_norm 6.4010 (8.6061/2.4461) mem 68106MB [2022-12-20 13:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][800/1519] eta 0:12:03 lr 0.000005 time 0.9343 (1.0066) model_time 0.9341 (1.0052) loss 0.7395 (0.8132) grad_norm 5.9890 (8.5924/2.4536) mem 68106MB [2022-12-20 13:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][810/1519] eta 0:11:53 lr 0.000005 time 0.9206 (1.0065) model_time 0.9204 (1.0052) loss 0.7417 (0.8130) grad_norm 9.8482 (8.6090/2.4617) mem 68106MB [2022-12-20 13:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][820/1519] eta 0:11:43 lr 0.000005 time 0.9719 (1.0065) model_time 0.9718 (1.0052) loss 0.6626 (0.8128) grad_norm 10.1286 (8.5957/2.4090) mem 68106MB [2022-12-20 13:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][830/1519] eta 0:11:33 lr 0.000005 time 1.0088 (1.0065) model_time 1.0087 (1.0052) loss 0.6663 (0.8120) grad_norm 8.9331 (8.6290/2.4153) mem 68106MB [2022-12-20 13:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][840/1519] eta 0:11:23 lr 0.000005 time 0.9234 (1.0064) model_time 0.9233 (1.0051) loss 0.7221 (0.8122) grad_norm 8.6136 (8.6516/2.4056) mem 68106MB [2022-12-20 13:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][850/1519] eta 0:11:13 lr 0.000005 time 0.9235 (1.0064) model_time 0.9234 (1.0051) loss 0.7252 (0.8127) grad_norm 6.6312 (8.6580/2.4056) mem 68106MB [2022-12-20 13:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][860/1519] eta 0:11:03 lr 0.000005 time 0.9218 (1.0063) model_time 0.9216 (1.0050) loss 0.8642 (0.8125) grad_norm 10.6329 (8.6919/2.4078) mem 68106MB [2022-12-20 13:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][870/1519] eta 0:10:53 lr 0.000005 time 0.9300 (1.0063) model_time 0.9299 (1.0050) loss 0.6599 (0.8125) grad_norm 11.9131 (8.6698/2.4185) mem 68106MB [2022-12-20 13:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][880/1519] eta 0:10:43 lr 0.000005 time 0.9392 (1.0063) model_time 0.9391 (1.0051) loss 1.0584 (0.8130) grad_norm 9.5611 (8.6810/2.4156) mem 68106MB [2022-12-20 13:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][890/1519] eta 0:10:32 lr 0.000005 time 0.9401 (1.0063) model_time 0.9398 (1.0051) loss 0.6892 (0.8128) grad_norm 8.6812 (8.6828/2.4152) mem 68106MB [2022-12-20 13:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][900/1519] eta 0:10:22 lr 0.000005 time 0.8877 (1.0063) model_time 0.8876 (1.0051) loss 0.8879 (0.8137) grad_norm 7.4238 (8.6608/2.4102) mem 68106MB [2022-12-20 13:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][910/1519] eta 0:10:12 lr 0.000005 time 0.9299 (1.0063) model_time 0.9298 (1.0051) loss 1.0265 (0.8143) grad_norm 6.8933 (8.6444/2.4067) mem 68106MB [2022-12-20 13:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][920/1519] eta 0:10:02 lr 0.000005 time 0.9323 (1.0062) model_time 0.9321 (1.0050) loss 0.6709 (0.8145) grad_norm 11.2693 (8.6515/2.4053) mem 68106MB [2022-12-20 13:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][930/1519] eta 0:09:52 lr 0.000005 time 0.9304 (1.0062) model_time 0.9303 (1.0050) loss 0.8576 (0.8146) grad_norm 9.0145 (8.6484/2.4076) mem 68106MB [2022-12-20 13:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][940/1519] eta 0:09:42 lr 0.000005 time 0.9269 (1.0060) model_time 0.9267 (1.0048) loss 0.6610 (0.8146) grad_norm 6.9030 (8.6227/2.4130) mem 68106MB [2022-12-20 13:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][950/1519] eta 0:09:32 lr 0.000005 time 0.9298 (1.0060) model_time 0.9296 (1.0048) loss 0.9951 (0.8143) grad_norm 8.1464 (8.6311/2.4154) mem 68106MB [2022-12-20 13:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][960/1519] eta 0:09:22 lr 0.000005 time 1.0044 (1.0060) model_time 1.0042 (1.0048) loss 0.6781 (0.8142) grad_norm 9.6994 (8.6482/2.4048) mem 68106MB [2022-12-20 13:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][970/1519] eta 0:09:12 lr 0.000005 time 0.9162 (1.0059) model_time 0.9160 (1.0047) loss 0.7778 (0.8145) grad_norm 10.4059 (8.6441/2.4218) mem 68106MB [2022-12-20 13:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][980/1519] eta 0:09:02 lr 0.000005 time 0.9310 (1.0059) model_time 0.9308 (1.0047) loss 0.8302 (0.8147) grad_norm 13.3323 (8.6486/2.4200) mem 68106MB [2022-12-20 13:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][990/1519] eta 0:08:52 lr 0.000005 time 0.9596 (1.0060) model_time 0.9594 (1.0048) loss 0.8845 (0.8151) grad_norm 7.6680 (8.6790/2.4249) mem 68106MB [2022-12-20 13:33:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1000/1519] eta 0:08:42 lr 0.000005 time 0.9801 (1.0059) model_time 0.9800 (1.0048) loss 0.6982 (0.8146) grad_norm 10.1048 (8.7152/2.4564) mem 68106MB [2022-12-20 13:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1010/1519] eta 0:08:31 lr 0.000005 time 0.9232 (1.0058) model_time 0.9230 (1.0047) loss 0.7841 (0.8139) grad_norm 7.9358 (8.7354/2.4592) mem 68106MB [2022-12-20 13:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1020/1519] eta 0:08:21 lr 0.000005 time 0.9157 (1.0059) model_time 0.9156 (1.0047) loss 1.0741 (0.8139) grad_norm 13.5863 (8.7730/2.5051) mem 68106MB [2022-12-20 13:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1030/1519] eta 0:08:11 lr 0.000005 time 0.9378 (1.0058) model_time 0.9377 (1.0047) loss 0.7816 (0.8141) grad_norm 8.6434 (8.7718/2.4864) mem 68106MB [2022-12-20 13:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1040/1519] eta 0:08:01 lr 0.000005 time 0.9292 (1.0058) model_time 0.9290 (1.0046) loss 1.0203 (0.8147) grad_norm 6.1570 (8.7797/2.5028) mem 68106MB [2022-12-20 13:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1050/1519] eta 0:07:51 lr 0.000005 time 0.9360 (1.0060) model_time 0.9358 (1.0049) loss 0.7727 (0.8144) grad_norm 8.5836 (8.8064/2.4969) mem 68106MB [2022-12-20 13:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1060/1519] eta 0:07:41 lr 0.000005 time 0.9279 (1.0059) model_time 0.9277 (1.0048) loss 0.8103 (0.8140) grad_norm 12.0538 (8.8044/2.5086) mem 68106MB [2022-12-20 13:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1070/1519] eta 0:07:31 lr 0.000005 time 0.9311 (1.0059) model_time 0.9309 (1.0048) loss 0.8482 (0.8137) grad_norm 8.9943 (8.7917/2.5056) mem 68106MB [2022-12-20 13:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1080/1519] eta 0:07:21 lr 0.000005 time 0.9338 (1.0059) model_time 0.9335 (1.0048) loss 0.7720 (0.8135) grad_norm 8.7314 (8.7767/2.4998) mem 68106MB [2022-12-20 13:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1090/1519] eta 0:07:11 lr 0.000005 time 0.9350 (1.0058) model_time 0.9348 (1.0047) loss 0.8185 (0.8135) grad_norm 8.0443 (8.7842/2.5056) mem 68106MB [2022-12-20 13:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1100/1519] eta 0:07:01 lr 0.000005 time 0.9267 (1.0057) model_time 0.9264 (1.0047) loss 0.8408 (0.8132) grad_norm 6.8094 (8.7706/2.5046) mem 68106MB [2022-12-20 13:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1110/1519] eta 0:06:51 lr 0.000005 time 0.9340 (1.0057) model_time 0.9339 (1.0046) loss 0.7260 (0.8130) grad_norm 8.6356 (8.7669/2.5026) mem 68106MB [2022-12-20 13:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1120/1519] eta 0:06:41 lr 0.000005 time 0.9270 (1.0056) model_time 0.9269 (1.0046) loss 0.8739 (0.8132) grad_norm 8.1793 (8.7026/2.1764) mem 68106MB [2022-12-20 13:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1130/1519] eta 0:06:31 lr 0.000005 time 0.9320 (1.0059) model_time 0.9318 (1.0048) loss 0.7021 (0.8139) grad_norm 11.3915 (8.7013/2.1791) mem 68106MB [2022-12-20 13:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1140/1519] eta 0:06:21 lr 0.000005 time 0.9780 (1.0058) model_time 0.9776 (1.0048) loss 0.7700 (0.8137) grad_norm 14.4854 (8.7323/2.2009) mem 68106MB [2022-12-20 13:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1150/1519] eta 0:06:11 lr 0.000005 time 0.9313 (1.0058) model_time 0.9312 (1.0048) loss 0.7531 (0.8134) grad_norm 6.8634 (8.6631/1.9923) mem 68106MB [2022-12-20 13:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1160/1519] eta 0:06:01 lr 0.000005 time 0.9865 (1.0059) model_time 0.9863 (1.0048) loss 1.0044 (0.8138) grad_norm 6.8816 (8.6918/2.0056) mem 68106MB [2022-12-20 13:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1170/1519] eta 0:05:51 lr 0.000005 time 0.9372 (1.0059) model_time 0.9371 (1.0049) loss 0.8893 (0.8139) grad_norm 9.3846 (8.6911/2.0090) mem 68106MB [2022-12-20 13:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1180/1519] eta 0:05:40 lr 0.000005 time 0.9395 (1.0059) model_time 0.9394 (1.0048) loss 0.7412 (0.8138) grad_norm 8.7900 (8.7264/2.0133) mem 68106MB [2022-12-20 13:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1190/1519] eta 0:05:30 lr 0.000005 time 0.9364 (1.0059) model_time 0.9363 (1.0049) loss 0.8005 (0.8139) grad_norm 7.3712 (8.7483/2.0112) mem 68106MB [2022-12-20 13:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1200/1519] eta 0:05:20 lr 0.000005 time 0.9310 (1.0058) model_time 0.9309 (1.0048) loss 0.6736 (0.8135) grad_norm 7.7665 (8.7485/2.0096) mem 68106MB [2022-12-20 13:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1210/1519] eta 0:05:10 lr 0.000005 time 0.9343 (1.0058) model_time 0.9342 (1.0048) loss 0.6881 (0.8136) grad_norm 9.1058 (8.7425/2.0023) mem 68106MB [2022-12-20 13:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1220/1519] eta 0:05:00 lr 0.000005 time 0.9342 (1.0060) model_time 0.9340 (1.0050) loss 0.6754 (0.8135) grad_norm 9.5039 (8.7540/1.9841) mem 68106MB [2022-12-20 13:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1230/1519] eta 0:04:50 lr 0.000005 time 0.9319 (1.0060) model_time 0.9318 (1.0050) loss 0.6743 (0.8133) grad_norm 6.8560 (8.7460/1.9873) mem 68106MB [2022-12-20 13:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1240/1519] eta 0:04:40 lr 0.000005 time 0.9252 (1.0059) model_time 0.9251 (1.0049) loss 0.6945 (0.8131) grad_norm 10.2632 (8.7315/1.9974) mem 68106MB [2022-12-20 13:37:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1250/1519] eta 0:04:30 lr 0.000005 time 0.9242 (1.0058) model_time 0.9241 (1.0048) loss 0.7773 (0.8135) grad_norm 13.5183 (8.7868/2.0278) mem 68106MB [2022-12-20 13:37:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1260/1519] eta 0:04:20 lr 0.000005 time 0.9215 (1.0058) model_time 0.9213 (1.0048) loss 0.6861 (0.8134) grad_norm 10.5442 (8.7984/2.0304) mem 68106MB [2022-12-20 13:38:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1270/1519] eta 0:04:10 lr 0.000005 time 0.9500 (1.0057) model_time 0.9498 (1.0047) loss 0.7586 (0.8133) grad_norm 8.6674 (8.7813/2.0278) mem 68106MB [2022-12-20 13:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1280/1519] eta 0:04:00 lr 0.000005 time 0.9343 (1.0058) model_time 0.9342 (1.0048) loss 1.0009 (0.8135) grad_norm 7.4663 (8.7904/2.0185) mem 68106MB [2022-12-20 13:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1290/1519] eta 0:03:50 lr 0.000005 time 0.9332 (1.0058) model_time 0.9329 (1.0048) loss 0.9415 (0.8135) grad_norm 8.1904 (8.8049/2.0233) mem 68106MB [2022-12-20 13:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1300/1519] eta 0:03:40 lr 0.000005 time 0.9188 (1.0060) model_time 0.9187 (1.0051) loss 0.7642 (0.8129) grad_norm 7.0300 (8.8080/2.0174) mem 68106MB [2022-12-20 13:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1310/1519] eta 0:03:30 lr 0.000005 time 0.9310 (1.0061) model_time 0.9308 (1.0051) loss 0.8124 (0.8130) grad_norm 8.8757 (8.7979/1.9840) mem 68106MB [2022-12-20 13:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1320/1519] eta 0:03:20 lr 0.000005 time 0.9759 (1.0061) model_time 0.9758 (1.0051) loss 0.6780 (0.8130) grad_norm 7.5793 (8.7906/1.9845) mem 68106MB [2022-12-20 13:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1330/1519] eta 0:03:10 lr 0.000005 time 0.9279 (1.0062) model_time 0.9278 (1.0052) loss 0.6800 (0.8131) grad_norm 10.2263 (8.8234/1.9955) mem 68106MB [2022-12-20 13:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1340/1519] eta 0:03:00 lr 0.000005 time 0.9931 (1.0062) model_time 0.9929 (1.0052) loss 0.8073 (0.8133) grad_norm 6.9643 (8.8069/2.0094) mem 68106MB [2022-12-20 13:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1350/1519] eta 0:02:50 lr 0.000005 time 0.9391 (1.0062) model_time 0.9389 (1.0052) loss 0.9784 (0.8135) grad_norm 11.2313 (8.7774/1.9094) mem 68106MB [2022-12-20 13:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1360/1519] eta 0:02:39 lr 0.000005 time 0.9136 (1.0062) model_time 0.9134 (1.0053) loss 0.7287 (0.8135) grad_norm 9.8833 (8.7758/1.9200) mem 68106MB [2022-12-20 13:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1370/1519] eta 0:02:29 lr 0.000005 time 0.9295 (1.0062) model_time 0.9289 (1.0053) loss 0.9628 (0.8140) grad_norm 15.7772 (8.7804/1.9698) mem 68106MB [2022-12-20 13:39:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1380/1519] eta 0:02:19 lr 0.000005 time 0.9285 (1.0062) model_time 0.9284 (1.0052) loss 1.0469 (0.8141) grad_norm 10.5064 (8.8485/1.9805) mem 68106MB [2022-12-20 13:40:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1390/1519] eta 0:02:09 lr 0.000005 time 0.9225 (1.0062) model_time 0.9224 (1.0052) loss 1.0095 (0.8147) grad_norm 8.1742 (8.8379/1.9550) mem 68106MB [2022-12-20 13:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1400/1519] eta 0:01:59 lr 0.000005 time 0.9296 (1.0062) model_time 0.9289 (1.0053) loss 0.6882 (0.8147) grad_norm 8.3020 (8.8399/1.9709) mem 68106MB [2022-12-20 13:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1410/1519] eta 0:01:49 lr 0.000005 time 0.9423 (1.0062) model_time 0.9422 (1.0052) loss 0.9185 (0.8145) grad_norm 9.4844 (8.8897/2.0004) mem 68106MB [2022-12-20 13:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1420/1519] eta 0:01:39 lr 0.000005 time 0.9278 (1.0062) model_time 0.9277 (1.0052) loss 0.6912 (0.8145) grad_norm 7.5256 (8.9010/2.0854) mem 68106MB [2022-12-20 13:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1430/1519] eta 0:01:29 lr 0.000005 time 0.9342 (1.0061) model_time 0.9341 (1.0052) loss 0.8051 (0.8145) grad_norm 6.4638 (8.8613/2.0834) mem 68106MB [2022-12-20 13:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1440/1519] eta 0:01:19 lr 0.000005 time 0.9329 (1.0061) model_time 0.9328 (1.0052) loss 0.9502 (0.8148) grad_norm 9.0088 (8.8576/2.0869) mem 68106MB [2022-12-20 13:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1450/1519] eta 0:01:09 lr 0.000005 time 0.9297 (1.0061) model_time 0.9295 (1.0052) loss 0.8935 (0.8146) grad_norm 6.7139 (8.8472/2.1188) mem 68106MB [2022-12-20 13:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1460/1519] eta 0:00:59 lr 0.000005 time 0.9158 (1.0061) model_time 0.9157 (1.0052) loss 0.7304 (0.8148) grad_norm 7.9928 (8.8326/2.0834) mem 68106MB [2022-12-20 13:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1470/1519] eta 0:00:49 lr 0.000005 time 0.9479 (1.0061) model_time 0.9477 (1.0052) loss 0.6687 (0.8151) grad_norm 8.6889 (8.8390/2.0805) mem 68106MB [2022-12-20 13:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1480/1519] eta 0:00:39 lr 0.000005 time 0.9187 (1.0061) model_time 0.9186 (1.0052) loss 1.0021 (0.8151) grad_norm 9.0437 (8.8459/2.0819) mem 68106MB [2022-12-20 13:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1490/1519] eta 0:00:29 lr 0.000005 time 0.9463 (1.0062) model_time 0.9461 (1.0053) loss 0.9257 (0.8153) grad_norm 6.4297 (8.8292/2.0871) mem 68106MB [2022-12-20 13:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1500/1519] eta 0:00:19 lr 0.000005 time 0.9592 (1.0061) model_time 0.9591 (1.0052) loss 0.6888 (0.8154) grad_norm 8.7265 (8.8181/2.0945) mem 68106MB [2022-12-20 13:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [75/100][1510/1519] eta 0:00:09 lr 0.000005 time 0.9217 (1.0061) model_time 0.9216 (1.0052) loss 1.0295 (0.8155) grad_norm 11.1281 (8.8273/2.0977) mem 68106MB [2022-12-20 13:42:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 75 training takes 0:25:28 [2022-12-20 13:42:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_75.pth saving...... [2022-12-20 13:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_75.pth saved !!! [2022-12-20 13:42:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.660 (0.660) Loss 0.5353 (0.5353) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 13:42:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.330) Loss 0.5276 (0.5055) Acc@1 92.014 (92.677) Acc@5 97.917 (98.516) Mem 68106MB [2022-12-20 13:42:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.316) Loss 0.4945 (0.5028) Acc@1 90.278 (92.609) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 13:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.310) Loss 0.6320 (0.5098) Acc@1 90.972 (92.328) Acc@5 98.264 (98.387) Mem 68106MB [2022-12-20 13:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.303 (0.307) Loss 0.4575 (0.4997) Acc@1 93.750 (92.412) Acc@5 99.306 (98.518) Mem 68106MB [2022-12-20 13:42:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.306) Loss 0.4845 (0.4969) Acc@1 91.667 (92.470) Acc@5 99.653 (98.570) Mem 68106MB [2022-12-20 13:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.5162 (0.4967) Acc@1 90.972 (92.435) Acc@5 98.264 (98.537) Mem 68106MB [2022-12-20 13:43:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5495 (0.4977) Acc@1 92.708 (92.410) Acc@5 97.917 (98.538) Mem 68106MB [2022-12-20 13:43:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.297 (0.303) Loss 0.4347 (0.4963) Acc@1 93.056 (92.460) Acc@5 98.264 (98.568) Mem 68106MB [2022-12-20 13:43:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:75] * Acc@1 92.428 Acc@5 98.576 [2022-12-20 13:43:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 13:43:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 13:43:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][0/1519] eta 0:45:08 lr 0.000005 time 1.7833 (1.7833) model_time 1.1226 (1.1226) loss 0.6598 (0.6598) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 13:43:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][10/1519] eta 0:26:55 lr 0.000005 time 0.9236 (1.0704) model_time 0.9235 (1.0099) loss 0.8431 (0.7492) grad_norm 11.8342 (8.3791/1.8105) mem 68106MB [2022-12-20 13:43:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][20/1519] eta 0:25:51 lr 0.000005 time 0.9216 (1.0354) model_time 0.9214 (1.0035) loss 0.6751 (0.7692) grad_norm 9.8040 (7.8658/1.6572) mem 68106MB [2022-12-20 13:43:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][30/1519] eta 0:25:24 lr 0.000005 time 0.9205 (1.0242) model_time 0.9204 (1.0025) loss 0.9736 (0.7725) grad_norm 8.3574 (8.1029/2.0613) mem 68106MB [2022-12-20 13:43:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][40/1519] eta 0:25:08 lr 0.000005 time 0.9308 (1.0197) model_time 0.9307 (1.0032) loss 0.8100 (0.7781) grad_norm 8.4008 (8.2588/1.8570) mem 68106MB [2022-12-20 13:43:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][50/1519] eta 0:24:52 lr 0.000005 time 0.9232 (1.0159) model_time 0.9230 (1.0026) loss 0.7023 (0.7798) grad_norm 8.4422 (8.3429/1.8236) mem 68106MB [2022-12-20 13:44:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][60/1519] eta 0:24:41 lr 0.000005 time 0.9234 (1.0155) model_time 0.9233 (1.0044) loss 0.7279 (0.7770) grad_norm 9.5387 (8.3288/1.7081) mem 68106MB [2022-12-20 13:44:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][70/1519] eta 0:24:28 lr 0.000005 time 0.9218 (1.0138) model_time 0.9217 (1.0041) loss 0.7864 (0.7725) grad_norm 7.3549 (8.4495/1.6968) mem 68106MB [2022-12-20 13:44:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][80/1519] eta 0:24:16 lr 0.000005 time 0.9298 (1.0123) model_time 0.9296 (1.0038) loss 0.6609 (0.7735) grad_norm 8.5846 (8.4424/1.6994) mem 68106MB [2022-12-20 13:44:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][90/1519] eta 0:24:05 lr 0.000005 time 0.9227 (1.0113) model_time 0.9225 (1.0037) loss 0.9393 (0.7728) grad_norm 8.0602 (8.4897/1.8433) mem 68106MB [2022-12-20 13:44:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][100/1519] eta 0:23:54 lr 0.000005 time 0.9912 (1.0108) model_time 0.9910 (1.0039) loss 0.8122 (0.7762) grad_norm 7.2456 (8.4273/1.8035) mem 68106MB [2022-12-20 13:44:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][110/1519] eta 0:23:42 lr 0.000005 time 0.9221 (1.0095) model_time 0.9220 (1.0033) loss 0.8076 (0.7780) grad_norm 7.7225 (8.4145/1.7501) mem 68106MB [2022-12-20 13:45:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][120/1519] eta 0:23:32 lr 0.000005 time 0.9196 (1.0094) model_time 0.9195 (1.0036) loss 0.8127 (0.7783) grad_norm 7.4118 (8.4599/1.7021) mem 68106MB [2022-12-20 13:45:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][130/1519] eta 0:23:21 lr 0.000005 time 0.9235 (1.0091) model_time 0.9234 (1.0037) loss 0.9285 (0.7788) grad_norm 8.0136 (8.4029/1.6496) mem 68106MB [2022-12-20 13:45:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][140/1519] eta 0:23:11 lr 0.000005 time 0.9354 (1.0091) model_time 0.9351 (1.0041) loss 1.1171 (0.7857) grad_norm 9.1631 (8.4092/1.6513) mem 68106MB [2022-12-20 13:45:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][150/1519] eta 0:23:00 lr 0.000005 time 0.9708 (1.0087) model_time 0.9706 (1.0040) loss 0.6790 (0.7863) grad_norm 6.4032 (8.4434/1.7114) mem 68106MB [2022-12-20 13:45:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][160/1519] eta 0:22:49 lr 0.000005 time 0.9324 (1.0079) model_time 0.9322 (1.0035) loss 0.6909 (0.7899) grad_norm 6.9285 (8.4403/1.6857) mem 68106MB [2022-12-20 13:45:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][170/1519] eta 0:22:39 lr 0.000005 time 0.9230 (1.0075) model_time 0.9228 (1.0033) loss 0.8233 (0.7877) grad_norm 5.4620 (8.3977/1.7168) mem 68106MB [2022-12-20 13:46:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][180/1519] eta 0:22:28 lr 0.000005 time 0.9307 (1.0069) model_time 0.9305 (1.0029) loss 0.6551 (0.7843) grad_norm 7.7293 (8.3719/1.6734) mem 68106MB [2022-12-20 13:46:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][190/1519] eta 0:22:18 lr 0.000005 time 1.0113 (1.0069) model_time 1.0112 (1.0031) loss 0.6978 (0.7837) grad_norm 5.8727 (8.4449/1.9069) mem 68106MB [2022-12-20 13:46:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][200/1519] eta 0:22:07 lr 0.000005 time 0.9203 (1.0064) model_time 0.9201 (1.0027) loss 0.7354 (0.7821) grad_norm 7.1403 (8.3685/1.8896) mem 68106MB [2022-12-20 13:46:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][210/1519] eta 0:21:56 lr 0.000005 time 0.9210 (1.0061) model_time 0.9208 (1.0026) loss 0.7073 (0.7823) grad_norm 14.6869 (8.3781/1.9658) mem 68106MB [2022-12-20 13:46:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][220/1519] eta 0:21:46 lr 0.000005 time 0.9254 (1.0057) model_time 0.9252 (1.0023) loss 1.0608 (0.7865) grad_norm 8.9064 (8.3595/1.9353) mem 68106MB [2022-12-20 13:46:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][230/1519] eta 0:21:36 lr 0.000005 time 0.9333 (1.0054) model_time 0.9332 (1.0022) loss 0.8027 (0.7870) grad_norm 9.0015 (8.4122/1.9156) mem 68106MB [2022-12-20 13:47:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][240/1519] eta 0:21:25 lr 0.000005 time 0.9371 (1.0051) model_time 0.9368 (1.0021) loss 0.6971 (0.7883) grad_norm 14.5802 (8.5496/2.0580) mem 68106MB [2022-12-20 13:47:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][250/1519] eta 0:21:15 lr 0.000005 time 0.9283 (1.0051) model_time 0.9281 (1.0021) loss 1.0176 (0.7899) grad_norm 12.3968 (8.5540/2.0578) mem 68106MB [2022-12-20 13:47:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][260/1519] eta 0:21:04 lr 0.000005 time 0.9304 (1.0048) model_time 0.9302 (1.0019) loss 0.9978 (0.7937) grad_norm 10.4126 (8.6390/2.0850) mem 68106MB [2022-12-20 13:47:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][270/1519] eta 0:20:55 lr 0.000005 time 0.9324 (1.0052) model_time 0.9323 (1.0024) loss 0.6617 (0.7975) grad_norm 10.1921 (8.6592/2.0526) mem 68106MB [2022-12-20 13:47:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][280/1519] eta 0:20:46 lr 0.000005 time 1.1814 (1.0060) model_time 1.1813 (1.0033) loss 0.6956 (0.7971) grad_norm 8.4632 (8.6141/2.0380) mem 68106MB [2022-12-20 13:47:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][290/1519] eta 0:20:35 lr 0.000005 time 0.9291 (1.0057) model_time 0.9290 (1.0031) loss 0.8928 (0.8007) grad_norm 9.3711 (8.6196/2.0157) mem 68106MB [2022-12-20 13:48:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][300/1519] eta 0:20:26 lr 0.000005 time 0.9269 (1.0061) model_time 0.9267 (1.0036) loss 0.9579 (0.8028) grad_norm 9.2423 (8.6375/1.9991) mem 68106MB [2022-12-20 13:48:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][310/1519] eta 0:20:16 lr 0.000005 time 0.9261 (1.0059) model_time 0.9260 (1.0034) loss 0.7044 (0.8014) grad_norm 8.6447 (8.6542/1.9976) mem 68106MB [2022-12-20 13:48:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][320/1519] eta 0:20:06 lr 0.000005 time 0.9232 (1.0059) model_time 0.9230 (1.0035) loss 0.6857 (0.7987) grad_norm 7.9616 (8.6304/1.9802) mem 68106MB [2022-12-20 13:48:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][330/1519] eta 0:19:55 lr 0.000005 time 0.9247 (1.0057) model_time 0.9245 (1.0033) loss 0.8359 (0.7981) grad_norm 11.8838 (8.6341/1.9716) mem 68106MB [2022-12-20 13:48:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][340/1519] eta 0:19:45 lr 0.000005 time 1.0700 (1.0059) model_time 1.0698 (1.0036) loss 0.8380 (0.7979) grad_norm 8.1481 (8.6532/1.9853) mem 68106MB [2022-12-20 13:48:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][350/1519] eta 0:19:35 lr 0.000005 time 0.8870 (1.0058) model_time 0.8868 (1.0036) loss 0.7476 (0.7987) grad_norm 8.8345 (8.6462/1.9604) mem 68106MB [2022-12-20 13:49:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][360/1519] eta 0:19:25 lr 0.000005 time 0.9202 (1.0056) model_time 0.9200 (1.0034) loss 0.6579 (0.7975) grad_norm 15.1861 (8.7129/2.0319) mem 68106MB [2022-12-20 13:49:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][370/1519] eta 0:19:15 lr 0.000005 time 1.0204 (1.0059) model_time 1.0202 (1.0037) loss 0.7013 (0.7967) grad_norm 13.0648 (8.7633/2.0719) mem 68106MB [2022-12-20 13:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][380/1519] eta 0:19:05 lr 0.000005 time 0.9229 (1.0060) model_time 0.9227 (1.0039) loss 0.6904 (0.7942) grad_norm 9.6692 (8.7585/2.0480) mem 68106MB [2022-12-20 13:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][390/1519] eta 0:18:55 lr 0.000005 time 0.9297 (1.0060) model_time 0.9296 (1.0040) loss 0.6794 (0.7951) grad_norm 13.3923 (8.8069/2.0753) mem 68106MB [2022-12-20 13:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][400/1519] eta 0:18:45 lr 0.000005 time 0.9162 (1.0059) model_time 0.9161 (1.0039) loss 0.6752 (0.7935) grad_norm 14.0084 (8.8376/2.1084) mem 68106MB [2022-12-20 13:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][410/1519] eta 0:18:35 lr 0.000005 time 0.9307 (1.0058) model_time 0.9306 (1.0038) loss 0.7391 (0.7935) grad_norm 8.5014 (8.8379/2.0916) mem 68106MB [2022-12-20 13:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][420/1519] eta 0:18:25 lr 0.000005 time 0.9318 (1.0056) model_time 0.9317 (1.0037) loss 1.0880 (0.7931) grad_norm 9.2108 (8.8309/2.0809) mem 68106MB [2022-12-20 13:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][430/1519] eta 0:18:15 lr 0.000005 time 0.9308 (1.0057) model_time 0.9306 (1.0038) loss 0.6583 (0.7924) grad_norm 8.5434 (8.8205/2.0711) mem 68106MB [2022-12-20 13:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][440/1519] eta 0:18:05 lr 0.000005 time 0.9311 (1.0056) model_time 0.9310 (1.0038) loss 0.8145 (0.7923) grad_norm 8.4456 (8.8487/2.0653) mem 68106MB [2022-12-20 13:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][450/1519] eta 0:17:55 lr 0.000005 time 0.9228 (1.0060) model_time 0.9226 (1.0041) loss 0.6571 (0.7927) grad_norm 7.4046 (8.8255/2.0510) mem 68106MB [2022-12-20 13:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][460/1519] eta 0:17:45 lr 0.000005 time 0.9332 (1.0059) model_time 0.9329 (1.0041) loss 0.6927 (0.7940) grad_norm 7.9681 (8.8236/2.0365) mem 68106MB [2022-12-20 13:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][470/1519] eta 0:17:35 lr 0.000005 time 0.9357 (1.0059) model_time 0.9355 (1.0041) loss 0.7293 (0.7939) grad_norm 8.4554 (8.8109/2.0177) mem 68106MB [2022-12-20 13:51:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][480/1519] eta 0:17:25 lr 0.000005 time 0.9244 (1.0058) model_time 0.9242 (1.0041) loss 1.0687 (0.7952) grad_norm 7.1409 (8.8054/2.0120) mem 68106MB [2022-12-20 13:51:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][490/1519] eta 0:17:14 lr 0.000005 time 0.9340 (1.0056) model_time 0.9338 (1.0039) loss 0.8136 (0.7950) grad_norm 10.9147 (8.8098/2.0035) mem 68106MB [2022-12-20 13:51:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][500/1519] eta 0:17:04 lr 0.000005 time 0.9276 (1.0055) model_time 0.9273 (1.0038) loss 0.6787 (0.7954) grad_norm 8.2884 (8.8261/1.9925) mem 68106MB [2022-12-20 13:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][510/1519] eta 0:16:54 lr 0.000005 time 0.9315 (1.0054) model_time 0.9314 (1.0038) loss 0.9717 (0.7957) grad_norm 7.3376 (8.7886/1.9920) mem 68106MB [2022-12-20 13:51:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][520/1519] eta 0:16:44 lr 0.000005 time 0.9974 (1.0054) model_time 0.9973 (1.0038) loss 0.6601 (0.7951) grad_norm 7.2052 (8.7735/1.9776) mem 68106MB [2022-12-20 13:52:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][530/1519] eta 0:16:34 lr 0.000005 time 0.9278 (1.0052) model_time 0.9277 (1.0036) loss 0.8159 (0.7956) grad_norm 8.5304 (8.7897/1.9693) mem 68106MB [2022-12-20 13:52:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][540/1519] eta 0:16:23 lr 0.000005 time 0.9323 (1.0051) model_time 0.9322 (1.0035) loss 0.7689 (0.7949) grad_norm 10.2082 (8.7869/1.9679) mem 68106MB [2022-12-20 13:52:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][550/1519] eta 0:16:13 lr 0.000005 time 0.9372 (1.0050) model_time 0.9371 (1.0034) loss 0.7648 (0.7952) grad_norm 9.6309 (8.7804/1.9610) mem 68106MB [2022-12-20 13:52:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][560/1519] eta 0:16:03 lr 0.000005 time 0.9452 (1.0050) model_time 0.9451 (1.0034) loss 0.7070 (0.7950) grad_norm 8.8038 (8.7860/1.9507) mem 68106MB [2022-12-20 13:52:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][570/1519] eta 0:15:54 lr 0.000005 time 0.9278 (1.0054) model_time 0.9276 (1.0039) loss 0.8282 (0.7957) grad_norm 5.7140 (8.7846/1.9537) mem 68106MB [2022-12-20 13:52:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][580/1519] eta 0:15:44 lr 0.000005 time 1.0075 (1.0055) model_time 1.0074 (1.0040) loss 0.8728 (0.7962) grad_norm 10.5835 (8.7648/1.9526) mem 68106MB [2022-12-20 13:53:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][590/1519] eta 0:15:33 lr 0.000005 time 0.9306 (1.0053) model_time 0.9301 (1.0038) loss 0.6764 (0.7960) grad_norm 10.8063 (8.7612/1.9514) mem 68106MB [2022-12-20 13:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][600/1519] eta 0:15:23 lr 0.000005 time 0.9375 (1.0054) model_time 0.9373 (1.0039) loss 0.7465 (0.7964) grad_norm 11.8170 (8.7593/1.9563) mem 68106MB [2022-12-20 13:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][610/1519] eta 0:15:14 lr 0.000005 time 0.9166 (1.0055) model_time 0.9165 (1.0041) loss 1.0044 (0.7963) grad_norm 8.1175 (8.7823/1.9676) mem 68106MB [2022-12-20 13:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][620/1519] eta 0:15:03 lr 0.000005 time 0.9330 (1.0054) model_time 0.9328 (1.0040) loss 1.0002 (0.7956) grad_norm 7.0075 (8.7774/1.9783) mem 68106MB [2022-12-20 13:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][630/1519] eta 0:14:53 lr 0.000005 time 0.9310 (1.0056) model_time 0.9308 (1.0042) loss 0.6836 (0.7963) grad_norm 9.6608 (8.7956/1.9993) mem 68106MB [2022-12-20 13:53:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][640/1519] eta 0:14:43 lr 0.000005 time 0.9284 (1.0055) model_time 0.9282 (1.0041) loss 0.8897 (0.7955) grad_norm 6.2696 (8.7627/2.0170) mem 68106MB [2022-12-20 13:54:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][650/1519] eta 0:14:33 lr 0.000005 time 0.9270 (1.0053) model_time 0.9269 (1.0039) loss 1.1111 (0.7970) grad_norm 9.4766 (8.7756/2.0574) mem 68106MB [2022-12-20 13:54:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][660/1519] eta 0:14:23 lr 0.000005 time 0.9360 (1.0052) model_time 0.9359 (1.0039) loss 0.8177 (0.7956) grad_norm 8.1575 (8.7683/2.0573) mem 68106MB [2022-12-20 13:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][670/1519] eta 0:14:13 lr 0.000005 time 0.9244 (1.0052) model_time 0.9243 (1.0038) loss 0.7514 (0.7966) grad_norm 13.3286 (8.7979/2.0933) mem 68106MB [2022-12-20 13:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][680/1519] eta 0:14:03 lr 0.000005 time 0.9237 (1.0052) model_time 0.9236 (1.0039) loss 0.8781 (0.7972) grad_norm 11.4258 (8.8005/2.0929) mem 68106MB [2022-12-20 13:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][690/1519] eta 0:13:53 lr 0.000005 time 0.9298 (1.0053) model_time 0.9296 (1.0040) loss 0.9290 (0.7971) grad_norm 6.7007 (8.7911/2.0951) mem 68106MB [2022-12-20 13:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][700/1519] eta 0:13:43 lr 0.000005 time 0.9200 (1.0052) model_time 0.9198 (1.0039) loss 0.7101 (0.7970) grad_norm 11.6108 (8.7994/2.0970) mem 68106MB [2022-12-20 13:55:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][710/1519] eta 0:13:33 lr 0.000005 time 0.9375 (1.0051) model_time 0.9374 (1.0038) loss 0.8373 (0.7976) grad_norm 7.6875 (8.7793/2.1049) mem 68106MB [2022-12-20 13:55:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][720/1519] eta 0:13:23 lr 0.000005 time 0.8951 (1.0052) model_time 0.8950 (1.0039) loss 0.6761 (0.7972) grad_norm 8.9765 (8.7591/2.1075) mem 68106MB [2022-12-20 13:55:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][730/1519] eta 0:13:13 lr 0.000005 time 0.9254 (1.0051) model_time 0.9253 (1.0038) loss 0.7796 (0.7974) grad_norm 8.4139 (8.7669/2.1090) mem 68106MB [2022-12-20 13:55:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][740/1519] eta 0:13:02 lr 0.000005 time 0.9120 (1.0051) model_time 0.9118 (1.0038) loss 0.9101 (0.7976) grad_norm 8.2840 (8.7581/2.1022) mem 68106MB [2022-12-20 13:55:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][750/1519] eta 0:12:53 lr 0.000005 time 1.1898 (1.0053) model_time 1.1896 (1.0041) loss 0.6544 (0.7977) grad_norm 7.0827 (8.7363/2.0894) mem 68106MB [2022-12-20 13:55:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][760/1519] eta 0:12:43 lr 0.000005 time 0.9185 (1.0053) model_time 0.9183 (1.0040) loss 0.9462 (0.7976) grad_norm 9.2226 (8.7375/2.0884) mem 68106MB [2022-12-20 13:56:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][770/1519] eta 0:12:33 lr 0.000005 time 0.9150 (1.0053) model_time 0.9143 (1.0041) loss 0.8452 (0.7975) grad_norm 8.3370 (8.7521/2.0737) mem 68106MB [2022-12-20 13:56:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][780/1519] eta 0:12:22 lr 0.000005 time 0.9382 (1.0053) model_time 0.9381 (1.0041) loss 0.6785 (0.7967) grad_norm 8.0163 (8.7546/2.0753) mem 68106MB [2022-12-20 13:56:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][790/1519] eta 0:12:12 lr 0.000005 time 0.9259 (1.0052) model_time 0.9257 (1.0040) loss 0.6671 (0.7973) grad_norm 10.7974 (8.7374/2.0069) mem 68106MB [2022-12-20 13:56:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][800/1519] eta 0:12:02 lr 0.000005 time 0.9392 (1.0052) model_time 0.9391 (1.0040) loss 1.0443 (0.7990) grad_norm 9.1269 (8.7459/2.0047) mem 68106MB [2022-12-20 13:56:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][810/1519] eta 0:11:52 lr 0.000005 time 0.9310 (1.0051) model_time 0.9308 (1.0039) loss 0.7471 (0.7989) grad_norm 10.9988 (8.7615/1.9761) mem 68106MB [2022-12-20 13:56:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][820/1519] eta 0:11:42 lr 0.000005 time 0.9351 (1.0051) model_time 0.9350 (1.0039) loss 0.7558 (0.7987) grad_norm 6.0191 (8.7549/1.9779) mem 68106MB [2022-12-20 13:57:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][830/1519] eta 0:11:32 lr 0.000005 time 0.9291 (1.0050) model_time 0.9284 (1.0038) loss 0.7041 (0.7983) grad_norm 7.3526 (8.7229/1.9786) mem 68106MB [2022-12-20 13:57:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][840/1519] eta 0:11:22 lr 0.000005 time 0.9280 (1.0050) model_time 0.9278 (1.0039) loss 0.6731 (0.7980) grad_norm 7.1512 (8.6585/1.9178) mem 68106MB [2022-12-20 13:57:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][850/1519] eta 0:11:12 lr 0.000005 time 0.9208 (1.0050) model_time 0.9206 (1.0039) loss 1.1375 (0.7980) grad_norm 7.5010 (8.6747/1.9286) mem 68106MB [2022-12-20 13:57:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][860/1519] eta 0:11:02 lr 0.000005 time 0.9352 (1.0049) model_time 0.9349 (1.0038) loss 0.7644 (0.7979) grad_norm 7.8336 (8.6289/1.9065) mem 68106MB [2022-12-20 13:57:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][870/1519] eta 0:10:52 lr 0.000005 time 0.9158 (1.0049) model_time 0.9157 (1.0038) loss 1.0135 (0.7982) grad_norm 7.8251 (8.6070/1.9064) mem 68106MB [2022-12-20 13:57:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][880/1519] eta 0:10:42 lr 0.000005 time 0.9365 (1.0049) model_time 0.9363 (1.0037) loss 0.8865 (0.7984) grad_norm 5.8196 (8.6065/1.9112) mem 68106MB [2022-12-20 13:58:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][890/1519] eta 0:10:32 lr 0.000005 time 0.9302 (1.0049) model_time 0.9301 (1.0037) loss 0.8058 (0.7988) grad_norm 8.8270 (8.6430/2.0701) mem 68106MB [2022-12-20 13:58:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][900/1519] eta 0:10:22 lr 0.000005 time 0.9230 (1.0049) model_time 0.9229 (1.0038) loss 0.8309 (0.7985) grad_norm 7.9402 (8.6153/2.0679) mem 68106MB [2022-12-20 13:58:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][910/1519] eta 0:10:11 lr 0.000005 time 0.9277 (1.0048) model_time 0.9275 (1.0037) loss 0.9153 (0.7986) grad_norm 8.0259 (8.5991/2.0624) mem 68106MB [2022-12-20 13:58:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][920/1519] eta 0:10:01 lr 0.000005 time 0.9514 (1.0048) model_time 0.9511 (1.0037) loss 0.6880 (0.7979) grad_norm 5.7896 (8.6040/2.0860) mem 68106MB [2022-12-20 13:58:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][930/1519] eta 0:09:51 lr 0.000005 time 1.0214 (1.0049) model_time 1.0212 (1.0038) loss 0.8727 (0.7977) grad_norm 8.3870 (8.5853/2.0949) mem 68106MB [2022-12-20 13:58:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][940/1519] eta 0:09:42 lr 0.000005 time 0.9788 (1.0052) model_time 0.9786 (1.0041) loss 0.6991 (0.7983) grad_norm 8.4646 (8.5667/2.0729) mem 68106MB [2022-12-20 13:59:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][950/1519] eta 0:09:31 lr 0.000005 time 0.9331 (1.0051) model_time 0.9329 (1.0041) loss 0.7585 (0.7988) grad_norm 8.8700 (8.5671/2.0772) mem 68106MB [2022-12-20 13:59:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][960/1519] eta 0:09:21 lr 0.000005 time 0.9297 (1.0051) model_time 0.9295 (1.0040) loss 0.9908 (0.7992) grad_norm 7.0366 (8.5019/2.0297) mem 68106MB [2022-12-20 13:59:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][970/1519] eta 0:09:11 lr 0.000005 time 0.9305 (1.0050) model_time 0.9304 (1.0039) loss 0.7023 (0.7993) grad_norm 8.2548 (8.4544/1.9890) mem 68106MB [2022-12-20 13:59:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][980/1519] eta 0:09:01 lr 0.000005 time 0.9342 (1.0049) model_time 0.9341 (1.0039) loss 0.9917 (0.7989) grad_norm 8.2564 (8.4512/2.0010) mem 68106MB [2022-12-20 13:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][990/1519] eta 0:08:51 lr 0.000005 time 1.0231 (1.0049) model_time 1.0229 (1.0039) loss 0.7906 (0.7994) grad_norm 9.8069 (8.4185/1.9672) mem 68106MB [2022-12-20 13:59:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1000/1519] eta 0:08:41 lr 0.000005 time 0.9302 (1.0049) model_time 0.9301 (1.0039) loss 0.7056 (0.7996) grad_norm 9.2091 (8.3895/1.9322) mem 68106MB [2022-12-20 14:00:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1010/1519] eta 0:08:31 lr 0.000005 time 0.9315 (1.0049) model_time 0.9314 (1.0039) loss 0.6812 (0.7988) grad_norm 9.6021 (8.4096/1.9476) mem 68106MB [2022-12-20 14:00:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1020/1519] eta 0:08:21 lr 0.000005 time 0.9358 (1.0048) model_time 0.9356 (1.0038) loss 0.6983 (0.7986) grad_norm 11.8544 (8.3975/1.9581) mem 68106MB [2022-12-20 14:00:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1030/1519] eta 0:08:11 lr 0.000005 time 0.9317 (1.0048) model_time 0.9315 (1.0038) loss 0.7459 (0.7982) grad_norm 7.4779 (8.3803/1.9585) mem 68106MB [2022-12-20 14:00:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1040/1519] eta 0:08:01 lr 0.000005 time 0.9334 (1.0048) model_time 0.9333 (1.0037) loss 0.6841 (0.7980) grad_norm 8.0314 (8.3361/1.9500) mem 68106MB [2022-12-20 14:00:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1050/1519] eta 0:07:51 lr 0.000005 time 0.9301 (1.0047) model_time 0.9300 (1.0037) loss 0.7936 (0.7981) grad_norm 7.6766 (8.3415/1.9505) mem 68106MB [2022-12-20 14:00:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1060/1519] eta 0:07:41 lr 0.000005 time 0.9394 (1.0047) model_time 0.9393 (1.0037) loss 0.9753 (0.7989) grad_norm 8.4291 (8.3309/1.9453) mem 68106MB [2022-12-20 14:01:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1070/1519] eta 0:07:31 lr 0.000005 time 1.0055 (1.0047) model_time 1.0054 (1.0037) loss 0.7191 (0.7986) grad_norm 7.7117 (8.3464/1.9628) mem 68106MB [2022-12-20 14:01:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1080/1519] eta 0:07:21 lr 0.000005 time 0.9315 (1.0048) model_time 0.9313 (1.0039) loss 0.9707 (0.7985) grad_norm 7.7293 (8.3446/1.9560) mem 68106MB [2022-12-20 14:01:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1090/1519] eta 0:07:11 lr 0.000005 time 0.9327 (1.0048) model_time 0.9325 (1.0039) loss 0.9691 (0.7990) grad_norm 8.3498 (8.3260/1.9537) mem 68106MB [2022-12-20 14:01:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1100/1519] eta 0:07:01 lr 0.000005 time 0.9332 (1.0048) model_time 0.9330 (1.0038) loss 0.6813 (0.7984) grad_norm 13.6091 (8.3194/1.9670) mem 68106MB [2022-12-20 14:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1110/1519] eta 0:06:50 lr 0.000005 time 0.9276 (1.0048) model_time 0.9275 (1.0038) loss 0.8653 (0.7992) grad_norm 8.5233 (8.3479/1.9604) mem 68106MB [2022-12-20 14:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1120/1519] eta 0:06:40 lr 0.000005 time 0.9325 (1.0047) model_time 0.9324 (1.0038) loss 1.0811 (0.7992) grad_norm 8.1891 (8.3555/1.9620) mem 68106MB [2022-12-20 14:02:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1130/1519] eta 0:06:30 lr 0.000005 time 0.9318 (1.0047) model_time 0.9316 (1.0037) loss 0.6761 (0.7990) grad_norm 8.1374 (8.3507/1.9654) mem 68106MB [2022-12-20 14:02:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1140/1519] eta 0:06:20 lr 0.000005 time 0.9348 (1.0046) model_time 0.9347 (1.0037) loss 0.6548 (0.7991) grad_norm 7.3846 (8.3451/1.9564) mem 68106MB [2022-12-20 14:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1150/1519] eta 0:06:10 lr 0.000005 time 0.9296 (1.0046) model_time 0.9294 (1.0036) loss 0.6804 (0.7989) grad_norm 9.5044 (8.3601/1.9621) mem 68106MB [2022-12-20 14:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1160/1519] eta 0:06:00 lr 0.000005 time 0.9427 (1.0046) model_time 0.9425 (1.0037) loss 0.7707 (0.7988) grad_norm 8.2890 (8.3343/1.9581) mem 68106MB [2022-12-20 14:02:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1170/1519] eta 0:05:50 lr 0.000005 time 1.0373 (1.0047) model_time 1.0372 (1.0037) loss 0.6994 (0.7985) grad_norm 8.8430 (8.3181/1.9442) mem 68106MB [2022-12-20 14:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1180/1519] eta 0:05:40 lr 0.000005 time 0.9249 (1.0046) model_time 0.9247 (1.0037) loss 0.7630 (0.7982) grad_norm 7.6044 (8.3303/1.9338) mem 68106MB [2022-12-20 14:03:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1190/1519] eta 0:05:30 lr 0.000005 time 0.9336 (1.0046) model_time 0.9334 (1.0037) loss 0.6613 (0.7982) grad_norm 9.2129 (8.3465/1.9403) mem 68106MB [2022-12-20 14:03:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1200/1519] eta 0:05:20 lr 0.000005 time 0.9298 (1.0048) model_time 0.9295 (1.0039) loss 0.6596 (0.7981) grad_norm 7.5080 (8.3397/1.9209) mem 68106MB [2022-12-20 14:03:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1210/1519] eta 0:05:10 lr 0.000005 time 0.9316 (1.0048) model_time 0.9315 (1.0039) loss 0.9837 (0.7985) grad_norm 10.1720 (8.3119/1.9038) mem 68106MB [2022-12-20 14:03:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1220/1519] eta 0:05:00 lr 0.000005 time 0.9272 (1.0048) model_time 0.9270 (1.0039) loss 0.8091 (0.7980) grad_norm 10.6570 (8.3409/1.8934) mem 68106MB [2022-12-20 14:03:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1230/1519] eta 0:04:50 lr 0.000005 time 0.9391 (1.0048) model_time 0.9389 (1.0038) loss 0.6986 (0.7981) grad_norm 9.4129 (8.3160/1.8382) mem 68106MB [2022-12-20 14:03:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1240/1519] eta 0:04:40 lr 0.000005 time 0.9211 (1.0048) model_time 0.9209 (1.0039) loss 0.7312 (0.7984) grad_norm 7.1579 (8.3434/1.8277) mem 68106MB [2022-12-20 14:04:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1250/1519] eta 0:04:30 lr 0.000005 time 1.0122 (1.0049) model_time 1.0120 (1.0040) loss 0.6497 (0.7984) grad_norm 11.5168 (8.3313/1.7819) mem 68106MB [2022-12-20 14:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1260/1519] eta 0:04:20 lr 0.000005 time 0.9322 (1.0049) model_time 0.9320 (1.0040) loss 0.8624 (0.7987) grad_norm 10.4028 (8.3349/1.7879) mem 68106MB [2022-12-20 14:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1270/1519] eta 0:04:10 lr 0.000005 time 0.9300 (1.0049) model_time 0.9299 (1.0040) loss 0.8537 (0.7990) grad_norm 8.2965 (8.2980/1.7379) mem 68106MB [2022-12-20 14:04:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1280/1519] eta 0:04:00 lr 0.000005 time 0.9353 (1.0048) model_time 0.9352 (1.0039) loss 0.9042 (0.7991) grad_norm 9.7856 (8.3118/1.7415) mem 68106MB [2022-12-20 14:04:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1290/1519] eta 0:03:50 lr 0.000005 time 0.9463 (1.0048) model_time 0.9461 (1.0039) loss 0.8440 (0.7994) grad_norm 9.8585 (8.3094/1.7084) mem 68106MB [2022-12-20 14:04:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1300/1519] eta 0:03:40 lr 0.000005 time 0.9352 (1.0047) model_time 0.9350 (1.0039) loss 0.6901 (0.7998) grad_norm 7.8048 (8.3074/1.7015) mem 68106MB [2022-12-20 14:05:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1310/1519] eta 0:03:29 lr 0.000005 time 0.9228 (1.0048) model_time 0.9225 (1.0039) loss 0.6999 (0.7994) grad_norm 9.0497 (8.3475/1.7295) mem 68106MB [2022-12-20 14:05:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1320/1519] eta 0:03:19 lr 0.000005 time 0.9372 (1.0047) model_time 0.9371 (1.0038) loss 0.7046 (0.7995) grad_norm 10.8977 (8.3525/1.7370) mem 68106MB [2022-12-20 14:05:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1330/1519] eta 0:03:09 lr 0.000005 time 0.9346 (1.0047) model_time 0.9344 (1.0038) loss 0.6650 (0.7992) grad_norm 8.1913 (8.3341/1.7449) mem 68106MB [2022-12-20 14:05:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1340/1519] eta 0:02:59 lr 0.000005 time 0.9335 (1.0046) model_time 0.9333 (1.0038) loss 1.1493 (0.7994) grad_norm 9.3948 (8.3509/1.7890) mem 68106MB [2022-12-20 14:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1350/1519] eta 0:02:49 lr 0.000005 time 0.9435 (1.0046) model_time 0.9434 (1.0038) loss 0.8430 (0.7996) grad_norm 6.2055 (8.3596/1.8127) mem 68106MB [2022-12-20 14:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1360/1519] eta 0:02:39 lr 0.000005 time 0.9286 (1.0046) model_time 0.9284 (1.0037) loss 0.6792 (0.8004) grad_norm 7.5153 (8.3596/1.8243) mem 68106MB [2022-12-20 14:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1370/1519] eta 0:02:29 lr 0.000005 time 0.9357 (1.0046) model_time 0.9355 (1.0037) loss 0.7846 (0.8003) grad_norm 7.8927 (8.3809/1.8378) mem 68106MB [2022-12-20 14:06:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1380/1519] eta 0:02:19 lr 0.000005 time 0.9329 (1.0045) model_time 0.9327 (1.0037) loss 0.7871 (0.8005) grad_norm 7.9947 (8.3815/1.8403) mem 68106MB [2022-12-20 14:06:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1390/1519] eta 0:02:09 lr 0.000005 time 0.9331 (1.0046) model_time 0.9329 (1.0037) loss 0.9766 (0.8003) grad_norm 8.0146 (8.3570/1.8433) mem 68106MB [2022-12-20 14:06:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1400/1519] eta 0:01:59 lr 0.000005 time 0.9365 (1.0046) model_time 0.9364 (1.0038) loss 0.8487 (0.8004) grad_norm 7.0228 (8.3830/1.8431) mem 68106MB [2022-12-20 14:06:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1410/1519] eta 0:01:49 lr 0.000005 time 0.9254 (1.0046) model_time 0.9252 (1.0038) loss 0.7148 (0.8000) grad_norm 7.0670 (8.3404/1.8370) mem 68106MB [2022-12-20 14:06:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1420/1519] eta 0:01:39 lr 0.000005 time 0.9318 (1.0048) model_time 0.9317 (1.0039) loss 0.7348 (0.8003) grad_norm 8.3395 (8.3601/1.8462) mem 68106MB [2022-12-20 14:07:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1430/1519] eta 0:01:29 lr 0.000005 time 0.9259 (1.0048) model_time 0.9257 (1.0039) loss 0.9500 (0.8006) grad_norm 10.1722 (8.3686/1.8632) mem 68106MB [2022-12-20 14:07:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1440/1519] eta 0:01:19 lr 0.000005 time 0.9286 (1.0048) model_time 0.9284 (1.0039) loss 0.8510 (0.8009) grad_norm 10.5218 (8.3930/1.8676) mem 68106MB [2022-12-20 14:07:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1450/1519] eta 0:01:09 lr 0.000005 time 0.9319 (1.0047) model_time 0.9317 (1.0039) loss 0.8239 (0.8008) grad_norm 7.1344 (8.3764/1.8409) mem 68106MB [2022-12-20 14:07:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1460/1519] eta 0:00:59 lr 0.000005 time 0.9313 (1.0047) model_time 0.9311 (1.0039) loss 0.6975 (0.8007) grad_norm 10.7725 (8.3892/1.8417) mem 68106MB [2022-12-20 14:07:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1470/1519] eta 0:00:49 lr 0.000005 time 0.9277 (1.0048) model_time 0.9275 (1.0040) loss 0.6991 (0.8006) grad_norm 6.7650 (8.4227/1.8733) mem 68106MB [2022-12-20 14:07:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1480/1519] eta 0:00:39 lr 0.000005 time 0.9322 (1.0048) model_time 0.9321 (1.0040) loss 0.7857 (0.8008) grad_norm 7.0913 (8.4396/1.8702) mem 68106MB [2022-12-20 14:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1490/1519] eta 0:00:29 lr 0.000005 time 0.9122 (1.0049) model_time 0.9120 (1.0041) loss 0.7503 (0.8008) grad_norm 8.5029 (8.3907/1.6824) mem 68106MB [2022-12-20 14:08:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1500/1519] eta 0:00:19 lr 0.000005 time 0.9294 (1.0050) model_time 0.9293 (1.0042) loss 0.7247 (0.8005) grad_norm 10.7930 (8.4357/1.7006) mem 68106MB [2022-12-20 14:08:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [76/100][1510/1519] eta 0:00:09 lr 0.000005 time 0.9167 (1.0050) model_time 0.9166 (1.0042) loss 0.6843 (0.8005) grad_norm 8.7222 (8.4510/1.7234) mem 68106MB [2022-12-20 14:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 76 training takes 0:25:26 [2022-12-20 14:08:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_76.pth saving...... [2022-12-20 14:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_76.pth saved !!! [2022-12-20 14:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.641 (0.641) Loss 0.5352 (0.5352) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 14:09:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.332) Loss 0.5238 (0.5015) Acc@1 92.014 (92.614) Acc@5 98.264 (98.485) Mem 68106MB [2022-12-20 14:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.315) Loss 0.4791 (0.4979) Acc@1 91.667 (92.626) Acc@5 99.306 (98.380) Mem 68106MB [2022-12-20 14:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6214 (0.5048) Acc@1 90.625 (92.350) Acc@5 97.917 (98.387) Mem 68106MB [2022-12-20 14:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.294 (0.307) Loss 0.4622 (0.4949) Acc@1 93.403 (92.429) Acc@5 98.958 (98.493) Mem 68106MB [2022-12-20 14:09:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.309 (0.306) Loss 0.4866 (0.4920) Acc@1 90.972 (92.490) Acc@5 99.653 (98.543) Mem 68106MB [2022-12-20 14:09:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.5120 (0.4918) Acc@1 90.972 (92.435) Acc@5 98.611 (98.531) Mem 68106MB [2022-12-20 14:09:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.301 (0.304) Loss 0.5446 (0.4933) Acc@1 92.361 (92.415) Acc@5 97.917 (98.518) Mem 68106MB [2022-12-20 14:09:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4271 (0.4914) Acc@1 93.056 (92.468) Acc@5 98.611 (98.551) Mem 68106MB [2022-12-20 14:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:76] * Acc@1 92.436 Acc@5 98.555 [2022-12-20 14:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 14:09:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 14:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][0/1519] eta 0:48:05 lr 0.000005 time 1.8999 (1.8999) model_time 1.1174 (1.1174) loss 0.8251 (0.8251) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 14:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][10/1519] eta 0:27:11 lr 0.000005 time 0.9287 (1.0812) model_time 0.9285 (1.0096) loss 0.7150 (0.8626) grad_norm 9.3678 (8.2949/1.5726) mem 68106MB [2022-12-20 14:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][20/1519] eta 0:26:02 lr 0.000005 time 0.9334 (1.0421) model_time 0.9333 (1.0045) loss 0.8108 (0.8662) grad_norm 8.4163 (8.2491/1.2212) mem 68106MB [2022-12-20 14:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][30/1519] eta 0:25:43 lr 0.000005 time 1.1792 (1.0366) model_time 1.1791 (1.0110) loss 0.7219 (0.8445) grad_norm 6.4773 (7.8708/1.1727) mem 68106MB [2022-12-20 14:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][40/1519] eta 0:25:21 lr 0.000005 time 0.9301 (1.0290) model_time 0.9299 (1.0094) loss 0.7631 (0.8255) grad_norm 8.2889 (8.2169/1.3023) mem 68106MB [2022-12-20 14:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][50/1519] eta 0:25:06 lr 0.000005 time 0.9362 (1.0256) model_time 0.9361 (1.0098) loss 0.7384 (0.8279) grad_norm 7.3329 (8.1854/1.3490) mem 68106MB [2022-12-20 14:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][60/1519] eta 0:24:49 lr 0.000004 time 0.9278 (1.0207) model_time 0.9276 (1.0075) loss 0.6705 (0.8205) grad_norm 6.6979 (8.1493/1.4671) mem 68106MB [2022-12-20 14:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][70/1519] eta 0:24:35 lr 0.000004 time 0.9316 (1.0186) model_time 0.9314 (1.0071) loss 1.2249 (0.8137) grad_norm 7.7478 (8.1867/1.4942) mem 68106MB [2022-12-20 14:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][80/1519] eta 0:24:24 lr 0.000004 time 0.9278 (1.0177) model_time 0.9277 (1.0076) loss 0.7617 (0.8247) grad_norm 11.2209 (8.3633/1.5308) mem 68106MB [2022-12-20 14:10:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][90/1519] eta 0:24:11 lr 0.000004 time 0.9378 (1.0156) model_time 0.9377 (1.0065) loss 0.7740 (0.8237) grad_norm 8.6702 (8.4065/1.5322) mem 68106MB [2022-12-20 14:11:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][100/1519] eta 0:24:00 lr 0.000004 time 0.9034 (1.0151) model_time 0.9032 (1.0069) loss 0.6991 (0.8284) grad_norm 7.7013 (8.3497/1.4814) mem 68106MB [2022-12-20 14:11:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][110/1519] eta 0:23:48 lr 0.000004 time 0.9433 (1.0139) model_time 0.9432 (1.0064) loss 0.6744 (0.8219) grad_norm 8.9311 (8.3270/1.4278) mem 68106MB [2022-12-20 14:11:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][120/1519] eta 0:23:36 lr 0.000004 time 0.9332 (1.0127) model_time 0.9331 (1.0058) loss 0.7730 (0.8239) grad_norm 7.8269 (8.3326/1.3786) mem 68106MB [2022-12-20 14:11:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][130/1519] eta 0:23:25 lr 0.000004 time 0.9414 (1.0117) model_time 0.9412 (1.0053) loss 0.8063 (0.8210) grad_norm 9.0599 (8.4384/1.4370) mem 68106MB [2022-12-20 14:11:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][140/1519] eta 0:23:14 lr 0.000004 time 0.9300 (1.0111) model_time 0.9297 (1.0051) loss 0.7937 (0.8222) grad_norm 5.0371 (8.3745/1.4556) mem 68106MB [2022-12-20 14:11:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][150/1519] eta 0:23:03 lr 0.000004 time 0.9329 (1.0103) model_time 0.9327 (1.0047) loss 0.6921 (0.8220) grad_norm 7.6740 (8.7949/3.3168) mem 68106MB [2022-12-20 14:12:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][160/1519] eta 0:22:52 lr 0.000004 time 0.9509 (1.0096) model_time 0.9507 (1.0044) loss 0.8782 (0.8225) grad_norm 8.5974 (8.7280/3.2419) mem 68106MB [2022-12-20 14:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][170/1519] eta 0:22:41 lr 0.000004 time 0.9449 (1.0091) model_time 0.9447 (1.0041) loss 1.0229 (0.8242) grad_norm 6.2936 (8.8523/3.3204) mem 68106MB [2022-12-20 14:12:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][180/1519] eta 0:22:31 lr 0.000004 time 0.9370 (1.0091) model_time 0.9367 (1.0044) loss 0.8024 (0.8227) grad_norm 9.2605 (8.8101/3.2656) mem 68106MB [2022-12-20 14:12:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][190/1519] eta 0:22:20 lr 0.000004 time 0.9230 (1.0086) model_time 0.9228 (1.0041) loss 1.0468 (0.8240) grad_norm 8.1236 (8.7579/3.1952) mem 68106MB [2022-12-20 14:12:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][200/1519] eta 0:22:09 lr 0.000004 time 0.9341 (1.0083) model_time 0.9339 (1.0040) loss 0.7532 (0.8217) grad_norm 12.8223 (8.7762/3.1508) mem 68106MB [2022-12-20 14:12:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][210/1519] eta 0:22:00 lr 0.000004 time 0.9285 (1.0089) model_time 0.9284 (1.0048) loss 0.9340 (0.8242) grad_norm 8.7351 (8.8385/3.1401) mem 68106MB [2022-12-20 14:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][220/1519] eta 0:21:50 lr 0.000004 time 0.9308 (1.0087) model_time 0.9306 (1.0048) loss 0.7031 (0.8236) grad_norm 9.2403 (8.8312/3.0735) mem 68106MB [2022-12-20 14:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][230/1519] eta 0:21:40 lr 0.000004 time 0.9316 (1.0086) model_time 0.9314 (1.0048) loss 0.6978 (0.8234) grad_norm 8.5494 (8.8090/3.0135) mem 68106MB [2022-12-20 14:13:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][240/1519] eta 0:21:29 lr 0.000004 time 0.9299 (1.0081) model_time 0.9298 (1.0045) loss 0.7539 (0.8233) grad_norm 9.6207 (8.8764/3.0227) mem 68106MB [2022-12-20 14:13:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][250/1519] eta 0:21:18 lr 0.000004 time 0.9255 (1.0077) model_time 0.9249 (1.0042) loss 0.6842 (0.8200) grad_norm 7.8474 (8.8307/2.9725) mem 68106MB [2022-12-20 14:13:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][260/1519] eta 0:21:08 lr 0.000004 time 0.9274 (1.0073) model_time 0.9271 (1.0039) loss 0.7017 (0.8211) grad_norm 9.3120 (8.8118/2.9268) mem 68106MB [2022-12-20 14:13:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][270/1519] eta 0:20:58 lr 0.000004 time 0.9122 (1.0073) model_time 0.9119 (1.0040) loss 0.6921 (0.8196) grad_norm 9.5517 (8.7939/2.8785) mem 68106MB [2022-12-20 14:14:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][280/1519] eta 0:20:48 lr 0.000004 time 0.9972 (1.0076) model_time 0.9970 (1.0044) loss 0.8406 (0.8202) grad_norm 9.1322 (8.7610/2.8361) mem 68106MB [2022-12-20 14:14:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][290/1519] eta 0:20:38 lr 0.000004 time 0.9251 (1.0077) model_time 0.9250 (1.0046) loss 0.7775 (0.8205) grad_norm 8.2709 (8.7561/2.8045) mem 68106MB [2022-12-20 14:14:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][300/1519] eta 0:20:27 lr 0.000004 time 0.9239 (1.0074) model_time 0.9238 (1.0044) loss 1.0240 (0.8234) grad_norm 8.4610 (8.7734/2.7661) mem 68106MB [2022-12-20 14:14:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][310/1519] eta 0:20:17 lr 0.000004 time 0.9977 (1.0073) model_time 0.9976 (1.0044) loss 0.7477 (0.8234) grad_norm 10.4675 (8.8010/2.7466) mem 68106MB [2022-12-20 14:14:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][320/1519] eta 0:20:07 lr 0.000004 time 0.9363 (1.0070) model_time 0.9361 (1.0042) loss 0.7362 (0.8226) grad_norm 6.8284 (8.7611/2.7177) mem 68106MB [2022-12-20 14:14:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][330/1519] eta 0:19:57 lr 0.000004 time 0.9412 (1.0069) model_time 0.9411 (1.0042) loss 1.0021 (0.8232) grad_norm 8.2599 (8.7469/2.6869) mem 68106MB [2022-12-20 14:15:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][340/1519] eta 0:19:46 lr 0.000004 time 0.9297 (1.0067) model_time 0.9295 (1.0041) loss 0.8067 (0.8240) grad_norm 6.9824 (8.7120/2.6699) mem 68106MB [2022-12-20 14:15:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][350/1519] eta 0:19:36 lr 0.000004 time 0.9299 (1.0066) model_time 0.9298 (1.0040) loss 0.7687 (0.8224) grad_norm 6.8968 (8.6931/2.6370) mem 68106MB [2022-12-20 14:15:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][360/1519] eta 0:19:26 lr 0.000004 time 0.9273 (1.0067) model_time 0.9272 (1.0042) loss 0.8013 (0.8224) grad_norm 8.1183 (8.6485/2.6153) mem 68106MB [2022-12-20 14:15:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][370/1519] eta 0:19:16 lr 0.000004 time 0.9355 (1.0067) model_time 0.9354 (1.0041) loss 0.8899 (0.8241) grad_norm 8.1749 (8.6298/2.5912) mem 68106MB [2022-12-20 14:15:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][380/1519] eta 0:19:06 lr 0.000004 time 0.9377 (1.0066) model_time 0.9375 (1.0041) loss 0.8334 (0.8241) grad_norm 10.0692 (8.6214/2.5632) mem 68106MB [2022-12-20 14:15:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][390/1519] eta 0:18:56 lr 0.000004 time 1.0307 (1.0069) model_time 1.0306 (1.0045) loss 0.6551 (0.8229) grad_norm 9.2895 (8.6232/2.5382) mem 68106MB [2022-12-20 14:16:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][400/1519] eta 0:18:46 lr 0.000004 time 0.9301 (1.0066) model_time 0.9300 (1.0043) loss 0.6997 (0.8204) grad_norm 8.7797 (8.5902/2.5223) mem 68106MB [2022-12-20 14:16:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][410/1519] eta 0:18:36 lr 0.000004 time 0.9496 (1.0065) model_time 0.9495 (1.0042) loss 0.7479 (0.8194) grad_norm 10.9698 (8.6278/2.5309) mem 68106MB [2022-12-20 14:16:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][420/1519] eta 0:18:25 lr 0.000004 time 0.9363 (1.0063) model_time 0.9360 (1.0041) loss 0.6906 (0.8174) grad_norm 9.0525 (8.6447/2.5177) mem 68106MB [2022-12-20 14:16:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][430/1519] eta 0:18:15 lr 0.000004 time 0.9373 (1.0062) model_time 0.9372 (1.0040) loss 0.6695 (0.8173) grad_norm 6.6866 (8.6146/2.5067) mem 68106MB [2022-12-20 14:16:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][440/1519] eta 0:18:05 lr 0.000004 time 0.9303 (1.0060) model_time 0.9302 (1.0038) loss 0.7003 (0.8150) grad_norm 9.2105 (8.6069/2.4866) mem 68106MB [2022-12-20 14:16:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][450/1519] eta 0:17:55 lr 0.000004 time 0.9306 (1.0059) model_time 0.9305 (1.0038) loss 0.7932 (0.8135) grad_norm 7.4468 (8.5961/2.4692) mem 68106MB [2022-12-20 14:17:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][460/1519] eta 0:17:45 lr 0.000004 time 0.9296 (1.0059) model_time 0.9294 (1.0038) loss 1.1500 (0.8147) grad_norm 9.4031 (8.5671/2.4566) mem 68106MB [2022-12-20 14:17:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][470/1519] eta 0:17:35 lr 0.000004 time 0.9321 (1.0060) model_time 0.9320 (1.0039) loss 0.6714 (0.8147) grad_norm 7.6660 (8.5665/2.4443) mem 68106MB [2022-12-20 14:17:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][480/1519] eta 0:17:25 lr 0.000004 time 0.9312 (1.0058) model_time 0.9311 (1.0038) loss 0.8756 (0.8141) grad_norm 5.7379 (8.5582/2.4385) mem 68106MB [2022-12-20 14:17:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][490/1519] eta 0:17:15 lr 0.000004 time 0.9939 (1.0059) model_time 0.9938 (1.0039) loss 0.7597 (0.8146) grad_norm 6.6462 (8.5405/2.4186) mem 68106MB [2022-12-20 14:17:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][500/1519] eta 0:17:04 lr 0.000004 time 0.9286 (1.0057) model_time 0.9284 (1.0037) loss 0.7280 (0.8134) grad_norm 7.2022 (8.5366/2.3986) mem 68106MB [2022-12-20 14:17:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][510/1519] eta 0:16:55 lr 0.000004 time 0.9303 (1.0061) model_time 0.9302 (1.0041) loss 0.9186 (0.8141) grad_norm 8.5471 (8.5490/2.3876) mem 68106MB [2022-12-20 14:18:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][520/1519] eta 0:16:45 lr 0.000004 time 0.9980 (1.0066) model_time 0.9979 (1.0047) loss 0.8447 (0.8136) grad_norm 9.0733 (8.5414/2.3697) mem 68106MB [2022-12-20 14:18:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][530/1519] eta 0:16:35 lr 0.000004 time 0.9159 (1.0066) model_time 0.9158 (1.0047) loss 0.6670 (0.8124) grad_norm 8.6403 (8.5253/2.3548) mem 68106MB [2022-12-20 14:18:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][540/1519] eta 0:16:25 lr 0.000004 time 0.9346 (1.0068) model_time 0.9345 (1.0049) loss 0.7394 (0.8120) grad_norm 8.4980 (8.5226/2.3394) mem 68106MB [2022-12-20 14:18:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][550/1519] eta 0:16:15 lr 0.000004 time 0.9362 (1.0067) model_time 0.9360 (1.0049) loss 1.0847 (0.8116) grad_norm 10.6042 (8.5468/2.3391) mem 68106MB [2022-12-20 14:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][560/1519] eta 0:16:05 lr 0.000004 time 0.9358 (1.0066) model_time 0.9357 (1.0049) loss 0.6737 (0.8105) grad_norm 10.7945 (8.5828/2.3540) mem 68106MB [2022-12-20 14:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][570/1519] eta 0:15:55 lr 0.000004 time 0.9337 (1.0065) model_time 0.9336 (1.0047) loss 0.8310 (0.8119) grad_norm 6.9372 (8.5802/2.3417) mem 68106MB [2022-12-20 14:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][580/1519] eta 0:15:45 lr 0.000004 time 0.9357 (1.0064) model_time 0.9356 (1.0047) loss 0.9727 (0.8116) grad_norm 9.6265 (8.5940/2.3269) mem 68106MB [2022-12-20 14:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][590/1519] eta 0:15:34 lr 0.000004 time 0.9261 (1.0063) model_time 0.9259 (1.0046) loss 0.7692 (0.8110) grad_norm 8.7123 (8.6120/2.3191) mem 68106MB [2022-12-20 14:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][600/1519] eta 0:15:24 lr 0.000004 time 0.9764 (1.0064) model_time 0.9763 (1.0048) loss 0.8606 (0.8111) grad_norm 6.3030 (8.6024/2.3079) mem 68106MB [2022-12-20 14:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][610/1519] eta 0:15:14 lr 0.000004 time 0.9318 (1.0063) model_time 0.9316 (1.0047) loss 0.6971 (0.8102) grad_norm 9.9967 (8.5957/2.3059) mem 68106MB [2022-12-20 14:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][620/1519] eta 0:15:04 lr 0.000004 time 0.9337 (1.0062) model_time 0.9336 (1.0046) loss 0.9659 (0.8090) grad_norm 8.4166 (8.6412/2.3806) mem 68106MB [2022-12-20 14:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][630/1519] eta 0:14:54 lr 0.000004 time 0.9346 (1.0061) model_time 0.9344 (1.0044) loss 0.7919 (0.8093) grad_norm 7.4387 (8.6697/2.3857) mem 68106MB [2022-12-20 14:20:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][640/1519] eta 0:14:44 lr 0.000004 time 0.9355 (1.0060) model_time 0.9354 (1.0044) loss 0.7214 (0.8091) grad_norm 6.8351 (8.6501/2.3829) mem 68106MB [2022-12-20 14:20:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][650/1519] eta 0:14:34 lr 0.000004 time 0.9288 (1.0059) model_time 0.9287 (1.0043) loss 0.7067 (0.8086) grad_norm 9.3953 (8.6677/2.3777) mem 68106MB [2022-12-20 14:20:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][660/1519] eta 0:14:23 lr 0.000004 time 0.9287 (1.0058) model_time 0.9286 (1.0043) loss 0.8051 (0.8081) grad_norm 9.7271 (8.6867/2.3657) mem 68106MB [2022-12-20 14:20:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][670/1519] eta 0:14:13 lr 0.000004 time 0.9322 (1.0058) model_time 0.9321 (1.0043) loss 0.8744 (0.8079) grad_norm 7.4288 (8.6924/2.3579) mem 68106MB [2022-12-20 14:20:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][680/1519] eta 0:14:03 lr 0.000004 time 0.9311 (1.0058) model_time 0.9310 (1.0043) loss 0.7209 (0.8077) grad_norm 9.3498 (8.7168/2.3937) mem 68106MB [2022-12-20 14:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][690/1519] eta 0:13:53 lr 0.000004 time 0.9493 (1.0058) model_time 0.9492 (1.0043) loss 0.8304 (0.8077) grad_norm 10.0627 (8.7414/2.4128) mem 68106MB [2022-12-20 14:21:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][700/1519] eta 0:13:43 lr 0.000004 time 1.0570 (1.0060) model_time 1.0569 (1.0045) loss 0.9003 (0.8089) grad_norm 6.4934 (8.7540/2.4145) mem 68106MB [2022-12-20 14:21:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][710/1519] eta 0:13:33 lr 0.000004 time 0.9293 (1.0060) model_time 0.9291 (1.0046) loss 0.7424 (0.8083) grad_norm 8.3539 (8.7610/2.4223) mem 68106MB [2022-12-20 14:21:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][720/1519] eta 0:13:23 lr 0.000004 time 0.9311 (1.0061) model_time 0.9309 (1.0046) loss 0.8933 (0.8085) grad_norm 5.3197 (8.7692/2.4380) mem 68106MB [2022-12-20 14:21:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][730/1519] eta 0:13:13 lr 0.000004 time 0.9363 (1.0060) model_time 0.9361 (1.0045) loss 0.9380 (0.8084) grad_norm 6.5995 (8.7378/2.4356) mem 68106MB [2022-12-20 14:21:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][740/1519] eta 0:13:03 lr 0.000004 time 0.9317 (1.0059) model_time 0.9316 (1.0044) loss 0.7504 (0.8087) grad_norm 8.7478 (8.7554/2.4268) mem 68106MB [2022-12-20 14:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][750/1519] eta 0:12:53 lr 0.000004 time 0.9327 (1.0058) model_time 0.9326 (1.0044) loss 0.8344 (0.8093) grad_norm 9.7070 (8.6534/1.9169) mem 68106MB [2022-12-20 14:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][760/1519] eta 0:12:43 lr 0.000004 time 0.9360 (1.0057) model_time 0.9358 (1.0043) loss 0.7889 (0.8092) grad_norm 10.8338 (8.6651/1.9140) mem 68106MB [2022-12-20 14:22:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][770/1519] eta 0:12:33 lr 0.000004 time 0.9301 (1.0058) model_time 0.9300 (1.0044) loss 0.8429 (0.8086) grad_norm 8.2437 (8.6338/1.8551) mem 68106MB [2022-12-20 14:22:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][780/1519] eta 0:12:23 lr 0.000004 time 0.9381 (1.0059) model_time 0.9380 (1.0045) loss 0.7116 (0.8088) grad_norm 6.8140 (8.6233/1.8453) mem 68106MB [2022-12-20 14:22:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][790/1519] eta 0:12:13 lr 0.000004 time 0.9434 (1.0058) model_time 0.9433 (1.0045) loss 0.8372 (0.8088) grad_norm 7.7743 (8.6311/1.8411) mem 68106MB [2022-12-20 14:22:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][800/1519] eta 0:12:03 lr 0.000004 time 0.9232 (1.0058) model_time 0.9230 (1.0044) loss 0.7714 (0.8084) grad_norm 8.6518 (8.6390/1.8319) mem 68106MB [2022-12-20 14:22:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][810/1519] eta 0:11:53 lr 0.000004 time 0.9296 (1.0057) model_time 0.9295 (1.0044) loss 0.6645 (0.8084) grad_norm 7.6721 (8.6047/1.7935) mem 68106MB [2022-12-20 14:23:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][820/1519] eta 0:11:42 lr 0.000004 time 0.9350 (1.0057) model_time 0.9349 (1.0043) loss 0.8010 (0.8091) grad_norm 6.8767 (8.6102/1.8015) mem 68106MB [2022-12-20 14:23:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][830/1519] eta 0:11:32 lr 0.000004 time 0.9272 (1.0056) model_time 0.9270 (1.0043) loss 1.0906 (0.8091) grad_norm 12.6903 (8.6310/1.8182) mem 68106MB [2022-12-20 14:23:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][840/1519] eta 0:11:22 lr 0.000004 time 0.9307 (1.0055) model_time 0.9306 (1.0042) loss 0.6894 (0.8090) grad_norm 9.4758 (8.6087/1.7780) mem 68106MB [2022-12-20 14:23:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][850/1519] eta 0:11:12 lr 0.000004 time 0.9021 (1.0057) model_time 0.9019 (1.0044) loss 1.1110 (0.8090) grad_norm 9.5285 (8.6238/1.7749) mem 68106MB [2022-12-20 14:23:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][860/1519] eta 0:11:02 lr 0.000004 time 0.9381 (1.0056) model_time 0.9379 (1.0043) loss 0.8903 (0.8088) grad_norm 7.0668 (8.6515/1.8213) mem 68106MB [2022-12-20 14:24:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][870/1519] eta 0:10:52 lr 0.000004 time 0.9325 (1.0056) model_time 0.9324 (1.0043) loss 0.8522 (0.8087) grad_norm 6.7751 (8.6702/1.8627) mem 68106MB [2022-12-20 14:24:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][880/1519] eta 0:10:42 lr 0.000004 time 0.9296 (1.0055) model_time 0.9294 (1.0043) loss 0.7736 (0.8081) grad_norm 9.6760 (8.6757/1.8610) mem 68106MB [2022-12-20 14:24:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][890/1519] eta 0:10:32 lr 0.000004 time 0.9363 (1.0056) model_time 0.9361 (1.0043) loss 0.7855 (0.8081) grad_norm 9.9798 (8.6728/1.8651) mem 68106MB [2022-12-20 14:24:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][900/1519] eta 0:10:22 lr 0.000004 time 0.9400 (1.0056) model_time 0.9399 (1.0044) loss 0.6617 (0.8079) grad_norm 7.6342 (8.6927/1.8965) mem 68106MB [2022-12-20 14:24:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][910/1519] eta 0:10:12 lr 0.000004 time 0.9285 (1.0056) model_time 0.9283 (1.0044) loss 0.6579 (0.8079) grad_norm 5.4059 (8.6548/1.8927) mem 68106MB [2022-12-20 14:24:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][920/1519] eta 0:10:02 lr 0.000004 time 0.9281 (1.0056) model_time 0.9280 (1.0044) loss 0.6871 (0.8078) grad_norm 6.0523 (8.6720/1.8965) mem 68106MB [2022-12-20 14:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][930/1519] eta 0:09:52 lr 0.000004 time 0.9327 (1.0055) model_time 0.9326 (1.0043) loss 0.8533 (0.8077) grad_norm 7.3065 (8.6728/1.9029) mem 68106MB [2022-12-20 14:25:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][940/1519] eta 0:09:42 lr 0.000004 time 0.9228 (1.0055) model_time 0.9226 (1.0043) loss 0.7410 (0.8074) grad_norm 7.7527 (8.6732/1.8978) mem 68106MB [2022-12-20 14:25:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][950/1519] eta 0:09:32 lr 0.000004 time 0.9299 (1.0054) model_time 0.9297 (1.0042) loss 1.0261 (0.8087) grad_norm 8.8425 (8.6768/1.8957) mem 68106MB [2022-12-20 14:25:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][960/1519] eta 0:09:21 lr 0.000004 time 0.9309 (1.0053) model_time 0.9308 (1.0041) loss 0.7116 (0.8086) grad_norm 9.9043 (8.7086/1.9063) mem 68106MB [2022-12-20 14:25:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][970/1519] eta 0:09:11 lr 0.000004 time 0.9291 (1.0053) model_time 0.9290 (1.0041) loss 0.6935 (0.8082) grad_norm 7.8822 (8.6921/1.9106) mem 68106MB [2022-12-20 14:25:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][980/1519] eta 0:09:01 lr 0.000004 time 0.9335 (1.0053) model_time 0.9333 (1.0041) loss 0.9355 (0.8081) grad_norm 5.8771 (8.6822/1.9245) mem 68106MB [2022-12-20 14:26:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][990/1519] eta 0:08:51 lr 0.000004 time 0.9346 (1.0052) model_time 0.9345 (1.0040) loss 0.7135 (0.8074) grad_norm 8.2716 (8.6800/1.9281) mem 68106MB [2022-12-20 14:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1000/1519] eta 0:08:41 lr 0.000004 time 0.9244 (1.0052) model_time 0.9243 (1.0041) loss 0.7244 (0.8066) grad_norm 7.0542 (8.6853/1.9194) mem 68106MB [2022-12-20 14:26:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1010/1519] eta 0:08:31 lr 0.000004 time 0.9715 (1.0052) model_time 0.9713 (1.0041) loss 0.6867 (0.8068) grad_norm 8.2173 (8.6609/1.8920) mem 68106MB [2022-12-20 14:26:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1020/1519] eta 0:08:21 lr 0.000004 time 0.9318 (1.0053) model_time 0.9317 (1.0042) loss 0.6597 (0.8064) grad_norm 8.6264 (8.6807/1.9522) mem 68106MB [2022-12-20 14:26:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1030/1519] eta 0:08:11 lr 0.000004 time 0.9295 (1.0053) model_time 0.9294 (1.0042) loss 0.6809 (0.8066) grad_norm 9.8801 (8.7069/1.9387) mem 68106MB [2022-12-20 14:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1040/1519] eta 0:08:01 lr 0.000004 time 0.9311 (1.0053) model_time 0.9310 (1.0042) loss 1.0345 (0.8075) grad_norm 10.5854 (8.7251/1.9352) mem 68106MB [2022-12-20 14:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1050/1519] eta 0:07:51 lr 0.000004 time 0.9326 (1.0053) model_time 0.9325 (1.0041) loss 0.6616 (0.8070) grad_norm 11.5167 (8.7307/1.9408) mem 68106MB [2022-12-20 14:27:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1060/1519] eta 0:07:41 lr 0.000004 time 0.9332 (1.0052) model_time 0.9331 (1.0041) loss 0.9712 (0.8076) grad_norm 10.7740 (8.7957/1.9785) mem 68106MB [2022-12-20 14:27:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1070/1519] eta 0:07:31 lr 0.000004 time 0.9322 (1.0051) model_time 0.9320 (1.0040) loss 0.6950 (0.8072) grad_norm 9.3634 (8.7970/1.9803) mem 68106MB [2022-12-20 14:27:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1080/1519] eta 0:07:21 lr 0.000004 time 0.9351 (1.0052) model_time 0.9349 (1.0041) loss 0.8109 (0.8068) grad_norm 10.0892 (8.8382/1.9872) mem 68106MB [2022-12-20 14:27:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1090/1519] eta 0:07:11 lr 0.000004 time 0.9434 (1.0054) model_time 0.9432 (1.0043) loss 0.9773 (0.8066) grad_norm 6.7547 (8.8392/1.9897) mem 68106MB [2022-12-20 14:27:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1100/1519] eta 0:07:01 lr 0.000004 time 0.9366 (1.0054) model_time 0.9363 (1.0043) loss 0.6592 (0.8059) grad_norm 8.8264 (8.8392/1.9991) mem 68106MB [2022-12-20 14:28:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1110/1519] eta 0:06:51 lr 0.000004 time 0.9234 (1.0054) model_time 0.9232 (1.0043) loss 0.6839 (0.8053) grad_norm 7.6122 (8.8237/1.9917) mem 68106MB [2022-12-20 14:28:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1120/1519] eta 0:06:41 lr 0.000004 time 0.9299 (1.0054) model_time 0.9298 (1.0043) loss 0.9745 (0.8056) grad_norm 9.1810 (8.8376/1.9902) mem 68106MB [2022-12-20 14:28:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1130/1519] eta 0:06:31 lr 0.000004 time 0.9344 (1.0053) model_time 0.9343 (1.0043) loss 0.6584 (0.8055) grad_norm 7.9164 (8.8492/1.9849) mem 68106MB [2022-12-20 14:28:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1140/1519] eta 0:06:21 lr 0.000004 time 0.9398 (1.0053) model_time 0.9396 (1.0042) loss 0.8295 (0.8058) grad_norm 8.0485 (8.8546/1.9824) mem 68106MB [2022-12-20 14:28:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1150/1519] eta 0:06:10 lr 0.000004 time 0.9339 (1.0053) model_time 0.9338 (1.0042) loss 0.9213 (0.8066) grad_norm 9.1174 (8.8310/1.9674) mem 68106MB [2022-12-20 14:28:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1160/1519] eta 0:06:00 lr 0.000004 time 0.9336 (1.0053) model_time 0.9334 (1.0042) loss 0.7060 (0.8066) grad_norm 9.1905 (8.7855/1.9378) mem 68106MB [2022-12-20 14:29:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1170/1519] eta 0:05:50 lr 0.000004 time 0.9312 (1.0053) model_time 0.9311 (1.0043) loss 0.8031 (0.8062) grad_norm 5.7020 (8.7769/1.9420) mem 68106MB [2022-12-20 14:29:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1180/1519] eta 0:05:40 lr 0.000004 time 0.9568 (1.0053) model_time 0.9567 (1.0043) loss 0.7219 (0.8059) grad_norm 7.5758 (8.7602/1.9381) mem 68106MB [2022-12-20 14:29:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1190/1519] eta 0:05:30 lr 0.000004 time 0.9297 (1.0052) model_time 0.9295 (1.0042) loss 0.8862 (0.8060) grad_norm 8.9553 (8.7399/1.9369) mem 68106MB [2022-12-20 14:29:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1200/1519] eta 0:05:20 lr 0.000004 time 0.9425 (1.0053) model_time 0.9423 (1.0042) loss 0.6825 (0.8061) grad_norm 10.1641 (8.7671/1.9398) mem 68106MB [2022-12-20 14:29:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1210/1519] eta 0:05:10 lr 0.000004 time 0.9146 (1.0052) model_time 0.9144 (1.0042) loss 0.8464 (0.8071) grad_norm 8.8054 (8.7783/1.9405) mem 68106MB [2022-12-20 14:29:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1220/1519] eta 0:05:00 lr 0.000004 time 0.9328 (1.0053) model_time 0.9327 (1.0043) loss 0.7182 (0.8071) grad_norm 9.5641 (8.7601/1.8841) mem 68106MB [2022-12-20 14:30:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1230/1519] eta 0:04:50 lr 0.000004 time 0.9342 (1.0052) model_time 0.9341 (1.0042) loss 0.7339 (0.8072) grad_norm 7.9125 (8.7529/1.8718) mem 68106MB [2022-12-20 14:30:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1240/1519] eta 0:04:40 lr 0.000004 time 0.9292 (1.0052) model_time 0.9291 (1.0042) loss 1.1428 (0.8074) grad_norm 6.4870 (8.7595/1.8756) mem 68106MB [2022-12-20 14:30:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1250/1519] eta 0:04:30 lr 0.000004 time 0.9159 (1.0052) model_time 0.9157 (1.0042) loss 0.7718 (0.8072) grad_norm 6.6554 (8.7439/1.8817) mem 68106MB [2022-12-20 14:30:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1260/1519] eta 0:04:20 lr 0.000004 time 0.9307 (1.0051) model_time 0.9306 (1.0041) loss 0.6764 (0.8074) grad_norm 7.3606 (8.7134/1.8930) mem 68106MB [2022-12-20 14:30:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1270/1519] eta 0:04:10 lr 0.000004 time 0.9321 (1.0050) model_time 0.9319 (1.0040) loss 1.1193 (0.8080) grad_norm 8.1733 (8.7417/1.9429) mem 68106MB [2022-12-20 14:30:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1280/1519] eta 0:04:00 lr 0.000004 time 0.9291 (1.0050) model_time 0.9289 (1.0040) loss 0.6791 (0.8080) grad_norm 10.0883 (8.7167/1.9100) mem 68106MB [2022-12-20 14:31:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1290/1519] eta 0:03:50 lr 0.000004 time 0.9310 (1.0049) model_time 0.9309 (1.0039) loss 0.9597 (0.8083) grad_norm 7.8563 (8.6745/1.8870) mem 68106MB [2022-12-20 14:31:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1300/1519] eta 0:03:40 lr 0.000004 time 0.9348 (1.0049) model_time 0.9347 (1.0039) loss 0.6969 (0.8086) grad_norm 7.4272 (8.6801/1.9046) mem 68106MB [2022-12-20 14:31:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1310/1519] eta 0:03:30 lr 0.000004 time 1.1799 (1.0051) model_time 1.1797 (1.0041) loss 1.3049 (0.8086) grad_norm 6.6420 (8.7033/1.9388) mem 68106MB [2022-12-20 14:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1320/1519] eta 0:03:19 lr 0.000004 time 0.9305 (1.0050) model_time 0.9303 (1.0040) loss 0.6610 (0.8083) grad_norm 8.9059 (8.6955/1.9242) mem 68106MB [2022-12-20 14:31:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1330/1519] eta 0:03:09 lr 0.000004 time 0.9198 (1.0051) model_time 0.9197 (1.0041) loss 0.7901 (0.8078) grad_norm 7.3825 (8.7036/1.9169) mem 68106MB [2022-12-20 14:31:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1340/1519] eta 0:02:59 lr 0.000004 time 0.9267 (1.0053) model_time 0.9265 (1.0043) loss 0.7816 (0.8080) grad_norm 9.7558 (8.7029/1.9148) mem 68106MB [2022-12-20 14:32:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1350/1519] eta 0:02:49 lr 0.000004 time 0.9319 (1.0052) model_time 0.9318 (1.0043) loss 0.9687 (0.8079) grad_norm 4.7765 (8.6750/1.9326) mem 68106MB [2022-12-20 14:32:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1360/1519] eta 0:02:39 lr 0.000004 time 0.9348 (1.0052) model_time 0.9343 (1.0042) loss 0.8242 (0.8081) grad_norm 11.3432 (8.6783/1.9404) mem 68106MB [2022-12-20 14:32:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1370/1519] eta 0:02:29 lr 0.000004 time 0.9385 (1.0052) model_time 0.9383 (1.0042) loss 1.0742 (0.8083) grad_norm 6.4532 (8.6669/1.9207) mem 68106MB [2022-12-20 14:32:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1380/1519] eta 0:02:19 lr 0.000004 time 0.9269 (1.0051) model_time 0.9267 (1.0042) loss 0.8955 (0.8084) grad_norm 7.0275 (8.6866/1.9433) mem 68106MB [2022-12-20 14:32:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1390/1519] eta 0:02:09 lr 0.000004 time 0.9271 (1.0052) model_time 0.9270 (1.0043) loss 0.8952 (0.8081) grad_norm 8.5035 (8.6678/1.9566) mem 68106MB [2022-12-20 14:32:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1400/1519] eta 0:01:59 lr 0.000004 time 1.0147 (1.0054) model_time 1.0146 (1.0045) loss 0.8177 (0.8084) grad_norm 7.1714 (8.6216/1.9606) mem 68106MB [2022-12-20 14:33:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1410/1519] eta 0:01:49 lr 0.000004 time 0.9329 (1.0054) model_time 0.9328 (1.0045) loss 0.8277 (0.8083) grad_norm 10.3582 (8.6382/1.9651) mem 68106MB [2022-12-20 14:33:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1420/1519] eta 0:01:39 lr 0.000004 time 0.9353 (1.0053) model_time 0.9351 (1.0044) loss 0.7115 (0.8084) grad_norm 12.6477 (8.6376/1.9845) mem 68106MB [2022-12-20 14:33:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1430/1519] eta 0:01:29 lr 0.000004 time 0.9313 (1.0054) model_time 0.9311 (1.0044) loss 0.6735 (0.8084) grad_norm 8.1017 (8.6184/1.9676) mem 68106MB [2022-12-20 14:33:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1440/1519] eta 0:01:19 lr 0.000004 time 0.9295 (1.0053) model_time 0.9294 (1.0044) loss 0.6883 (0.8088) grad_norm 11.4955 (8.6428/1.9770) mem 68106MB [2022-12-20 14:33:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1450/1519] eta 0:01:09 lr 0.000004 time 0.9188 (1.0053) model_time 0.9186 (1.0043) loss 0.8183 (0.8090) grad_norm 7.2635 (8.6556/1.9916) mem 68106MB [2022-12-20 14:33:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1460/1519] eta 0:00:59 lr 0.000004 time 0.9268 (1.0052) model_time 0.9266 (1.0043) loss 0.6726 (0.8089) grad_norm 7.9262 (8.6307/1.9476) mem 68106MB [2022-12-20 14:34:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1470/1519] eta 0:00:49 lr 0.000004 time 0.9362 (1.0052) model_time 0.9360 (1.0043) loss 0.7002 (0.8088) grad_norm 6.0248 (8.6041/1.9137) mem 68106MB [2022-12-20 14:34:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1480/1519] eta 0:00:39 lr 0.000004 time 0.9312 (1.0052) model_time 0.9310 (1.0043) loss 0.8401 (0.8086) grad_norm 10.7643 (8.5998/1.9248) mem 68106MB [2022-12-20 14:34:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1490/1519] eta 0:00:29 lr 0.000004 time 0.9248 (1.0052) model_time 0.9247 (1.0043) loss 0.9042 (0.8085) grad_norm 8.1099 (8.6018/1.9102) mem 68106MB [2022-12-20 14:34:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1500/1519] eta 0:00:19 lr 0.000004 time 0.9361 (1.0052) model_time 0.9359 (1.0043) loss 0.8438 (0.8086) grad_norm 10.2484 (8.5951/1.9188) mem 68106MB [2022-12-20 14:34:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [77/100][1510/1519] eta 0:00:09 lr 0.000004 time 0.9220 (1.0054) model_time 0.9219 (1.0044) loss 1.2519 (0.8092) grad_norm 7.2228 (8.6124/1.9193) mem 68106MB [2022-12-20 14:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 77 training takes 0:25:27 [2022-12-20 14:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_77.pth saving...... [2022-12-20 14:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_77.pth saved !!! [2022-12-20 14:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.618 (0.618) Loss 0.5344 (0.5344) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 14:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.326) Loss 0.5311 (0.5064) Acc@1 92.014 (92.551) Acc@5 97.917 (98.390) Mem 68106MB [2022-12-20 14:35:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.312) Loss 0.4971 (0.5028) Acc@1 91.319 (92.560) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 14:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.303 (0.309) Loss 0.6258 (0.5080) Acc@1 90.625 (92.350) Acc@5 97.569 (98.353) Mem 68106MB [2022-12-20 14:35:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.307) Loss 0.4593 (0.4984) Acc@1 93.403 (92.412) Acc@5 99.306 (98.484) Mem 68106MB [2022-12-20 14:35:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.302 (0.305) Loss 0.4841 (0.4956) Acc@1 91.667 (92.463) Acc@5 99.653 (98.536) Mem 68106MB [2022-12-20 14:35:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.304 (0.305) Loss 0.5124 (0.4954) Acc@1 91.319 (92.447) Acc@5 98.611 (98.520) Mem 68106MB [2022-12-20 14:35:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.296 (0.304) Loss 0.5487 (0.4968) Acc@1 93.056 (92.395) Acc@5 98.264 (98.528) Mem 68106MB [2022-12-20 14:35:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.305 (0.303) Loss 0.4357 (0.4955) Acc@1 93.750 (92.460) Acc@5 98.264 (98.551) Mem 68106MB [2022-12-20 14:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:77] * Acc@1 92.420 Acc@5 98.555 [2022-12-20 14:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 14:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 14:35:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][0/1519] eta 0:47:22 lr 0.000004 time 1.8710 (1.8710) model_time 1.1580 (1.1580) loss 0.6862 (0.6862) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 14:35:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][10/1519] eta 0:27:12 lr 0.000004 time 0.9177 (1.0820) model_time 0.9176 (1.0167) loss 0.9091 (0.8033) grad_norm 7.1986 (8.8839/2.0392) mem 68106MB [2022-12-20 14:36:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][20/1519] eta 0:26:02 lr 0.000004 time 0.9230 (1.0426) model_time 0.9229 (1.0081) loss 0.7124 (0.8019) grad_norm 6.7224 (8.8585/1.6919) mem 68106MB [2022-12-20 14:36:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][30/1519] eta 0:25:32 lr 0.000004 time 0.9365 (1.0293) model_time 0.9364 (1.0058) loss 0.7073 (0.8187) grad_norm 8.4680 (8.5969/1.8315) mem 68106MB [2022-12-20 14:36:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][40/1519] eta 0:25:11 lr 0.000004 time 0.9312 (1.0218) model_time 0.9311 (1.0040) loss 0.8749 (0.8307) grad_norm 10.7937 (8.6970/1.7044) mem 68106MB [2022-12-20 14:36:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][50/1519] eta 0:24:57 lr 0.000004 time 0.9453 (1.0197) model_time 0.9451 (1.0053) loss 0.9280 (0.8281) grad_norm 8.6207 (8.8122/1.6169) mem 68106MB [2022-12-20 14:36:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][60/1519] eta 0:24:44 lr 0.000004 time 0.9790 (1.0172) model_time 0.9789 (1.0050) loss 0.6870 (0.8139) grad_norm 6.6837 (8.6536/1.6008) mem 68106MB [2022-12-20 14:36:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][70/1519] eta 0:24:33 lr 0.000004 time 1.0206 (1.0167) model_time 1.0205 (1.0062) loss 0.7045 (0.8120) grad_norm 8.1043 (8.6348/1.5276) mem 68106MB [2022-12-20 14:37:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][80/1519] eta 0:24:21 lr 0.000004 time 0.9381 (1.0154) model_time 0.9379 (1.0061) loss 0.8268 (0.8203) grad_norm 8.7981 (8.5547/1.4856) mem 68106MB [2022-12-20 14:37:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][90/1519] eta 0:24:08 lr 0.000004 time 0.9316 (1.0138) model_time 0.9313 (1.0055) loss 0.7491 (0.8169) grad_norm 11.1195 (8.5349/1.5074) mem 68106MB [2022-12-20 14:37:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][100/1519] eta 0:23:57 lr 0.000004 time 0.9193 (1.0134) model_time 0.9191 (1.0059) loss 0.7596 (0.8236) grad_norm 8.6001 (8.4885/1.4617) mem 68106MB [2022-12-20 14:37:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][110/1519] eta 0:23:46 lr 0.000004 time 0.9417 (1.0127) model_time 0.9415 (1.0058) loss 1.0402 (0.8238) grad_norm 6.3229 (8.5084/1.5233) mem 68106MB [2022-12-20 14:37:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][120/1519] eta 0:23:36 lr 0.000004 time 0.9389 (1.0125) model_time 0.9387 (1.0061) loss 0.7849 (0.8234) grad_norm 9.8780 (8.5496/1.5192) mem 68106MB [2022-12-20 14:37:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][130/1519] eta 0:23:25 lr 0.000004 time 0.9278 (1.0116) model_time 0.9276 (1.0057) loss 0.6658 (0.8218) grad_norm 7.5460 (8.5816/1.5229) mem 68106MB [2022-12-20 14:38:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][140/1519] eta 0:23:13 lr 0.000004 time 0.9309 (1.0106) model_time 0.9305 (1.0051) loss 0.6974 (0.8236) grad_norm 9.3479 (8.5740/1.5669) mem 68106MB [2022-12-20 14:38:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][150/1519] eta 0:23:02 lr 0.000004 time 0.9298 (1.0099) model_time 0.9297 (1.0047) loss 0.8667 (0.8217) grad_norm 9.7854 (8.5416/1.5880) mem 68106MB [2022-12-20 14:38:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][160/1519] eta 0:22:52 lr 0.000004 time 0.9307 (1.0102) model_time 0.9305 (1.0054) loss 0.7290 (0.8198) grad_norm 6.3607 (8.4757/1.5867) mem 68106MB [2022-12-20 14:38:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][170/1519] eta 0:22:41 lr 0.000004 time 0.9400 (1.0096) model_time 0.9398 (1.0049) loss 0.6856 (0.8211) grad_norm 11.7508 (8.5759/1.6554) mem 68106MB [2022-12-20 14:38:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][180/1519] eta 0:22:32 lr 0.000004 time 0.9920 (1.0099) model_time 0.9918 (1.0055) loss 0.7697 (0.8185) grad_norm 8.6983 (8.6135/1.6546) mem 68106MB [2022-12-20 14:38:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][190/1519] eta 0:22:22 lr 0.000004 time 0.9384 (1.0098) model_time 0.9382 (1.0056) loss 0.7342 (0.8159) grad_norm 10.2818 (8.6282/1.6368) mem 68106MB [2022-12-20 14:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][200/1519] eta 0:22:11 lr 0.000004 time 0.9328 (1.0093) model_time 0.9327 (1.0053) loss 0.7515 (0.8147) grad_norm 8.1120 (8.6142/1.6108) mem 68106MB [2022-12-20 14:39:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][210/1519] eta 0:22:01 lr 0.000004 time 0.9383 (1.0093) model_time 0.9381 (1.0054) loss 0.9726 (0.8144) grad_norm 7.3652 (8.6112/1.5802) mem 68106MB [2022-12-20 14:39:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][220/1519] eta 0:21:50 lr 0.000004 time 0.9353 (1.0088) model_time 0.9351 (1.0051) loss 0.7255 (0.8141) grad_norm 8.0726 (8.6030/1.5540) mem 68106MB [2022-12-20 14:39:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][230/1519] eta 0:21:40 lr 0.000004 time 0.9303 (1.0087) model_time 0.9301 (1.0052) loss 0.7248 (0.8137) grad_norm 6.4078 (8.6018/1.5382) mem 68106MB [2022-12-20 14:39:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][240/1519] eta 0:21:29 lr 0.000004 time 0.9347 (1.0085) model_time 0.9345 (1.0051) loss 0.7348 (0.8137) grad_norm 7.8000 (8.6010/1.5586) mem 68106MB [2022-12-20 14:39:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][250/1519] eta 0:21:19 lr 0.000004 time 0.9342 (1.0080) model_time 0.9340 (1.0047) loss 0.8818 (0.8153) grad_norm 9.0146 (8.6028/1.5470) mem 68106MB [2022-12-20 14:40:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][260/1519] eta 0:21:08 lr 0.000004 time 0.9273 (1.0078) model_time 0.9271 (1.0046) loss 0.6674 (0.8154) grad_norm 11.5727 (8.6594/1.5521) mem 68106MB [2022-12-20 14:40:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][270/1519] eta 0:20:58 lr 0.000004 time 0.9705 (1.0078) model_time 0.9703 (1.0048) loss 0.7349 (0.8145) grad_norm 15.8296 (8.6735/1.6701) mem 68106MB [2022-12-20 14:40:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][280/1519] eta 0:20:48 lr 0.000004 time 0.9303 (1.0075) model_time 0.9301 (1.0045) loss 1.0005 (0.8160) grad_norm 10.8497 (8.6738/1.6820) mem 68106MB [2022-12-20 14:40:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][290/1519] eta 0:20:37 lr 0.000004 time 0.9287 (1.0072) model_time 0.9285 (1.0043) loss 0.6723 (0.8143) grad_norm 13.0554 (8.7261/1.7497) mem 68106MB [2022-12-20 14:40:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][300/1519] eta 0:20:27 lr 0.000004 time 0.9237 (1.0072) model_time 0.9236 (1.0044) loss 0.7910 (0.8131) grad_norm 6.9260 (8.6998/1.7301) mem 68106MB [2022-12-20 14:40:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][310/1519] eta 0:20:17 lr 0.000004 time 0.9369 (1.0070) model_time 0.9366 (1.0043) loss 0.7578 (0.8107) grad_norm 15.4063 (8.7491/1.7883) mem 68106MB [2022-12-20 14:41:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][320/1519] eta 0:20:07 lr 0.000004 time 0.9267 (1.0067) model_time 0.9265 (1.0041) loss 1.0126 (0.8100) grad_norm 7.9856 (8.7513/1.7651) mem 68106MB [2022-12-20 14:41:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][330/1519] eta 0:19:56 lr 0.000004 time 0.9317 (1.0065) model_time 0.9316 (1.0039) loss 0.7466 (0.8083) grad_norm 8.0963 (8.7664/1.7932) mem 68106MB [2022-12-20 14:41:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][340/1519] eta 0:19:46 lr 0.000004 time 0.9209 (1.0062) model_time 0.9207 (1.0037) loss 0.7273 (0.8087) grad_norm 9.1786 (8.7658/1.7756) mem 68106MB [2022-12-20 14:41:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][350/1519] eta 0:19:36 lr 0.000004 time 0.9209 (1.0061) model_time 0.9207 (1.0037) loss 0.7055 (0.8070) grad_norm 11.3889 (8.7877/1.7815) mem 68106MB [2022-12-20 14:41:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][360/1519] eta 0:19:26 lr 0.000004 time 1.0027 (1.0062) model_time 1.0025 (1.0038) loss 0.8108 (0.8060) grad_norm 9.9727 (8.7939/1.7808) mem 68106MB [2022-12-20 14:41:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][370/1519] eta 0:19:15 lr 0.000004 time 0.9399 (1.0060) model_time 0.9395 (1.0036) loss 0.7072 (0.8058) grad_norm 6.0040 (8.7467/1.7883) mem 68106MB [2022-12-20 14:42:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][380/1519] eta 0:19:05 lr 0.000004 time 0.9349 (1.0059) model_time 0.9348 (1.0036) loss 1.0093 (0.8061) grad_norm 7.5740 (8.7511/1.7690) mem 68106MB [2022-12-20 14:42:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][390/1519] eta 0:18:56 lr 0.000004 time 0.9157 (1.0067) model_time 0.9156 (1.0044) loss 0.7066 (0.8051) grad_norm 9.6762 (8.7253/1.7772) mem 68106MB [2022-12-20 14:42:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][400/1519] eta 0:18:46 lr 0.000004 time 0.9517 (1.0066) model_time 0.9515 (1.0044) loss 0.7908 (0.8053) grad_norm 10.7485 (8.7176/1.7840) mem 68106MB [2022-12-20 14:42:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][410/1519] eta 0:18:36 lr 0.000004 time 1.0297 (1.0066) model_time 1.0296 (1.0044) loss 0.7426 (0.8047) grad_norm 8.4289 (8.6966/1.7707) mem 68106MB [2022-12-20 14:42:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][420/1519] eta 0:18:26 lr 0.000004 time 0.9403 (1.0065) model_time 0.9402 (1.0044) loss 0.7818 (0.8046) grad_norm 7.6553 (8.6756/1.7628) mem 68106MB [2022-12-20 14:42:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][430/1519] eta 0:18:16 lr 0.000004 time 0.9332 (1.0065) model_time 0.9331 (1.0044) loss 0.9371 (0.8054) grad_norm 5.5183 (8.6424/1.7649) mem 68106MB [2022-12-20 14:43:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][440/1519] eta 0:18:06 lr 0.000004 time 0.9887 (1.0066) model_time 0.9886 (1.0046) loss 0.7346 (0.8054) grad_norm 8.0193 (8.6581/1.7747) mem 68106MB [2022-12-20 14:43:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][450/1519] eta 0:17:55 lr 0.000004 time 0.9393 (1.0064) model_time 0.9391 (1.0044) loss 0.6710 (0.8053) grad_norm 7.8047 (8.6728/1.7615) mem 68106MB [2022-12-20 14:43:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][460/1519] eta 0:17:45 lr 0.000004 time 0.9318 (1.0063) model_time 0.9317 (1.0043) loss 1.0017 (0.8064) grad_norm 7.5094 (8.6803/1.7516) mem 68106MB [2022-12-20 14:43:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][470/1519] eta 0:17:35 lr 0.000004 time 0.9383 (1.0064) model_time 0.9382 (1.0044) loss 0.6707 (0.8059) grad_norm 9.3611 (8.6716/1.7472) mem 68106MB [2022-12-20 14:43:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][480/1519] eta 0:17:25 lr 0.000004 time 0.9301 (1.0062) model_time 0.9300 (1.0043) loss 0.7622 (0.8082) grad_norm 8.7739 (8.6438/1.7444) mem 68106MB [2022-12-20 14:43:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][490/1519] eta 0:17:15 lr 0.000004 time 0.9310 (1.0062) model_time 0.9309 (1.0043) loss 0.7491 (0.8085) grad_norm 7.7820 (8.6984/1.8468) mem 68106MB [2022-12-20 14:44:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][500/1519] eta 0:17:05 lr 0.000004 time 0.9270 (1.0062) model_time 0.9268 (1.0043) loss 0.7133 (0.8086) grad_norm 8.6985 (8.7038/1.8413) mem 68106MB [2022-12-20 14:44:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][510/1519] eta 0:16:55 lr 0.000004 time 0.9324 (1.0060) model_time 0.9322 (1.0042) loss 0.8857 (0.8087) grad_norm 10.2800 (8.6786/1.8437) mem 68106MB [2022-12-20 14:44:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][520/1519] eta 0:16:45 lr 0.000004 time 0.9277 (1.0060) model_time 0.9274 (1.0042) loss 0.8487 (0.8080) grad_norm 9.0007 (8.6647/1.8381) mem 68106MB [2022-12-20 14:44:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][530/1519] eta 0:16:34 lr 0.000004 time 0.9324 (1.0060) model_time 0.9322 (1.0042) loss 0.8972 (0.8077) grad_norm 9.8750 (8.6528/1.8350) mem 68106MB [2022-12-20 14:44:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][540/1519] eta 0:16:24 lr 0.000004 time 0.9341 (1.0059) model_time 0.9339 (1.0042) loss 0.7592 (0.8075) grad_norm 8.5956 (8.6373/1.8301) mem 68106MB [2022-12-20 14:44:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][550/1519] eta 0:16:14 lr 0.000004 time 0.9188 (1.0058) model_time 0.9186 (1.0041) loss 0.6641 (0.8073) grad_norm 6.5711 (8.6263/1.8298) mem 68106MB [2022-12-20 14:45:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][560/1519] eta 0:16:04 lr 0.000004 time 0.9322 (1.0057) model_time 0.9321 (1.0040) loss 0.9836 (0.8094) grad_norm 11.4369 (8.6284/1.8235) mem 68106MB [2022-12-20 14:45:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][570/1519] eta 0:15:54 lr 0.000004 time 0.9326 (1.0056) model_time 0.9325 (1.0039) loss 0.8084 (0.8090) grad_norm 8.8497 (8.6147/1.8182) mem 68106MB [2022-12-20 14:45:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][580/1519] eta 0:15:44 lr 0.000004 time 0.9604 (1.0055) model_time 0.9602 (1.0038) loss 0.6826 (0.8106) grad_norm 8.0021 (8.6167/1.8086) mem 68106MB [2022-12-20 14:45:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][590/1519] eta 0:15:34 lr 0.000004 time 0.9266 (1.0055) model_time 0.9264 (1.0039) loss 0.6729 (0.8103) grad_norm 10.8233 (8.6392/1.8307) mem 68106MB [2022-12-20 14:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][600/1519] eta 0:15:23 lr 0.000004 time 0.9298 (1.0054) model_time 0.9296 (1.0038) loss 0.8277 (0.8099) grad_norm 8.1044 (8.6264/1.8200) mem 68106MB [2022-12-20 14:45:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][610/1519] eta 0:15:14 lr 0.000004 time 0.9321 (1.0056) model_time 0.9319 (1.0040) loss 0.6993 (0.8096) grad_norm 9.0019 (8.6400/1.8677) mem 68106MB [2022-12-20 14:46:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][620/1519] eta 0:15:03 lr 0.000004 time 0.9866 (1.0055) model_time 0.9863 (1.0040) loss 0.6989 (0.8098) grad_norm 11.8203 (8.6682/1.9076) mem 68106MB [2022-12-20 14:46:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][630/1519] eta 0:14:53 lr 0.000004 time 0.9283 (1.0055) model_time 0.9281 (1.0039) loss 0.6736 (0.8099) grad_norm 7.3837 (8.6777/1.9044) mem 68106MB [2022-12-20 14:46:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][640/1519] eta 0:14:43 lr 0.000004 time 0.9501 (1.0054) model_time 0.9498 (1.0039) loss 0.8099 (0.8096) grad_norm 9.9535 (8.7021/1.9224) mem 68106MB [2022-12-20 14:46:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][650/1519] eta 0:14:33 lr 0.000004 time 0.9234 (1.0053) model_time 0.9232 (1.0038) loss 0.6597 (0.8095) grad_norm 9.1875 (8.7102/1.9477) mem 68106MB [2022-12-20 14:46:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][660/1519] eta 0:14:23 lr 0.000004 time 0.9216 (1.0052) model_time 0.9215 (1.0037) loss 1.0881 (0.8103) grad_norm 9.0683 (8.7292/1.9557) mem 68106MB [2022-12-20 14:46:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][670/1519] eta 0:14:13 lr 0.000004 time 0.9357 (1.0052) model_time 0.9356 (1.0038) loss 0.8692 (0.8105) grad_norm 8.7712 (8.7327/1.9570) mem 68106MB [2022-12-20 14:47:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][680/1519] eta 0:14:03 lr 0.000004 time 0.9209 (1.0054) model_time 0.9207 (1.0039) loss 1.1652 (0.8104) grad_norm 8.1013 (8.7336/1.9551) mem 68106MB [2022-12-20 14:47:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][690/1519] eta 0:13:53 lr 0.000004 time 0.9303 (1.0053) model_time 0.9302 (1.0039) loss 0.7487 (0.8096) grad_norm 9.6408 (8.7534/1.9508) mem 68106MB [2022-12-20 14:47:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][700/1519] eta 0:13:43 lr 0.000004 time 0.9197 (1.0054) model_time 0.9196 (1.0040) loss 0.7779 (0.8095) grad_norm 7.1817 (8.7634/1.9538) mem 68106MB [2022-12-20 14:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][710/1519] eta 0:13:33 lr 0.000004 time 0.9234 (1.0054) model_time 0.9232 (1.0039) loss 0.9648 (0.8096) grad_norm 6.5366 (8.7485/1.9497) mem 68106MB [2022-12-20 14:47:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][720/1519] eta 0:13:23 lr 0.000004 time 0.9258 (1.0053) model_time 0.9256 (1.0039) loss 0.7619 (0.8090) grad_norm 8.1166 (8.7360/1.9488) mem 68106MB [2022-12-20 14:47:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][730/1519] eta 0:13:13 lr 0.000004 time 0.9398 (1.0055) model_time 0.9397 (1.0041) loss 0.6956 (0.8090) grad_norm 10.0904 (8.7508/1.9445) mem 68106MB [2022-12-20 14:48:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][740/1519] eta 0:13:03 lr 0.000004 time 1.0024 (1.0055) model_time 1.0022 (1.0041) loss 0.7000 (0.8090) grad_norm 8.6912 (8.7396/1.9374) mem 68106MB [2022-12-20 14:48:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][750/1519] eta 0:12:53 lr 0.000004 time 0.9289 (1.0055) model_time 0.9287 (1.0041) loss 0.6713 (0.8087) grad_norm 9.1564 (8.7469/1.9222) mem 68106MB [2022-12-20 14:48:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][760/1519] eta 0:12:43 lr 0.000004 time 0.9266 (1.0054) model_time 0.9264 (1.0041) loss 0.9984 (0.8102) grad_norm 7.6327 (8.7575/1.9160) mem 68106MB [2022-12-20 14:48:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][770/1519] eta 0:12:33 lr 0.000004 time 0.9197 (1.0054) model_time 0.9196 (1.0041) loss 0.6871 (0.8105) grad_norm 8.3116 (8.7454/1.9332) mem 68106MB [2022-12-20 14:48:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][780/1519] eta 0:12:23 lr 0.000004 time 0.9976 (1.0055) model_time 0.9974 (1.0042) loss 0.9283 (0.8110) grad_norm 8.8310 (8.7258/1.9248) mem 68106MB [2022-12-20 14:48:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][790/1519] eta 0:12:13 lr 0.000004 time 0.9214 (1.0055) model_time 0.9212 (1.0042) loss 0.7478 (0.8120) grad_norm 6.3835 (8.6982/1.9332) mem 68106MB [2022-12-20 14:49:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][800/1519] eta 0:12:02 lr 0.000004 time 1.0192 (1.0055) model_time 1.0189 (1.0042) loss 1.0567 (0.8132) grad_norm 11.2162 (8.7002/1.9373) mem 68106MB [2022-12-20 14:49:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][810/1519] eta 0:11:52 lr 0.000004 time 0.9180 (1.0055) model_time 0.9178 (1.0042) loss 0.7880 (0.8135) grad_norm 8.2395 (8.6965/1.9488) mem 68106MB [2022-12-20 14:49:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][820/1519] eta 0:11:42 lr 0.000004 time 0.9253 (1.0053) model_time 0.9251 (1.0041) loss 0.8234 (0.8130) grad_norm 7.8723 (8.7211/1.9731) mem 68106MB [2022-12-20 14:49:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][830/1519] eta 0:11:32 lr 0.000004 time 0.9301 (1.0053) model_time 0.9300 (1.0040) loss 0.7293 (0.8125) grad_norm 6.8243 (8.6979/1.9811) mem 68106MB [2022-12-20 14:49:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][840/1519] eta 0:11:22 lr 0.000004 time 0.9292 (1.0052) model_time 0.9290 (1.0039) loss 0.8784 (0.8126) grad_norm 9.9023 (8.7052/1.9672) mem 68106MB [2022-12-20 14:49:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][850/1519] eta 0:11:12 lr 0.000004 time 0.9182 (1.0052) model_time 0.9180 (1.0040) loss 0.6691 (0.8129) grad_norm 9.1232 (8.6998/1.9622) mem 68106MB [2022-12-20 14:50:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][860/1519] eta 0:11:02 lr 0.000004 time 0.9344 (1.0052) model_time 0.9343 (1.0039) loss 0.9333 (0.8133) grad_norm 8.7972 (8.6736/1.9590) mem 68106MB [2022-12-20 14:50:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][870/1519] eta 0:10:52 lr 0.000004 time 0.9328 (1.0051) model_time 0.9326 (1.0039) loss 0.7732 (0.8136) grad_norm 7.1713 (8.6663/1.9090) mem 68106MB [2022-12-20 14:50:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][880/1519] eta 0:10:42 lr 0.000004 time 0.9284 (1.0050) model_time 0.9282 (1.0038) loss 0.8118 (0.8133) grad_norm 9.0306 (8.6533/1.9002) mem 68106MB [2022-12-20 14:50:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][890/1519] eta 0:10:32 lr 0.000004 time 0.9315 (1.0050) model_time 0.9313 (1.0038) loss 0.8642 (0.8129) grad_norm 11.9066 (8.6366/1.8689) mem 68106MB [2022-12-20 14:50:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][900/1519] eta 0:10:22 lr 0.000004 time 0.9307 (1.0050) model_time 0.9305 (1.0038) loss 0.8061 (0.8128) grad_norm 12.9079 (8.7097/1.9985) mem 68106MB [2022-12-20 14:50:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][910/1519] eta 0:10:12 lr 0.000004 time 0.9337 (1.0051) model_time 0.9335 (1.0039) loss 0.7257 (0.8132) grad_norm 7.8562 (8.6756/1.9618) mem 68106MB [2022-12-20 14:51:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][920/1519] eta 0:10:02 lr 0.000004 time 1.1890 (1.0053) model_time 1.1888 (1.0041) loss 0.7583 (0.8135) grad_norm 6.0805 (8.6432/1.9752) mem 68106MB [2022-12-20 14:51:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][930/1519] eta 0:09:52 lr 0.000004 time 0.9318 (1.0053) model_time 0.9315 (1.0041) loss 0.8371 (0.8131) grad_norm 11.4314 (8.6480/1.9789) mem 68106MB [2022-12-20 14:51:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][940/1519] eta 0:09:42 lr 0.000004 time 0.9321 (1.0052) model_time 0.9320 (1.0041) loss 0.6680 (0.8129) grad_norm 11.3004 (8.6526/1.9862) mem 68106MB [2022-12-20 14:51:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][950/1519] eta 0:09:31 lr 0.000004 time 0.9293 (1.0052) model_time 0.9292 (1.0040) loss 0.9470 (0.8131) grad_norm 8.8684 (8.6291/1.9771) mem 68106MB [2022-12-20 14:51:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][960/1519] eta 0:09:21 lr 0.000004 time 1.0233 (1.0052) model_time 1.0232 (1.0040) loss 0.8479 (0.8133) grad_norm 7.2593 (8.6170/1.9718) mem 68106MB [2022-12-20 14:51:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][970/1519] eta 0:09:11 lr 0.000004 time 0.9320 (1.0052) model_time 0.9319 (1.0040) loss 1.1018 (0.8133) grad_norm 8.9082 (8.6563/1.9624) mem 68106MB [2022-12-20 14:52:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][980/1519] eta 0:09:01 lr 0.000004 time 0.9428 (1.0051) model_time 0.9426 (1.0040) loss 0.6684 (0.8136) grad_norm 9.1214 (8.6530/1.9663) mem 68106MB [2022-12-20 14:52:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][990/1519] eta 0:08:51 lr 0.000004 time 0.9327 (1.0052) model_time 0.9323 (1.0040) loss 0.7298 (0.8133) grad_norm 12.0589 (8.6922/1.9633) mem 68106MB [2022-12-20 14:52:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1000/1519] eta 0:08:41 lr 0.000004 time 0.9309 (1.0052) model_time 0.9308 (1.0041) loss 0.8308 (0.8132) grad_norm 8.7325 (8.7013/1.9511) mem 68106MB [2022-12-20 14:52:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1010/1519] eta 0:08:31 lr 0.000004 time 1.0015 (1.0053) model_time 1.0014 (1.0042) loss 0.7018 (0.8126) grad_norm 8.0069 (8.7028/1.9508) mem 68106MB [2022-12-20 14:52:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1020/1519] eta 0:08:21 lr 0.000004 time 0.9430 (1.0053) model_time 0.9428 (1.0042) loss 1.0599 (0.8128) grad_norm 9.4372 (8.7060/1.9478) mem 68106MB [2022-12-20 14:52:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1030/1519] eta 0:08:11 lr 0.000004 time 0.9283 (1.0053) model_time 0.9282 (1.0042) loss 0.6845 (0.8130) grad_norm 8.1331 (8.7342/1.9332) mem 68106MB [2022-12-20 14:53:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1040/1519] eta 0:08:01 lr 0.000004 time 0.9344 (1.0053) model_time 0.9342 (1.0042) loss 0.6799 (0.8137) grad_norm 8.2739 (8.7068/1.9220) mem 68106MB [2022-12-20 14:53:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1050/1519] eta 0:07:51 lr 0.000004 time 0.9308 (1.0052) model_time 0.9307 (1.0041) loss 0.7423 (0.8136) grad_norm 8.3845 (8.6781/1.9262) mem 68106MB [2022-12-20 14:53:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1060/1519] eta 0:07:41 lr 0.000004 time 0.9651 (1.0053) model_time 0.9650 (1.0042) loss 0.9111 (0.8139) grad_norm 7.0518 (8.6686/1.9445) mem 68106MB [2022-12-20 14:53:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1070/1519] eta 0:07:31 lr 0.000004 time 0.9332 (1.0052) model_time 0.9330 (1.0041) loss 0.6861 (0.8138) grad_norm 11.4250 (8.6980/1.9483) mem 68106MB [2022-12-20 14:53:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1080/1519] eta 0:07:21 lr 0.000004 time 0.9353 (1.0053) model_time 0.9352 (1.0042) loss 0.9509 (0.8142) grad_norm 9.6827 (8.7142/1.9466) mem 68106MB [2022-12-20 14:53:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1090/1519] eta 0:07:11 lr 0.000004 time 0.9842 (1.0053) model_time 0.9840 (1.0042) loss 0.7330 (0.8140) grad_norm 12.6379 (8.6885/1.8719) mem 68106MB [2022-12-20 14:54:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1100/1519] eta 0:07:01 lr 0.000004 time 0.9431 (1.0052) model_time 0.9430 (1.0041) loss 0.6764 (0.8141) grad_norm 10.4807 (8.6824/1.8707) mem 68106MB [2022-12-20 14:54:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1110/1519] eta 0:06:51 lr 0.000004 time 0.9272 (1.0052) model_time 0.9270 (1.0041) loss 0.8546 (0.8146) grad_norm 8.0431 (8.6904/1.8623) mem 68106MB [2022-12-20 14:54:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1120/1519] eta 0:06:41 lr 0.000004 time 0.9793 (1.0052) model_time 0.9791 (1.0042) loss 0.8974 (0.8141) grad_norm 7.3539 (8.6947/1.8555) mem 68106MB [2022-12-20 14:54:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1130/1519] eta 0:06:31 lr 0.000004 time 0.9378 (1.0052) model_time 0.9377 (1.0042) loss 0.7390 (0.8138) grad_norm 9.0479 (8.6966/1.8495) mem 68106MB [2022-12-20 14:54:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1140/1519] eta 0:06:20 lr 0.000004 time 0.9330 (1.0052) model_time 0.9329 (1.0041) loss 0.7620 (0.8132) grad_norm 6.1181 (8.6941/1.8598) mem 68106MB [2022-12-20 14:54:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1150/1519] eta 0:06:10 lr 0.000004 time 0.9343 (1.0051) model_time 0.9342 (1.0041) loss 0.6658 (0.8126) grad_norm 11.6893 (8.7164/1.8555) mem 68106MB [2022-12-20 14:55:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1160/1519] eta 0:06:00 lr 0.000004 time 0.9786 (1.0051) model_time 0.9784 (1.0041) loss 0.7097 (0.8121) grad_norm 5.8440 (8.6997/1.8628) mem 68106MB [2022-12-20 14:55:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1170/1519] eta 0:05:50 lr 0.000004 time 0.9303 (1.0053) model_time 0.9301 (1.0043) loss 0.9899 (0.8122) grad_norm 12.3296 (8.7166/1.8719) mem 68106MB [2022-12-20 14:55:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1180/1519] eta 0:05:40 lr 0.000004 time 0.9399 (1.0053) model_time 0.9398 (1.0043) loss 0.9170 (0.8124) grad_norm 6.5197 (8.7141/1.8759) mem 68106MB [2022-12-20 14:55:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1190/1519] eta 0:05:30 lr 0.000004 time 0.9314 (1.0056) model_time 0.9313 (1.0046) loss 0.8307 (0.8124) grad_norm 9.1372 (8.6942/1.8524) mem 68106MB [2022-12-20 14:55:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1200/1519] eta 0:05:20 lr 0.000004 time 0.9297 (1.0056) model_time 0.9294 (1.0046) loss 0.8189 (0.8119) grad_norm 8.0063 (8.7107/1.8603) mem 68106MB [2022-12-20 14:55:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1210/1519] eta 0:05:10 lr 0.000004 time 0.9269 (1.0055) model_time 0.9268 (1.0045) loss 0.9173 (0.8117) grad_norm 7.3044 (8.6785/1.8014) mem 68106MB [2022-12-20 14:56:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1220/1519] eta 0:05:00 lr 0.000004 time 0.9326 (1.0056) model_time 0.9324 (1.0046) loss 0.6722 (0.8117) grad_norm 7.0791 (8.6445/1.7662) mem 68106MB [2022-12-20 14:56:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1230/1519] eta 0:04:50 lr 0.000004 time 0.9288 (1.0056) model_time 0.9286 (1.0046) loss 0.7650 (0.8117) grad_norm 7.6923 (8.6549/1.7810) mem 68106MB [2022-12-20 14:56:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1240/1519] eta 0:04:40 lr 0.000004 time 0.9342 (1.0056) model_time 0.9340 (1.0046) loss 0.6769 (0.8123) grad_norm 8.4980 (8.6399/1.7693) mem 68106MB [2022-12-20 14:56:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1250/1519] eta 0:04:30 lr 0.000004 time 0.9299 (1.0056) model_time 0.9297 (1.0046) loss 0.9709 (0.8121) grad_norm 8.1272 (8.6070/1.7404) mem 68106MB [2022-12-20 14:56:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1260/1519] eta 0:04:20 lr 0.000004 time 0.9282 (1.0056) model_time 0.9280 (1.0046) loss 1.0908 (0.8121) grad_norm 8.4252 (8.5922/1.7236) mem 68106MB [2022-12-20 14:56:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1270/1519] eta 0:04:10 lr 0.000004 time 0.9993 (1.0056) model_time 0.9992 (1.0046) loss 0.8651 (0.8132) grad_norm 7.9034 (8.5803/1.7207) mem 68106MB [2022-12-20 14:57:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1280/1519] eta 0:04:00 lr 0.000004 time 0.9341 (1.0056) model_time 0.9339 (1.0046) loss 0.9191 (0.8128) grad_norm 8.4866 (8.5887/1.7242) mem 68106MB [2022-12-20 14:57:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1290/1519] eta 0:03:50 lr 0.000004 time 0.9331 (1.0055) model_time 0.9330 (1.0045) loss 0.8713 (0.8129) grad_norm 8.6098 (8.5750/1.7172) mem 68106MB [2022-12-20 14:57:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1300/1519] eta 0:03:40 lr 0.000004 time 0.9723 (1.0056) model_time 0.9722 (1.0047) loss 0.8103 (0.8127) grad_norm 8.2216 (8.5795/1.7177) mem 68106MB [2022-12-20 14:57:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1310/1519] eta 0:03:30 lr 0.000004 time 0.9361 (1.0056) model_time 0.9360 (1.0047) loss 0.7635 (0.8133) grad_norm 12.2482 (8.6081/1.7350) mem 68106MB [2022-12-20 14:57:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1320/1519] eta 0:03:20 lr 0.000004 time 0.9318 (1.0056) model_time 0.9316 (1.0046) loss 0.6883 (0.8133) grad_norm 7.9339 (8.5986/1.7370) mem 68106MB [2022-12-20 14:58:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1330/1519] eta 0:03:10 lr 0.000004 time 0.9371 (1.0056) model_time 0.9369 (1.0046) loss 0.8912 (0.8134) grad_norm 5.9351 (8.5648/1.7391) mem 68106MB [2022-12-20 14:58:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1340/1519] eta 0:02:59 lr 0.000004 time 0.9868 (1.0056) model_time 0.9867 (1.0046) loss 0.7027 (0.8135) grad_norm 7.7460 (8.5653/1.7342) mem 68106MB [2022-12-20 14:58:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1350/1519] eta 0:02:49 lr 0.000004 time 0.9222 (1.0056) model_time 0.9221 (1.0047) loss 0.6737 (0.8128) grad_norm 6.3056 (8.5608/1.7496) mem 68106MB [2022-12-20 14:58:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1360/1519] eta 0:02:39 lr 0.000004 time 0.9311 (1.0055) model_time 0.9310 (1.0046) loss 1.1528 (0.8131) grad_norm 8.8396 (8.5687/1.7553) mem 68106MB [2022-12-20 14:58:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1370/1519] eta 0:02:29 lr 0.000004 time 0.9349 (1.0055) model_time 0.9348 (1.0046) loss 0.6817 (0.8128) grad_norm 12.3407 (8.5594/1.7236) mem 68106MB [2022-12-20 14:58:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1380/1519] eta 0:02:19 lr 0.000004 time 0.9364 (1.0055) model_time 0.9363 (1.0045) loss 1.3093 (0.8130) grad_norm 8.1742 (8.5484/1.7301) mem 68106MB [2022-12-20 14:59:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1390/1519] eta 0:02:09 lr 0.000004 time 0.9308 (1.0055) model_time 0.9307 (1.0046) loss 1.0191 (0.8131) grad_norm 5.8185 (8.5464/1.7283) mem 68106MB [2022-12-20 14:59:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1400/1519] eta 0:01:59 lr 0.000004 time 0.9610 (1.0054) model_time 0.9609 (1.0045) loss 0.9508 (0.8133) grad_norm 10.2877 (8.5561/1.7382) mem 68106MB [2022-12-20 14:59:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1410/1519] eta 0:01:49 lr 0.000004 time 0.9413 (1.0054) model_time 0.9412 (1.0045) loss 0.9535 (0.8137) grad_norm 7.7975 (8.5438/1.7303) mem 68106MB [2022-12-20 14:59:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1420/1519] eta 0:01:39 lr 0.000004 time 0.9321 (1.0056) model_time 0.9319 (1.0047) loss 0.6814 (0.8131) grad_norm 12.4478 (8.5233/1.7163) mem 68106MB [2022-12-20 14:59:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1430/1519] eta 0:01:29 lr 0.000004 time 0.9333 (1.0056) model_time 0.9332 (1.0046) loss 0.8392 (0.8128) grad_norm 8.6679 (8.5398/1.7109) mem 68106MB [2022-12-20 14:59:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1440/1519] eta 0:01:19 lr 0.000004 time 0.9303 (1.0055) model_time 0.9301 (1.0046) loss 0.7382 (0.8125) grad_norm 9.3024 (8.5476/1.7298) mem 68106MB [2022-12-20 15:00:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1450/1519] eta 0:01:09 lr 0.000004 time 0.9282 (1.0054) model_time 0.9281 (1.0046) loss 0.6598 (0.8126) grad_norm 7.0338 (8.5541/1.7446) mem 68106MB [2022-12-20 15:00:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1460/1519] eta 0:00:59 lr 0.000004 time 0.9284 (1.0054) model_time 0.9282 (1.0045) loss 0.7642 (0.8128) grad_norm 6.2957 (8.5636/1.7524) mem 68106MB [2022-12-20 15:00:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1470/1519] eta 0:00:49 lr 0.000004 time 0.9339 (1.0053) model_time 0.9337 (1.0045) loss 1.0017 (0.8129) grad_norm 6.7414 (8.5645/1.7537) mem 68106MB [2022-12-20 15:00:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1480/1519] eta 0:00:39 lr 0.000004 time 0.9288 (1.0054) model_time 0.9286 (1.0045) loss 1.1797 (0.8132) grad_norm 5.5340 (8.5636/1.7689) mem 68106MB [2022-12-20 15:00:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1490/1519] eta 0:00:29 lr 0.000004 time 0.9281 (1.0053) model_time 0.9279 (1.0044) loss 0.8781 (0.8130) grad_norm 6.9436 (8.5685/1.7971) mem 68106MB [2022-12-20 15:00:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1500/1519] eta 0:00:19 lr 0.000004 time 0.9190 (1.0054) model_time 0.9188 (1.0045) loss 0.9664 (0.8132) grad_norm 11.4869 (8.5002/1.6595) mem 68106MB [2022-12-20 15:01:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [78/100][1510/1519] eta 0:00:09 lr 0.000004 time 0.9113 (1.0054) model_time 0.9112 (1.0046) loss 0.7966 (0.8133) grad_norm 7.0732 (8.4934/1.6721) mem 68106MB [2022-12-20 15:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 78 training takes 0:25:27 [2022-12-20 15:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_78.pth saving...... [2022-12-20 15:01:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_78.pth saved !!! [2022-12-20 15:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.666 (0.666) Loss 0.5412 (0.5412) Acc@1 91.319 (91.319) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 15:01:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.331) Loss 0.5298 (0.5084) Acc@1 92.361 (92.487) Acc@5 97.917 (98.327) Mem 68106MB [2022-12-20 15:01:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.315) Loss 0.4906 (0.5032) Acc@1 90.278 (92.460) Acc@5 98.958 (98.347) Mem 68106MB [2022-12-20 15:01:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.302 (0.310) Loss 0.6404 (0.5102) Acc@1 90.278 (92.272) Acc@5 97.917 (98.342) Mem 68106MB [2022-12-20 15:01:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.308) Loss 0.4644 (0.5006) Acc@1 93.750 (92.420) Acc@5 99.306 (98.493) Mem 68106MB [2022-12-20 15:01:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.306) Loss 0.4852 (0.4981) Acc@1 92.361 (92.490) Acc@5 99.653 (98.543) Mem 68106MB [2022-12-20 15:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.295 (0.305) Loss 0.5112 (0.4974) Acc@1 90.972 (92.469) Acc@5 98.264 (98.520) Mem 68106MB [2022-12-20 15:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.305 (0.304) Loss 0.5470 (0.4986) Acc@1 93.056 (92.420) Acc@5 97.917 (98.508) Mem 68106MB [2022-12-20 15:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.302 (0.303) Loss 0.4360 (0.4973) Acc@1 93.056 (92.490) Acc@5 98.264 (98.543) Mem 68106MB [2022-12-20 15:02:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:78] * Acc@1 92.453 Acc@5 98.543 [2022-12-20 15:02:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 15:02:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.51% [2022-12-20 15:02:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][0/1519] eta 0:44:38 lr 0.000004 time 1.7635 (1.7635) model_time 0.9576 (0.9576) loss 0.8852 (0.8852) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 15:02:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][10/1519] eta 0:27:11 lr 0.000004 time 0.9282 (1.0812) model_time 0.9278 (1.0076) loss 0.7241 (0.8153) grad_norm 7.7992 (7.5203/0.8403) mem 68106MB [2022-12-20 15:02:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][20/1519] eta 0:26:08 lr 0.000004 time 0.9728 (1.0465) model_time 0.9726 (1.0077) loss 1.1841 (0.8280) grad_norm 8.0598 (7.4763/0.9135) mem 68106MB [2022-12-20 15:02:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][30/1519] eta 0:25:38 lr 0.000004 time 0.9358 (1.0332) model_time 0.9357 (1.0068) loss 0.7061 (0.8130) grad_norm 7.4657 (7.8086/1.0513) mem 68106MB [2022-12-20 15:02:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][40/1519] eta 0:25:16 lr 0.000004 time 0.9365 (1.0254) model_time 0.9364 (1.0054) loss 0.7784 (0.8228) grad_norm 11.2731 (8.0993/1.3330) mem 68106MB [2022-12-20 15:02:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][50/1519] eta 0:24:58 lr 0.000004 time 0.9193 (1.0202) model_time 0.9192 (1.0040) loss 0.7160 (0.8143) grad_norm 6.8388 (8.2767/1.3641) mem 68106MB [2022-12-20 15:03:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][60/1519] eta 0:24:45 lr 0.000004 time 0.9340 (1.0183) model_time 0.9338 (1.0047) loss 0.6791 (0.8043) grad_norm 6.9531 (8.1393/1.3138) mem 68106MB [2022-12-20 15:03:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][70/1519] eta 0:24:34 lr 0.000004 time 0.9373 (1.0173) model_time 0.9372 (1.0055) loss 0.8500 (0.8139) grad_norm 7.4113 (8.1500/1.2685) mem 68106MB [2022-12-20 15:03:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][80/1519] eta 0:24:20 lr 0.000004 time 0.9246 (1.0149) model_time 0.9245 (1.0045) loss 0.8483 (0.8205) grad_norm 14.7522 (8.3623/1.6806) mem 68106MB [2022-12-20 15:03:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][90/1519] eta 0:24:09 lr 0.000004 time 0.9274 (1.0140) model_time 0.9270 (1.0048) loss 0.8228 (0.8113) grad_norm 11.4538 (8.7314/2.3707) mem 68106MB [2022-12-20 15:03:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][100/1519] eta 0:23:57 lr 0.000004 time 0.9330 (1.0130) model_time 0.9327 (1.0046) loss 0.7249 (0.8096) grad_norm 8.6390 (8.7030/2.3267) mem 68106MB [2022-12-20 15:03:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][110/1519] eta 0:23:50 lr 0.000004 time 0.9347 (1.0150) model_time 0.9344 (1.0073) loss 0.7101 (0.8081) grad_norm 6.6752 (8.6344/2.2725) mem 68106MB [2022-12-20 15:04:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][120/1519] eta 0:23:39 lr 0.000004 time 0.9397 (1.0143) model_time 0.9395 (1.0072) loss 0.8768 (0.8099) grad_norm 9.6683 (8.5760/2.1994) mem 68106MB [2022-12-20 15:04:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][130/1519] eta 0:23:27 lr 0.000004 time 0.9330 (1.0130) model_time 0.9327 (1.0065) loss 0.7677 (0.8151) grad_norm 9.6405 (8.5505/2.1361) mem 68106MB [2022-12-20 15:04:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][140/1519] eta 0:23:16 lr 0.000004 time 0.9258 (1.0126) model_time 0.9255 (1.0065) loss 0.8882 (0.8175) grad_norm 11.7071 (8.6624/2.1895) mem 68106MB [2022-12-20 15:04:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][150/1519] eta 0:23:05 lr 0.000004 time 0.9288 (1.0121) model_time 0.9286 (1.0064) loss 0.7380 (0.8153) grad_norm 10.5261 (8.7332/2.2253) mem 68106MB [2022-12-20 15:04:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][160/1519] eta 0:22:54 lr 0.000004 time 0.9313 (1.0114) model_time 0.9312 (1.0060) loss 0.9255 (0.8161) grad_norm 10.4883 (8.7905/2.1957) mem 68106MB [2022-12-20 15:04:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][170/1519] eta 0:22:46 lr 0.000004 time 0.9407 (1.0126) model_time 0.9405 (1.0075) loss 0.8834 (0.8139) grad_norm 9.7254 (8.7383/2.1579) mem 68106MB [2022-12-20 15:05:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][180/1519] eta 0:22:35 lr 0.000004 time 0.9306 (1.0122) model_time 0.9304 (1.0074) loss 0.9134 (0.8155) grad_norm 8.2529 (8.7727/2.1370) mem 68106MB [2022-12-20 15:05:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][190/1519] eta 0:22:25 lr 0.000004 time 0.9280 (1.0120) model_time 0.9278 (1.0074) loss 0.6591 (0.8122) grad_norm 9.4310 (8.7501/2.1046) mem 68106MB [2022-12-20 15:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][200/1519] eta 0:22:14 lr 0.000004 time 0.9375 (1.0118) model_time 0.9373 (1.0074) loss 0.8967 (0.8117) grad_norm 8.7056 (8.7268/2.0624) mem 68106MB [2022-12-20 15:05:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][210/1519] eta 0:22:03 lr 0.000004 time 0.9376 (1.0114) model_time 0.9375 (1.0072) loss 0.7048 (0.8117) grad_norm 8.3074 (8.6933/2.0376) mem 68106MB [2022-12-20 15:05:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][220/1519] eta 0:21:53 lr 0.000004 time 0.9305 (1.0109) model_time 0.9302 (1.0068) loss 0.6746 (0.8118) grad_norm 11.7092 (8.6697/2.0468) mem 68106MB [2022-12-20 15:05:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][230/1519] eta 0:21:42 lr 0.000004 time 0.9300 (1.0103) model_time 0.9298 (1.0064) loss 1.0235 (0.8102) grad_norm 10.2406 (8.6691/2.0094) mem 68106MB [2022-12-20 15:06:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][240/1519] eta 0:21:31 lr 0.000004 time 0.9276 (1.0098) model_time 0.9274 (1.0060) loss 0.7993 (0.8107) grad_norm 11.1933 (8.7043/2.0273) mem 68106MB [2022-12-20 15:06:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][250/1519] eta 0:21:21 lr 0.000004 time 0.9296 (1.0100) model_time 0.9294 (1.0064) loss 0.6643 (0.8075) grad_norm 7.1798 (8.6387/2.0403) mem 68106MB [2022-12-20 15:06:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][260/1519] eta 0:21:11 lr 0.000004 time 0.9343 (1.0096) model_time 0.9341 (1.0061) loss 0.6876 (0.8046) grad_norm 5.4761 (8.6286/2.0405) mem 68106MB [2022-12-20 15:06:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][270/1519] eta 0:21:00 lr 0.000004 time 0.9262 (1.0092) model_time 0.9260 (1.0058) loss 0.7199 (0.8041) grad_norm 7.8397 (8.6043/2.0312) mem 68106MB [2022-12-20 15:06:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][280/1519] eta 0:20:50 lr 0.000004 time 0.9335 (1.0089) model_time 0.9333 (1.0056) loss 0.8442 (0.8029) grad_norm 7.8209 (8.5591/2.0104) mem 68106MB [2022-12-20 15:06:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][290/1519] eta 0:20:39 lr 0.000004 time 0.9300 (1.0088) model_time 0.9299 (1.0056) loss 0.7507 (0.8012) grad_norm 10.9011 (8.5866/2.0374) mem 68106MB [2022-12-20 15:07:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][300/1519] eta 0:20:29 lr 0.000004 time 1.0074 (1.0089) model_time 1.0072 (1.0058) loss 1.1482 (0.8027) grad_norm 11.5596 (8.6308/2.0319) mem 68106MB [2022-12-20 15:07:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][310/1519] eta 0:20:19 lr 0.000004 time 0.9297 (1.0086) model_time 0.9294 (1.0056) loss 0.6612 (0.8045) grad_norm 6.7025 (8.5993/2.0126) mem 68106MB [2022-12-20 15:07:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][320/1519] eta 0:20:09 lr 0.000004 time 0.9343 (1.0084) model_time 0.9341 (1.0055) loss 0.9067 (0.8063) grad_norm 12.7310 (8.6787/2.1432) mem 68106MB [2022-12-20 15:07:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][330/1519] eta 0:19:58 lr 0.000004 time 0.9294 (1.0083) model_time 0.9293 (1.0055) loss 1.0965 (0.8054) grad_norm 7.2069 (8.6974/2.1319) mem 68106MB [2022-12-20 15:07:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][340/1519] eta 0:19:49 lr 0.000004 time 0.9416 (1.0085) model_time 0.9415 (1.0057) loss 0.6757 (0.8056) grad_norm 8.6647 (8.6938/2.1240) mem 68106MB [2022-12-20 15:07:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][350/1519] eta 0:19:38 lr 0.000004 time 0.9298 (1.0082) model_time 0.9295 (1.0055) loss 0.8481 (0.8063) grad_norm 8.2270 (8.6791/2.0993) mem 68106MB [2022-12-20 15:08:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][360/1519] eta 0:19:28 lr 0.000004 time 0.9319 (1.0081) model_time 0.9317 (1.0054) loss 1.3444 (0.8073) grad_norm 8.4406 (8.7117/2.1127) mem 68106MB [2022-12-20 15:08:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][370/1519] eta 0:19:18 lr 0.000004 time 0.9625 (1.0079) model_time 0.9623 (1.0053) loss 1.0845 (0.8099) grad_norm 7.9862 (8.6958/2.1022) mem 68106MB [2022-12-20 15:08:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][380/1519] eta 0:19:07 lr 0.000004 time 0.9358 (1.0076) model_time 0.9356 (1.0050) loss 0.8755 (0.8126) grad_norm 7.8355 (8.6852/2.0824) mem 68106MB [2022-12-20 15:08:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][390/1519] eta 0:18:57 lr 0.000004 time 0.9253 (1.0075) model_time 0.9251 (1.0050) loss 0.7026 (0.8117) grad_norm 18.9780 (8.7006/2.2001) mem 68106MB [2022-12-20 15:08:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][400/1519] eta 0:18:47 lr 0.000004 time 0.9319 (1.0075) model_time 0.9318 (1.0051) loss 0.6673 (0.8103) grad_norm 11.2186 (8.7085/2.2099) mem 68106MB [2022-12-20 15:08:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][410/1519] eta 0:18:37 lr 0.000004 time 0.9357 (1.0074) model_time 0.9355 (1.0050) loss 0.6512 (0.8095) grad_norm 7.5058 (8.7334/2.2098) mem 68106MB [2022-12-20 15:09:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][420/1519] eta 0:18:27 lr 0.000004 time 0.9337 (1.0073) model_time 0.9335 (1.0050) loss 0.8418 (0.8094) grad_norm 8.6369 (8.7287/2.1849) mem 68106MB [2022-12-20 15:09:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][430/1519] eta 0:18:17 lr 0.000004 time 0.9124 (1.0074) model_time 0.9123 (1.0052) loss 0.7269 (0.8088) grad_norm 8.8390 (8.7419/2.1706) mem 68106MB [2022-12-20 15:09:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][440/1519] eta 0:18:06 lr 0.000004 time 0.9469 (1.0073) model_time 0.9467 (1.0051) loss 0.7080 (0.8083) grad_norm 7.8381 (8.7441/2.1620) mem 68106MB [2022-12-20 15:09:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][450/1519] eta 0:17:56 lr 0.000004 time 0.9360 (1.0071) model_time 0.9359 (1.0049) loss 0.7311 (0.8090) grad_norm 11.5455 (8.7660/2.1514) mem 68106MB [2022-12-20 15:09:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][460/1519] eta 0:17:46 lr 0.000004 time 0.9320 (1.0069) model_time 0.9318 (1.0048) loss 0.9368 (0.8083) grad_norm 8.9368 (8.7714/2.1321) mem 68106MB [2022-12-20 15:09:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][470/1519] eta 0:17:36 lr 0.000004 time 0.9304 (1.0070) model_time 0.9303 (1.0048) loss 0.9571 (0.8077) grad_norm 8.2548 (8.7495/2.1174) mem 68106MB [2022-12-20 15:10:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][480/1519] eta 0:17:27 lr 0.000004 time 1.2571 (1.0077) model_time 1.2569 (1.0056) loss 0.8382 (0.8087) grad_norm 7.9373 (8.7432/2.1000) mem 68106MB [2022-12-20 15:10:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][490/1519] eta 0:17:16 lr 0.000004 time 0.9276 (1.0077) model_time 0.9275 (1.0056) loss 0.6633 (0.8086) grad_norm 7.8259 (8.7382/2.0906) mem 68106MB [2022-12-20 15:10:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][500/1519] eta 0:17:06 lr 0.000004 time 0.9321 (1.0075) model_time 0.9319 (1.0055) loss 0.6973 (0.8067) grad_norm 6.8583 (8.7117/2.0799) mem 68106MB [2022-12-20 15:10:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][510/1519] eta 0:16:56 lr 0.000004 time 0.9449 (1.0076) model_time 0.9447 (1.0057) loss 0.6981 (0.8063) grad_norm 6.6966 (8.7280/2.1176) mem 68106MB [2022-12-20 15:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][520/1519] eta 0:16:46 lr 0.000004 time 0.9288 (1.0076) model_time 0.9287 (1.0057) loss 0.7554 (0.8056) grad_norm 7.1583 (8.7079/2.1062) mem 68106MB [2022-12-20 15:10:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][530/1519] eta 0:16:36 lr 0.000004 time 0.9312 (1.0075) model_time 0.9311 (1.0055) loss 0.7677 (0.8061) grad_norm 9.6580 (8.7099/2.0880) mem 68106MB [2022-12-20 15:11:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][540/1519] eta 0:16:25 lr 0.000004 time 0.9265 (1.0071) model_time 0.9264 (1.0052) loss 1.3398 (0.8073) grad_norm 8.3422 (8.6968/2.0775) mem 68106MB [2022-12-20 15:11:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][550/1519] eta 0:16:15 lr 0.000004 time 0.9173 (1.0069) model_time 0.9172 (1.0050) loss 0.8590 (0.8072) grad_norm 9.4497 (8.6967/2.0638) mem 68106MB [2022-12-20 15:11:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][560/1519] eta 0:16:05 lr 0.000004 time 0.9069 (1.0070) model_time 0.9068 (1.0051) loss 0.7975 (0.8074) grad_norm 8.6163 (8.7095/2.0750) mem 68106MB [2022-12-20 15:11:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][570/1519] eta 0:15:55 lr 0.000004 time 0.9423 (1.0069) model_time 0.9421 (1.0051) loss 0.7364 (0.8087) grad_norm 6.1985 (8.6931/2.0700) mem 68106MB [2022-12-20 15:11:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][580/1519] eta 0:15:45 lr 0.000004 time 0.9457 (1.0069) model_time 0.9455 (1.0051) loss 1.0196 (0.8085) grad_norm 7.7683 (8.6709/2.0629) mem 68106MB [2022-12-20 15:11:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][590/1519] eta 0:15:35 lr 0.000004 time 0.9374 (1.0069) model_time 0.9371 (1.0051) loss 0.8062 (0.8080) grad_norm 9.1449 (8.6679/2.0548) mem 68106MB [2022-12-20 15:12:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][600/1519] eta 0:15:25 lr 0.000004 time 0.9428 (1.0069) model_time 0.9423 (1.0052) loss 0.6898 (0.8082) grad_norm 7.7822 (8.6615/2.0460) mem 68106MB [2022-12-20 15:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][610/1519] eta 0:15:15 lr 0.000004 time 0.9341 (1.0069) model_time 0.9340 (1.0051) loss 0.7973 (0.8069) grad_norm 7.4385 (8.6600/2.0409) mem 68106MB [2022-12-20 15:12:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][620/1519] eta 0:15:05 lr 0.000004 time 0.9368 (1.0069) model_time 0.9366 (1.0052) loss 0.8244 (0.8079) grad_norm 11.0978 (8.6895/2.0411) mem 68106MB [2022-12-20 15:12:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][630/1519] eta 0:14:55 lr 0.000004 time 0.9341 (1.0068) model_time 0.9340 (1.0051) loss 0.8723 (0.8083) grad_norm 7.3909 (8.6938/2.0560) mem 68106MB [2022-12-20 15:12:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][640/1519] eta 0:14:44 lr 0.000004 time 0.9343 (1.0068) model_time 0.9341 (1.0052) loss 0.7812 (0.8083) grad_norm 5.8716 (8.6797/2.0614) mem 68106MB [2022-12-20 15:12:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][650/1519] eta 0:14:34 lr 0.000004 time 0.9304 (1.0067) model_time 0.9303 (1.0051) loss 0.6683 (0.8079) grad_norm 8.2558 (8.6518/2.0559) mem 68106MB [2022-12-20 15:13:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][660/1519] eta 0:14:24 lr 0.000004 time 0.9354 (1.0066) model_time 0.9352 (1.0050) loss 0.7493 (0.8074) grad_norm 11.0312 (8.6913/2.0708) mem 68106MB [2022-12-20 15:13:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][670/1519] eta 0:14:14 lr 0.000004 time 0.9295 (1.0065) model_time 0.9293 (1.0049) loss 0.7958 (0.8072) grad_norm 8.4661 (8.7135/2.0851) mem 68106MB [2022-12-20 15:13:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][680/1519] eta 0:14:04 lr 0.000004 time 0.9360 (1.0065) model_time 0.9358 (1.0049) loss 0.8137 (0.8073) grad_norm 7.6187 (8.7012/2.0812) mem 68106MB [2022-12-20 15:13:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][690/1519] eta 0:13:54 lr 0.000004 time 0.9295 (1.0065) model_time 0.9294 (1.0049) loss 0.6598 (0.8059) grad_norm 7.3171 (8.6492/1.9654) mem 68106MB [2022-12-20 15:13:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][700/1519] eta 0:13:44 lr 0.000004 time 0.9146 (1.0064) model_time 0.9144 (1.0048) loss 0.7267 (0.8065) grad_norm 6.2598 (8.6135/1.9589) mem 68106MB [2022-12-20 15:13:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][710/1519] eta 0:13:34 lr 0.000004 time 0.9380 (1.0064) model_time 0.9379 (1.0048) loss 0.6975 (0.8062) grad_norm 8.5639 (8.6288/1.9581) mem 68106MB [2022-12-20 15:14:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][720/1519] eta 0:13:23 lr 0.000004 time 0.9306 (1.0062) model_time 0.9305 (1.0047) loss 0.6736 (0.8056) grad_norm 8.1290 (8.6490/1.9572) mem 68106MB [2022-12-20 15:14:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][730/1519] eta 0:13:13 lr 0.000004 time 0.9331 (1.0061) model_time 0.9329 (1.0046) loss 0.6749 (0.8062) grad_norm 6.1705 (8.6369/1.9619) mem 68106MB [2022-12-20 15:14:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][740/1519] eta 0:13:03 lr 0.000004 time 0.9327 (1.0060) model_time 0.9326 (1.0045) loss 0.7920 (0.8059) grad_norm 8.3096 (8.6155/1.9494) mem 68106MB [2022-12-20 15:14:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][750/1519] eta 0:12:53 lr 0.000004 time 0.9369 (1.0060) model_time 0.9368 (1.0046) loss 0.6713 (0.8057) grad_norm 7.3249 (8.5724/1.9220) mem 68106MB [2022-12-20 15:14:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][760/1519] eta 0:12:43 lr 0.000004 time 0.9422 (1.0060) model_time 0.9421 (1.0045) loss 0.6911 (0.8055) grad_norm 8.1483 (8.5911/1.9904) mem 68106MB [2022-12-20 15:14:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][770/1519] eta 0:12:33 lr 0.000004 time 0.9359 (1.0060) model_time 0.9357 (1.0045) loss 0.8417 (0.8052) grad_norm 8.5807 (8.6237/1.9926) mem 68106MB [2022-12-20 15:15:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][780/1519] eta 0:12:23 lr 0.000004 time 0.9375 (1.0060) model_time 0.9373 (1.0046) loss 0.7071 (0.8049) grad_norm 7.1262 (8.6034/1.9887) mem 68106MB [2022-12-20 15:15:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][790/1519] eta 0:12:13 lr 0.000004 time 0.9110 (1.0060) model_time 0.9108 (1.0045) loss 0.6915 (0.8053) grad_norm 8.3526 (8.6092/1.9879) mem 68106MB [2022-12-20 15:15:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][800/1519] eta 0:12:03 lr 0.000004 time 0.9956 (1.0061) model_time 0.9954 (1.0047) loss 0.6564 (0.8054) grad_norm 8.6943 (8.6125/1.9905) mem 68106MB [2022-12-20 15:15:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][810/1519] eta 0:11:53 lr 0.000004 time 0.9356 (1.0061) model_time 0.9354 (1.0047) loss 0.6951 (0.8056) grad_norm 8.2247 (8.6040/1.9897) mem 68106MB [2022-12-20 15:15:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][820/1519] eta 0:11:43 lr 0.000004 time 0.9590 (1.0061) model_time 0.9589 (1.0047) loss 0.9510 (0.8060) grad_norm 8.5504 (8.6216/1.9939) mem 68106MB [2022-12-20 15:15:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][830/1519] eta 0:11:33 lr 0.000004 time 0.9313 (1.0060) model_time 0.9311 (1.0047) loss 0.8947 (0.8060) grad_norm 9.2781 (8.6117/1.9930) mem 68106MB [2022-12-20 15:16:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][840/1519] eta 0:11:23 lr 0.000004 time 0.9402 (1.0061) model_time 0.9401 (1.0047) loss 0.8242 (0.8061) grad_norm 7.7423 (8.6069/1.9797) mem 68106MB [2022-12-20 15:16:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][850/1519] eta 0:11:13 lr 0.000004 time 0.9368 (1.0061) model_time 0.9366 (1.0047) loss 0.6977 (0.8057) grad_norm 9.5276 (8.6311/1.9717) mem 68106MB [2022-12-20 15:16:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][860/1519] eta 0:11:02 lr 0.000004 time 0.9303 (1.0060) model_time 0.9301 (1.0047) loss 0.7962 (0.8047) grad_norm 14.2081 (8.6432/1.9889) mem 68106MB [2022-12-20 15:16:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][870/1519] eta 0:10:52 lr 0.000004 time 0.9780 (1.0060) model_time 0.9778 (1.0047) loss 0.6994 (0.8042) grad_norm 6.9638 (8.6380/1.9776) mem 68106MB [2022-12-20 15:16:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][880/1519] eta 0:10:42 lr 0.000004 time 0.9541 (1.0061) model_time 0.9540 (1.0048) loss 0.7404 (0.8049) grad_norm 7.0564 (8.6479/1.9744) mem 68106MB [2022-12-20 15:16:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][890/1519] eta 0:10:32 lr 0.000004 time 0.9352 (1.0060) model_time 0.9351 (1.0047) loss 0.6938 (0.8050) grad_norm 10.4592 (8.6681/1.9580) mem 68106MB [2022-12-20 15:17:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][900/1519] eta 0:10:22 lr 0.000004 time 0.9182 (1.0061) model_time 0.9181 (1.0048) loss 0.8707 (0.8051) grad_norm 13.2260 (8.6422/1.9745) mem 68106MB [2022-12-20 15:17:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][910/1519] eta 0:10:12 lr 0.000004 time 0.9435 (1.0061) model_time 0.9434 (1.0048) loss 0.6828 (0.8050) grad_norm 8.0866 (8.6226/1.9738) mem 68106MB [2022-12-20 15:17:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][920/1519] eta 0:10:02 lr 0.000004 time 0.9322 (1.0061) model_time 0.9321 (1.0048) loss 0.6974 (0.8046) grad_norm 7.6313 (8.5818/1.8994) mem 68106MB [2022-12-20 15:17:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][930/1519] eta 0:09:52 lr 0.000004 time 0.9348 (1.0061) model_time 0.9347 (1.0048) loss 0.7835 (0.8043) grad_norm 6.5821 (8.5489/1.8798) mem 68106MB [2022-12-20 15:17:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][940/1519] eta 0:09:42 lr 0.000004 time 0.9643 (1.0060) model_time 0.9641 (1.0047) loss 0.7882 (0.8038) grad_norm 8.5538 (8.5478/1.8730) mem 68106MB [2022-12-20 15:17:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][950/1519] eta 0:09:32 lr 0.000004 time 0.9448 (1.0060) model_time 0.9445 (1.0047) loss 0.6628 (0.8039) grad_norm 4.9946 (8.5329/1.8873) mem 68106MB [2022-12-20 15:18:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][960/1519] eta 0:09:22 lr 0.000004 time 0.9294 (1.0059) model_time 0.9293 (1.0047) loss 0.8436 (0.8044) grad_norm 9.2481 (8.5220/1.8616) mem 68106MB [2022-12-20 15:18:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][970/1519] eta 0:09:12 lr 0.000004 time 0.9301 (1.0059) model_time 0.9300 (1.0046) loss 0.7701 (0.8042) grad_norm 13.7945 (8.5293/1.8797) mem 68106MB [2022-12-20 15:18:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][980/1519] eta 0:09:02 lr 0.000004 time 0.9350 (1.0058) model_time 0.9348 (1.0046) loss 0.8737 (0.8037) grad_norm 7.0788 (8.5177/1.8851) mem 68106MB [2022-12-20 15:18:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][990/1519] eta 0:08:52 lr 0.000004 time 0.9295 (1.0058) model_time 0.9293 (1.0046) loss 0.6667 (0.8030) grad_norm 9.2685 (8.5414/1.8785) mem 68106MB [2022-12-20 15:18:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1000/1519] eta 0:08:41 lr 0.000004 time 0.9320 (1.0057) model_time 0.9318 (1.0045) loss 0.6993 (0.8027) grad_norm 8.2505 (8.5092/1.7590) mem 68106MB [2022-12-20 15:18:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1010/1519] eta 0:08:31 lr 0.000004 time 0.9270 (1.0057) model_time 0.9268 (1.0045) loss 0.7106 (0.8028) grad_norm 8.6180 (8.4609/1.7375) mem 68106MB [2022-12-20 15:19:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1020/1519] eta 0:08:21 lr 0.000004 time 0.9314 (1.0057) model_time 0.9313 (1.0045) loss 0.7548 (0.8033) grad_norm 13.3544 (8.4875/1.7648) mem 68106MB [2022-12-20 15:19:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1030/1519] eta 0:08:11 lr 0.000004 time 0.9377 (1.0056) model_time 0.9376 (1.0044) loss 1.1560 (0.8031) grad_norm 7.2896 (8.4695/1.7633) mem 68106MB [2022-12-20 15:19:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1040/1519] eta 0:08:01 lr 0.000004 time 0.9419 (1.0056) model_time 0.9418 (1.0044) loss 0.8102 (0.8031) grad_norm 7.2555 (8.4349/1.7639) mem 68106MB [2022-12-20 15:19:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1050/1519] eta 0:07:51 lr 0.000004 time 0.9324 (1.0055) model_time 0.9323 (1.0044) loss 0.9014 (0.8031) grad_norm 7.2631 (8.4261/1.7603) mem 68106MB [2022-12-20 15:19:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1060/1519] eta 0:07:41 lr 0.000004 time 0.9371 (1.0056) model_time 0.9370 (1.0044) loss 0.6847 (0.8031) grad_norm 7.7246 (8.4232/1.8170) mem 68106MB [2022-12-20 15:19:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1070/1519] eta 0:07:31 lr 0.000004 time 0.9309 (1.0055) model_time 0.9307 (1.0044) loss 0.7791 (0.8025) grad_norm 7.8373 (8.4387/1.8336) mem 68106MB [2022-12-20 15:20:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1080/1519] eta 0:07:21 lr 0.000004 time 0.9313 (1.0055) model_time 0.9311 (1.0044) loss 0.7884 (0.8026) grad_norm 7.3966 (8.4373/1.8555) mem 68106MB [2022-12-20 15:20:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1090/1519] eta 0:07:11 lr 0.000004 time 0.9625 (1.0055) model_time 0.9623 (1.0043) loss 0.8781 (0.8029) grad_norm 8.8945 (8.4704/1.8953) mem 68106MB [2022-12-20 15:20:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1100/1519] eta 0:07:01 lr 0.000004 time 0.9312 (1.0054) model_time 0.9311 (1.0043) loss 0.7075 (0.8030) grad_norm 9.1231 (8.4943/1.9054) mem 68106MB [2022-12-20 15:20:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1110/1519] eta 0:06:51 lr 0.000004 time 0.9314 (1.0059) model_time 0.9313 (1.0048) loss 0.9514 (0.8030) grad_norm 9.3450 (8.4846/1.8515) mem 68106MB [2022-12-20 15:20:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1120/1519] eta 0:06:41 lr 0.000004 time 0.9306 (1.0059) model_time 0.9304 (1.0047) loss 0.9495 (0.8035) grad_norm 5.6219 (8.5127/1.8674) mem 68106MB [2022-12-20 15:20:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1130/1519] eta 0:06:31 lr 0.000004 time 0.9306 (1.0059) model_time 0.9304 (1.0048) loss 0.7063 (0.8036) grad_norm 9.3469 (8.5185/1.8831) mem 68106MB [2022-12-20 15:21:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1140/1519] eta 0:06:21 lr 0.000004 time 0.9345 (1.0059) model_time 0.9344 (1.0048) loss 0.7177 (0.8033) grad_norm 7.3965 (8.5458/1.8985) mem 68106MB [2022-12-20 15:21:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1150/1519] eta 0:06:11 lr 0.000004 time 0.9339 (1.0062) model_time 0.9338 (1.0051) loss 0.8363 (0.8031) grad_norm 7.8804 (8.5248/1.8998) mem 68106MB [2022-12-20 15:21:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1160/1519] eta 0:06:01 lr 0.000004 time 0.9349 (1.0061) model_time 0.9348 (1.0050) loss 0.7358 (0.8029) grad_norm 11.5276 (8.5545/1.9320) mem 68106MB [2022-12-20 15:21:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1170/1519] eta 0:05:51 lr 0.000004 time 0.9355 (1.0061) model_time 0.9354 (1.0050) loss 0.6746 (0.8030) grad_norm 5.6965 (8.5521/1.9322) mem 68106MB [2022-12-20 15:21:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1180/1519] eta 0:05:41 lr 0.000004 time 0.9313 (1.0061) model_time 0.9312 (1.0050) loss 0.7607 (0.8035) grad_norm 6.7376 (8.5841/1.9485) mem 68106MB [2022-12-20 15:21:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1190/1519] eta 0:05:31 lr 0.000004 time 0.9267 (1.0061) model_time 0.9265 (1.0051) loss 1.0771 (0.8039) grad_norm 9.4580 (8.5827/1.9516) mem 68106MB [2022-12-20 15:22:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1200/1519] eta 0:05:20 lr 0.000004 time 0.9313 (1.0061) model_time 0.9312 (1.0050) loss 0.7193 (0.8042) grad_norm 6.8830 (8.5864/1.9487) mem 68106MB [2022-12-20 15:22:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1210/1519] eta 0:05:10 lr 0.000004 time 0.9314 (1.0061) model_time 0.9313 (1.0050) loss 0.8039 (0.8053) grad_norm 9.8100 (8.6019/1.9549) mem 68106MB [2022-12-20 15:22:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1220/1519] eta 0:05:00 lr 0.000004 time 0.9290 (1.0061) model_time 0.9288 (1.0050) loss 0.7983 (0.8048) grad_norm 5.6559 (8.5621/1.9673) mem 68106MB [2022-12-20 15:22:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1230/1519] eta 0:04:50 lr 0.000004 time 0.9372 (1.0061) model_time 0.9371 (1.0050) loss 0.6785 (0.8046) grad_norm 7.5323 (8.5621/1.9527) mem 68106MB [2022-12-20 15:22:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1240/1519] eta 0:04:40 lr 0.000004 time 0.9334 (1.0060) model_time 0.9332 (1.0050) loss 0.8333 (0.8049) grad_norm 6.0417 (8.5641/1.9557) mem 68106MB [2022-12-20 15:22:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1250/1519] eta 0:04:30 lr 0.000004 time 0.9323 (1.0060) model_time 0.9322 (1.0049) loss 0.8813 (0.8050) grad_norm 6.9598 (8.5834/2.0117) mem 68106MB [2022-12-20 15:23:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1260/1519] eta 0:04:20 lr 0.000004 time 0.9818 (1.0060) model_time 0.9816 (1.0050) loss 0.7612 (0.8053) grad_norm 8.3983 (8.6313/2.3214) mem 68106MB [2022-12-20 15:23:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1270/1519] eta 0:04:10 lr 0.000004 time 0.9317 (1.0059) model_time 0.9315 (1.0049) loss 0.6808 (0.8050) grad_norm 12.4089 (8.6252/2.3162) mem 68106MB [2022-12-20 15:23:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1280/1519] eta 0:04:00 lr 0.000004 time 0.9301 (1.0059) model_time 0.9299 (1.0048) loss 0.6956 (0.8052) grad_norm 10.4898 (8.6492/2.3309) mem 68106MB [2022-12-20 15:23:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1290/1519] eta 0:03:50 lr 0.000004 time 0.9331 (1.0059) model_time 0.9329 (1.0049) loss 0.6721 (0.8053) grad_norm 12.2460 (8.6387/2.3326) mem 68106MB [2022-12-20 15:23:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1300/1519] eta 0:03:40 lr 0.000004 time 0.9375 (1.0059) model_time 0.9373 (1.0049) loss 0.6596 (0.8053) grad_norm 8.5288 (8.6583/2.3233) mem 68106MB [2022-12-20 15:23:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1310/1519] eta 0:03:30 lr 0.000004 time 0.9329 (1.0059) model_time 0.9328 (1.0050) loss 0.6690 (0.8050) grad_norm 6.2069 (8.6503/2.3284) mem 68106MB [2022-12-20 15:24:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1320/1519] eta 0:03:20 lr 0.000004 time 0.9332 (1.0061) model_time 0.9331 (1.0051) loss 0.9317 (0.8047) grad_norm 8.3900 (8.6366/2.3320) mem 68106MB [2022-12-20 15:24:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1330/1519] eta 0:03:10 lr 0.000004 time 0.9304 (1.0061) model_time 0.9302 (1.0051) loss 0.7685 (0.8055) grad_norm 5.9034 (8.6330/2.3390) mem 68106MB [2022-12-20 15:24:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1340/1519] eta 0:03:00 lr 0.000004 time 0.9317 (1.0060) model_time 0.9315 (1.0050) loss 0.7247 (0.8053) grad_norm 8.2610 (8.6277/2.3316) mem 68106MB [2022-12-20 15:24:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1350/1519] eta 0:02:50 lr 0.000004 time 0.9319 (1.0059) model_time 0.9318 (1.0050) loss 0.6655 (0.8049) grad_norm 9.3241 (8.6391/2.3266) mem 68106MB [2022-12-20 15:24:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1360/1519] eta 0:02:39 lr 0.000004 time 0.9343 (1.0059) model_time 0.9341 (1.0049) loss 0.7124 (0.8046) grad_norm 7.7415 (8.5902/2.2641) mem 68106MB [2022-12-20 15:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1370/1519] eta 0:02:29 lr 0.000004 time 0.9333 (1.0059) model_time 0.9332 (1.0050) loss 0.7606 (0.8045) grad_norm 9.6335 (8.5587/2.2572) mem 68106MB [2022-12-20 15:25:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1380/1519] eta 0:02:19 lr 0.000004 time 0.9336 (1.0059) model_time 0.9334 (1.0049) loss 0.8287 (0.8049) grad_norm 8.1710 (8.5544/2.2543) mem 68106MB [2022-12-20 15:25:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1390/1519] eta 0:02:09 lr 0.000004 time 0.9707 (1.0059) model_time 0.9706 (1.0049) loss 0.8246 (0.8044) grad_norm 7.3063 (8.5657/2.2790) mem 68106MB [2022-12-20 15:25:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1400/1519] eta 0:01:59 lr 0.000004 time 0.9305 (1.0058) model_time 0.9303 (1.0049) loss 0.7670 (0.8043) grad_norm 6.7747 (8.5511/2.2935) mem 68106MB [2022-12-20 15:25:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1410/1519] eta 0:01:49 lr 0.000004 time 0.9272 (1.0058) model_time 0.9270 (1.0049) loss 0.7872 (0.8043) grad_norm 9.5153 (8.5810/2.2918) mem 68106MB [2022-12-20 15:25:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1420/1519] eta 0:01:39 lr 0.000004 time 0.9314 (1.0058) model_time 0.9313 (1.0049) loss 1.0982 (0.8051) grad_norm 11.4999 (8.5864/2.2886) mem 68106MB [2022-12-20 15:26:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1430/1519] eta 0:01:29 lr 0.000004 time 0.9371 (1.0058) model_time 0.9369 (1.0048) loss 0.8407 (0.8051) grad_norm 7.8470 (8.5910/2.2847) mem 68106MB [2022-12-20 15:26:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1440/1519] eta 0:01:19 lr 0.000004 time 1.0105 (1.0058) model_time 1.0104 (1.0049) loss 0.7618 (0.8056) grad_norm 8.1474 (8.5863/2.2794) mem 68106MB [2022-12-20 15:26:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1450/1519] eta 0:01:09 lr 0.000004 time 0.9333 (1.0058) model_time 0.9330 (1.0048) loss 0.7330 (0.8055) grad_norm 7.6422 (8.5714/2.2643) mem 68106MB [2022-12-20 15:26:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1460/1519] eta 0:00:59 lr 0.000004 time 0.9304 (1.0058) model_time 0.9303 (1.0049) loss 1.0578 (0.8056) grad_norm 8.2775 (8.5411/2.2448) mem 68106MB [2022-12-20 15:26:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1470/1519] eta 0:00:49 lr 0.000004 time 0.9362 (1.0058) model_time 0.9361 (1.0049) loss 0.7684 (0.8055) grad_norm 14.1566 (8.5719/2.2672) mem 68106MB [2022-12-20 15:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1480/1519] eta 0:00:39 lr 0.000004 time 0.9368 (1.0058) model_time 0.9367 (1.0049) loss 0.7206 (0.8054) grad_norm 8.5583 (8.5958/2.2816) mem 68106MB [2022-12-20 15:27:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1490/1519] eta 0:00:29 lr 0.000004 time 0.9352 (1.0058) model_time 0.9351 (1.0049) loss 0.9622 (0.8050) grad_norm 7.0813 (8.5631/2.2772) mem 68106MB [2022-12-20 15:27:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1500/1519] eta 0:00:19 lr 0.000004 time 0.9353 (1.0057) model_time 0.9352 (1.0048) loss 0.7374 (0.8052) grad_norm 7.3693 (8.5506/2.2583) mem 68106MB [2022-12-20 15:27:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [79/100][1510/1519] eta 0:00:09 lr 0.000004 time 0.9355 (1.0057) model_time 0.9354 (1.0048) loss 1.0536 (0.8051) grad_norm 6.3429 (8.5867/2.3222) mem 68106MB [2022-12-20 15:27:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 79 training takes 0:25:27 [2022-12-20 15:27:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_79.pth saving...... [2022-12-20 15:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_79.pth saved !!! [2022-12-20 15:27:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.690 (0.690) Loss 0.5299 (0.5299) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 15:27:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.334) Loss 0.5207 (0.5023) Acc@1 92.361 (92.551) Acc@5 97.917 (98.485) Mem 68106MB [2022-12-20 15:28:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.317) Loss 0.4934 (0.4981) Acc@1 90.625 (92.642) Acc@5 98.958 (98.479) Mem 68106MB [2022-12-20 15:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.311) Loss 0.6305 (0.5050) Acc@1 90.625 (92.462) Acc@5 98.264 (98.477) Mem 68106MB [2022-12-20 15:28:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.303 (0.308) Loss 0.4662 (0.4957) Acc@1 93.750 (92.581) Acc@5 98.958 (98.577) Mem 68106MB [2022-12-20 15:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.307) Loss 0.4889 (0.4935) Acc@1 92.014 (92.647) Acc@5 99.653 (98.611) Mem 68106MB [2022-12-20 15:28:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.300 (0.306) Loss 0.5062 (0.4930) Acc@1 90.972 (92.594) Acc@5 98.264 (98.571) Mem 68106MB [2022-12-20 15:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5409 (0.4944) Acc@1 92.361 (92.508) Acc@5 97.917 (98.543) Mem 68106MB [2022-12-20 15:28:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.296 (0.303) Loss 0.4329 (0.4928) Acc@1 93.403 (92.575) Acc@5 98.264 (98.564) Mem 68106MB [2022-12-20 15:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:79] * Acc@1 92.526 Acc@5 98.567 [2022-12-20 15:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 15:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 15:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 15:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.53% [2022-12-20 15:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][0/1519] eta 0:35:30 lr 0.000004 time 1.4024 (1.4024) model_time 0.9364 (0.9364) loss 0.8855 (0.8855) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 15:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][10/1519] eta 0:26:10 lr 0.000004 time 0.9224 (1.0404) model_time 0.9222 (0.9976) loss 0.8345 (0.7736) grad_norm 12.5971 (8.9995/1.8259) mem 68106MB [2022-12-20 15:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][20/1519] eta 0:25:37 lr 0.000004 time 0.9571 (1.0256) model_time 0.9570 (1.0029) loss 0.8590 (0.7950) grad_norm 10.5483 (9.9411/3.2470) mem 68106MB [2022-12-20 15:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][30/1519] eta 0:25:14 lr 0.000004 time 0.9327 (1.0175) model_time 0.9325 (1.0020) loss 0.8649 (0.7895) grad_norm 9.0083 (9.5985/2.8884) mem 68106MB [2022-12-20 15:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][40/1519] eta 0:24:57 lr 0.000004 time 0.9273 (1.0125) model_time 0.9271 (1.0007) loss 0.7150 (0.8132) grad_norm 8.5526 (9.2153/2.5975) mem 68106MB [2022-12-20 15:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][50/1519] eta 0:24:44 lr 0.000004 time 0.9281 (1.0105) model_time 0.9279 (1.0008) loss 0.7504 (0.8049) grad_norm 7.9269 (9.0389/2.4057) mem 68106MB [2022-12-20 15:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][60/1519] eta 0:24:31 lr 0.000004 time 0.9230 (1.0084) model_time 0.9227 (1.0003) loss 1.0924 (0.8172) grad_norm 11.5737 (9.0534/2.2713) mem 68106MB [2022-12-20 15:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][70/1519] eta 0:24:20 lr 0.000004 time 0.9410 (1.0078) model_time 0.9409 (1.0008) loss 0.7358 (0.8128) grad_norm 6.7406 (8.8776/2.1574) mem 68106MB [2022-12-20 15:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][80/1519] eta 0:24:09 lr 0.000004 time 0.9124 (1.0070) model_time 0.9122 (1.0008) loss 0.7993 (0.8091) grad_norm 7.6017 (8.8742/2.0880) mem 68106MB [2022-12-20 15:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][90/1519] eta 0:23:59 lr 0.000004 time 0.9427 (1.0072) model_time 0.9424 (1.0016) loss 0.8919 (0.8158) grad_norm 8.6767 (8.9567/2.0349) mem 68106MB [2022-12-20 15:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][100/1519] eta 0:23:49 lr 0.000004 time 1.0315 (1.0073) model_time 1.0313 (1.0023) loss 0.6747 (0.8161) grad_norm 7.7699 (8.8470/1.9909) mem 68106MB [2022-12-20 15:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][110/1519] eta 0:23:40 lr 0.000004 time 0.9877 (1.0081) model_time 0.9875 (1.0034) loss 0.7818 (0.8064) grad_norm 15.6043 (8.8992/2.1418) mem 68106MB [2022-12-20 15:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][120/1519] eta 0:23:31 lr 0.000004 time 0.9266 (1.0093) model_time 0.9265 (1.0050) loss 0.7264 (0.8038) grad_norm 7.2420 (8.8406/2.0950) mem 68106MB [2022-12-20 15:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][130/1519] eta 0:23:21 lr 0.000004 time 0.9193 (1.0092) model_time 0.9191 (1.0052) loss 0.8057 (0.8090) grad_norm 6.6606 (8.8160/2.0317) mem 68106MB [2022-12-20 15:31:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][140/1519] eta 0:23:13 lr 0.000003 time 0.9022 (1.0108) model_time 0.9019 (1.0071) loss 0.7154 (0.8116) grad_norm 7.3017 (8.8086/2.0147) mem 68106MB [2022-12-20 15:31:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][150/1519] eta 0:23:02 lr 0.000003 time 0.9312 (1.0101) model_time 0.9310 (1.0066) loss 0.6793 (0.8099) grad_norm 12.2960 (8.8387/2.0042) mem 68106MB [2022-12-20 15:31:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][160/1519] eta 0:22:53 lr 0.000003 time 0.9873 (1.0106) model_time 0.9872 (1.0072) loss 0.7887 (0.8088) grad_norm 8.7164 (8.8214/1.9672) mem 68106MB [2022-12-20 15:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][170/1519] eta 0:22:42 lr 0.000003 time 0.9356 (1.0098) model_time 0.9354 (1.0067) loss 0.7125 (0.8107) grad_norm 9.4778 (8.7944/1.9218) mem 68106MB [2022-12-20 15:31:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][180/1519] eta 0:22:31 lr 0.000003 time 0.9352 (1.0091) model_time 0.9350 (1.0061) loss 1.0637 (0.8132) grad_norm 9.5598 (8.7339/1.8995) mem 68106MB [2022-12-20 15:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][190/1519] eta 0:22:20 lr 0.000003 time 0.9278 (1.0085) model_time 0.9276 (1.0056) loss 0.6578 (0.8100) grad_norm 8.7933 (8.7072/1.9067) mem 68106MB [2022-12-20 15:32:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][200/1519] eta 0:22:10 lr 0.000003 time 0.9410 (1.0084) model_time 0.9408 (1.0056) loss 0.6764 (0.8070) grad_norm 7.2638 (8.7022/1.8924) mem 68106MB [2022-12-20 15:32:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][210/1519] eta 0:22:00 lr 0.000003 time 0.9344 (1.0084) model_time 0.9343 (1.0058) loss 0.7184 (0.8063) grad_norm 5.9343 (8.6779/1.8905) mem 68106MB [2022-12-20 15:32:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][220/1519] eta 0:21:49 lr 0.000003 time 0.9366 (1.0080) model_time 0.9364 (1.0055) loss 0.6714 (0.8029) grad_norm 6.7126 (8.7013/1.9468) mem 68106MB [2022-12-20 15:32:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][230/1519] eta 0:21:38 lr 0.000003 time 0.9317 (1.0077) model_time 0.9315 (1.0053) loss 0.8204 (0.8030) grad_norm 6.9684 (8.6997/1.9343) mem 68106MB [2022-12-20 15:32:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][240/1519] eta 0:21:28 lr 0.000003 time 0.9284 (1.0075) model_time 0.9283 (1.0052) loss 1.0397 (0.8031) grad_norm 7.0437 (8.6864/1.9375) mem 68106MB [2022-12-20 15:32:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][250/1519] eta 0:21:18 lr 0.000003 time 0.9354 (1.0074) model_time 0.9352 (1.0051) loss 0.8778 (0.8017) grad_norm 5.7701 (8.6520/1.9221) mem 68106MB [2022-12-20 15:33:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][260/1519] eta 0:21:08 lr 0.000003 time 0.9390 (1.0074) model_time 0.9387 (1.0052) loss 0.9421 (0.8003) grad_norm 6.2208 (8.6718/1.9619) mem 68106MB [2022-12-20 15:33:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][270/1519] eta 0:20:58 lr 0.000003 time 0.9577 (1.0072) model_time 0.9575 (1.0051) loss 0.8422 (0.8007) grad_norm 8.2709 (8.6588/1.9352) mem 68106MB [2022-12-20 15:33:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][280/1519] eta 0:20:47 lr 0.000003 time 0.9293 (1.0070) model_time 0.9292 (1.0049) loss 0.8081 (0.7991) grad_norm 8.8116 (8.6190/1.9162) mem 68106MB [2022-12-20 15:33:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][290/1519] eta 0:20:37 lr 0.000003 time 0.9310 (1.0067) model_time 0.9308 (1.0047) loss 0.8332 (0.7973) grad_norm 6.8220 (8.6290/1.9000) mem 68106MB [2022-12-20 15:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][300/1519] eta 0:20:27 lr 0.000003 time 0.9159 (1.0067) model_time 0.9156 (1.0047) loss 1.0516 (0.7958) grad_norm 7.2486 (8.6131/1.8822) mem 68106MB [2022-12-20 15:33:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][310/1519] eta 0:20:17 lr 0.000003 time 0.9185 (1.0068) model_time 0.9176 (1.0048) loss 0.9386 (0.7968) grad_norm 7.9889 (8.6481/1.8686) mem 68106MB [2022-12-20 15:34:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][320/1519] eta 0:20:07 lr 0.000003 time 0.9351 (1.0068) model_time 0.9350 (1.0049) loss 1.1125 (0.7981) grad_norm 10.4929 (8.6778/1.8614) mem 68106MB [2022-12-20 15:34:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][330/1519] eta 0:19:57 lr 0.000003 time 0.9335 (1.0068) model_time 0.9334 (1.0049) loss 0.8692 (0.7978) grad_norm 6.5213 (8.6983/1.8954) mem 68106MB [2022-12-20 15:34:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][340/1519] eta 0:19:47 lr 0.000003 time 0.9879 (1.0073) model_time 0.9877 (1.0055) loss 1.2145 (0.7986) grad_norm 6.6889 (8.7418/2.0346) mem 68106MB [2022-12-20 15:34:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][350/1519] eta 0:19:37 lr 0.000003 time 0.9259 (1.0072) model_time 0.9258 (1.0054) loss 0.7007 (0.7975) grad_norm 8.8675 (8.7760/2.0480) mem 68106MB [2022-12-20 15:34:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][360/1519] eta 0:19:27 lr 0.000003 time 0.9291 (1.0069) model_time 0.9290 (1.0052) loss 0.8157 (0.7993) grad_norm 6.4263 (8.7592/2.0314) mem 68106MB [2022-12-20 15:34:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][370/1519] eta 0:19:16 lr 0.000003 time 0.9346 (1.0067) model_time 0.9345 (1.0050) loss 0.8063 (0.7998) grad_norm 13.7514 (8.7955/2.0478) mem 68106MB [2022-12-20 15:35:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][380/1519] eta 0:19:06 lr 0.000003 time 0.9255 (1.0065) model_time 0.9253 (1.0048) loss 0.7922 (0.7982) grad_norm 8.5733 (8.8112/2.0425) mem 68106MB [2022-12-20 15:35:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][390/1519] eta 0:18:56 lr 0.000003 time 0.9710 (1.0064) model_time 0.9708 (1.0048) loss 0.8387 (0.7999) grad_norm 7.8505 (8.7923/2.0207) mem 68106MB [2022-12-20 15:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][400/1519] eta 0:18:46 lr 0.000003 time 0.9299 (1.0063) model_time 0.9297 (1.0047) loss 0.8064 (0.8001) grad_norm 9.4166 (8.7773/2.0184) mem 68106MB [2022-12-20 15:35:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][410/1519] eta 0:18:35 lr 0.000003 time 0.9530 (1.0063) model_time 0.9529 (1.0047) loss 1.0376 (0.8005) grad_norm 7.4341 (8.7733/2.0020) mem 68106MB [2022-12-20 15:35:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][420/1519] eta 0:18:27 lr 0.000003 time 0.9339 (1.0077) model_time 0.9338 (1.0062) loss 0.7271 (0.7995) grad_norm 6.7440 (8.8114/2.0273) mem 68106MB [2022-12-20 15:35:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][430/1519] eta 0:18:17 lr 0.000003 time 0.9302 (1.0077) model_time 0.9300 (1.0062) loss 0.6604 (0.7996) grad_norm 8.5972 (8.7797/2.0183) mem 68106MB [2022-12-20 15:36:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][440/1519] eta 0:18:07 lr 0.000003 time 0.9194 (1.0075) model_time 0.9193 (1.0060) loss 0.7746 (0.7977) grad_norm 6.1172 (8.7754/2.0176) mem 68106MB [2022-12-20 15:36:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][450/1519] eta 0:17:57 lr 0.000003 time 0.9253 (1.0076) model_time 0.9251 (1.0061) loss 1.0119 (0.7980) grad_norm 7.3014 (8.7661/2.0045) mem 68106MB [2022-12-20 15:36:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][460/1519] eta 0:17:47 lr 0.000003 time 0.9283 (1.0076) model_time 0.9281 (1.0062) loss 0.6741 (0.7981) grad_norm 8.9346 (8.7730/1.9986) mem 68106MB [2022-12-20 15:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][470/1519] eta 0:17:37 lr 0.000003 time 0.9935 (1.0077) model_time 0.9933 (1.0063) loss 0.6748 (0.7982) grad_norm 6.1459 (8.7485/1.9882) mem 68106MB [2022-12-20 15:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][480/1519] eta 0:17:26 lr 0.000003 time 0.9205 (1.0076) model_time 0.9202 (1.0062) loss 0.6646 (0.7986) grad_norm 10.1553 (8.7473/1.9739) mem 68106MB [2022-12-20 15:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][490/1519] eta 0:17:16 lr 0.000003 time 0.9540 (1.0075) model_time 0.9538 (1.0061) loss 0.7728 (0.7997) grad_norm 10.9835 (8.7464/1.9703) mem 68106MB [2022-12-20 15:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][500/1519] eta 0:17:06 lr 0.000003 time 0.9365 (1.0074) model_time 0.9364 (1.0060) loss 0.7527 (0.8000) grad_norm 16.0736 (8.7788/2.0122) mem 68106MB [2022-12-20 15:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][510/1519] eta 0:16:56 lr 0.000003 time 0.9309 (1.0071) model_time 0.9308 (1.0058) loss 0.9800 (0.7991) grad_norm 7.3007 (8.7808/2.0052) mem 68106MB [2022-12-20 15:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][520/1519] eta 0:16:46 lr 0.000003 time 0.9839 (1.0074) model_time 0.9837 (1.0061) loss 0.7282 (0.7996) grad_norm 7.1794 (8.7939/2.0336) mem 68106MB [2022-12-20 15:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][530/1519] eta 0:16:36 lr 0.000003 time 0.9220 (1.0073) model_time 0.9218 (1.0060) loss 0.7205 (0.7990) grad_norm 11.8012 (8.8216/2.0408) mem 68106MB [2022-12-20 15:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][540/1519] eta 0:16:26 lr 0.000003 time 0.9352 (1.0072) model_time 0.9350 (1.0060) loss 0.7708 (0.7983) grad_norm 10.2990 (8.8202/2.0272) mem 68106MB [2022-12-20 15:37:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][550/1519] eta 0:16:15 lr 0.000003 time 0.9228 (1.0071) model_time 0.9226 (1.0059) loss 0.8864 (0.7976) grad_norm 11.1674 (8.8222/2.0372) mem 68106MB [2022-12-20 15:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][560/1519] eta 0:16:05 lr 0.000003 time 0.9267 (1.0070) model_time 0.9265 (1.0058) loss 0.9362 (0.7978) grad_norm 6.6035 (8.7989/2.0347) mem 68106MB [2022-12-20 15:38:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][570/1519] eta 0:15:55 lr 0.000003 time 1.0112 (1.0071) model_time 1.0110 (1.0059) loss 0.8723 (0.7992) grad_norm 7.7989 (8.8287/2.0941) mem 68106MB [2022-12-20 15:38:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][580/1519] eta 0:15:45 lr 0.000003 time 0.9165 (1.0071) model_time 0.9163 (1.0059) loss 1.0815 (0.7989) grad_norm 6.9253 (8.8497/2.1027) mem 68106MB [2022-12-20 15:38:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][590/1519] eta 0:15:35 lr 0.000003 time 0.9289 (1.0070) model_time 0.9287 (1.0058) loss 0.6719 (0.7988) grad_norm 12.6048 (8.8856/2.1218) mem 68106MB [2022-12-20 15:38:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][600/1519] eta 0:15:25 lr 0.000003 time 0.9226 (1.0075) model_time 0.9221 (1.0063) loss 1.0377 (0.7997) grad_norm 8.1305 (8.8577/2.1174) mem 68106MB [2022-12-20 15:38:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][610/1519] eta 0:15:15 lr 0.000003 time 0.9730 (1.0074) model_time 0.9728 (1.0062) loss 0.8341 (0.7990) grad_norm 7.8399 (8.8503/2.1227) mem 68106MB [2022-12-20 15:39:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][620/1519] eta 0:15:05 lr 0.000003 time 0.9312 (1.0073) model_time 0.9309 (1.0062) loss 1.2897 (0.7996) grad_norm 8.2084 (8.8178/2.0614) mem 68106MB [2022-12-20 15:39:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][630/1519] eta 0:14:55 lr 0.000003 time 0.9601 (1.0074) model_time 0.9599 (1.0063) loss 0.9687 (0.7995) grad_norm 7.7137 (8.8358/2.0656) mem 68106MB [2022-12-20 15:39:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][640/1519] eta 0:14:45 lr 0.000003 time 0.9359 (1.0073) model_time 0.9358 (1.0062) loss 0.6972 (0.7990) grad_norm 8.2168 (8.8718/2.0844) mem 68106MB [2022-12-20 15:39:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][650/1519] eta 0:14:35 lr 0.000003 time 0.9996 (1.0075) model_time 0.9994 (1.0063) loss 0.6765 (0.7988) grad_norm 8.0518 (8.8658/2.0846) mem 68106MB [2022-12-20 15:39:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][660/1519] eta 0:14:25 lr 0.000003 time 0.9360 (1.0075) model_time 0.9359 (1.0063) loss 0.7169 (0.7998) grad_norm 9.5413 (8.8532/2.0812) mem 68106MB [2022-12-20 15:40:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][670/1519] eta 0:14:15 lr 0.000003 time 0.9399 (1.0074) model_time 0.9397 (1.0062) loss 0.8231 (0.7998) grad_norm 10.2623 (8.8811/2.0845) mem 68106MB [2022-12-20 15:40:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][680/1519] eta 0:14:05 lr 0.000003 time 0.9350 (1.0073) model_time 0.9349 (1.0062) loss 0.7701 (0.7996) grad_norm 9.8179 (8.8789/2.0796) mem 68106MB [2022-12-20 15:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][690/1519] eta 0:13:54 lr 0.000003 time 0.9404 (1.0072) model_time 0.9403 (1.0061) loss 0.7177 (0.7986) grad_norm 6.9453 (8.8597/2.0912) mem 68106MB [2022-12-20 15:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][700/1519] eta 0:13:44 lr 0.000003 time 0.9429 (1.0071) model_time 0.9428 (1.0060) loss 0.6954 (0.7986) grad_norm 5.8120 (8.8571/2.0932) mem 68106MB [2022-12-20 15:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][710/1519] eta 0:13:34 lr 0.000003 time 0.9474 (1.0071) model_time 0.9473 (1.0060) loss 0.6925 (0.7981) grad_norm 6.7719 (8.8355/2.0616) mem 68106MB [2022-12-20 15:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][720/1519] eta 0:13:24 lr 0.000003 time 0.9365 (1.0071) model_time 0.9364 (1.0060) loss 0.9774 (0.7978) grad_norm 11.4373 (8.8521/2.0662) mem 68106MB [2022-12-20 15:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][730/1519] eta 0:13:14 lr 0.000003 time 0.9626 (1.0071) model_time 0.9624 (1.0061) loss 0.7424 (0.7986) grad_norm 9.0775 (8.8586/2.0658) mem 68106MB [2022-12-20 15:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][740/1519] eta 0:13:04 lr 0.000003 time 0.9313 (1.0072) model_time 0.9312 (1.0061) loss 0.8293 (0.7981) grad_norm 9.4495 (8.8650/2.0719) mem 68106MB [2022-12-20 15:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][750/1519] eta 0:12:54 lr 0.000003 time 0.9449 (1.0071) model_time 0.9447 (1.0061) loss 0.7140 (0.7982) grad_norm 8.1672 (8.8609/2.0708) mem 68106MB [2022-12-20 15:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][760/1519] eta 0:12:44 lr 0.000003 time 0.9354 (1.0072) model_time 0.9352 (1.0062) loss 0.6910 (0.7975) grad_norm 13.2468 (8.8672/2.0838) mem 68106MB [2022-12-20 15:41:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][770/1519] eta 0:12:34 lr 0.000003 time 0.9288 (1.0073) model_time 0.9286 (1.0062) loss 0.6730 (0.7971) grad_norm 8.2709 (8.8374/2.1043) mem 68106MB [2022-12-20 15:41:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][780/1519] eta 0:12:24 lr 0.000003 time 0.9235 (1.0072) model_time 0.9233 (1.0061) loss 0.8100 (0.7974) grad_norm 8.8107 (8.8536/2.1015) mem 68106MB [2022-12-20 15:42:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][790/1519] eta 0:12:14 lr 0.000003 time 0.9999 (1.0072) model_time 0.9996 (1.0062) loss 0.7605 (0.7982) grad_norm 8.0270 (8.8564/2.0996) mem 68106MB [2022-12-20 15:42:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][800/1519] eta 0:12:04 lr 0.000003 time 0.9215 (1.0071) model_time 0.9213 (1.0061) loss 1.0233 (0.7981) grad_norm 7.1385 (8.8708/2.1093) mem 68106MB [2022-12-20 15:42:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][810/1519] eta 0:11:54 lr 0.000003 time 0.9477 (1.0071) model_time 0.9475 (1.0061) loss 1.2219 (0.7992) grad_norm 8.0681 (8.8780/2.1250) mem 68106MB [2022-12-20 15:42:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][820/1519] eta 0:11:43 lr 0.000003 time 0.9285 (1.0070) model_time 0.9284 (1.0060) loss 0.8102 (0.7991) grad_norm 6.0044 (8.8369/2.1160) mem 68106MB [2022-12-20 15:42:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][830/1519] eta 0:11:34 lr 0.000003 time 0.9523 (1.0074) model_time 0.9520 (1.0064) loss 0.9380 (0.7993) grad_norm 7.1871 (8.8439/2.1170) mem 68106MB [2022-12-20 15:42:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][840/1519] eta 0:11:24 lr 0.000003 time 0.9321 (1.0074) model_time 0.9320 (1.0064) loss 0.9985 (0.8002) grad_norm 7.9327 (8.8351/2.1087) mem 68106MB [2022-12-20 15:43:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][850/1519] eta 0:11:13 lr 0.000003 time 0.9235 (1.0073) model_time 0.9233 (1.0063) loss 0.9671 (0.8017) grad_norm 7.2898 (8.8336/2.1111) mem 68106MB [2022-12-20 15:43:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][860/1519] eta 0:11:03 lr 0.000003 time 0.9353 (1.0072) model_time 0.9351 (1.0062) loss 0.9324 (0.8023) grad_norm 14.0467 (8.8409/2.1159) mem 68106MB [2022-12-20 15:43:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][870/1519] eta 0:10:53 lr 0.000003 time 0.9395 (1.0072) model_time 0.9393 (1.0062) loss 0.8424 (0.8028) grad_norm 6.1342 (8.8314/2.1222) mem 68106MB [2022-12-20 15:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][880/1519] eta 0:10:43 lr 0.000003 time 0.9323 (1.0071) model_time 0.9322 (1.0061) loss 1.1007 (0.8033) grad_norm 9.3792 (8.8380/2.1227) mem 68106MB [2022-12-20 15:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][890/1519] eta 0:10:33 lr 0.000003 time 0.9282 (1.0070) model_time 0.9277 (1.0061) loss 0.6697 (0.8036) grad_norm 14.5494 (8.8367/2.1549) mem 68106MB [2022-12-20 15:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][900/1519] eta 0:10:23 lr 0.000003 time 0.9261 (1.0070) model_time 0.9259 (1.0061) loss 0.7525 (0.8034) grad_norm 7.9605 (8.8451/2.1524) mem 68106MB [2022-12-20 15:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][910/1519] eta 0:10:13 lr 0.000003 time 0.9268 (1.0070) model_time 0.9265 (1.0061) loss 0.6647 (0.8027) grad_norm 6.8281 (8.8613/2.2627) mem 68106MB [2022-12-20 15:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][920/1519] eta 0:10:03 lr 0.000003 time 0.9254 (1.0071) model_time 0.9251 (1.0061) loss 1.2313 (0.8039) grad_norm 6.3111 (8.8245/2.2678) mem 68106MB [2022-12-20 15:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][930/1519] eta 0:09:53 lr 0.000003 time 0.9057 (1.0070) model_time 0.9056 (1.0061) loss 0.7605 (0.8037) grad_norm 8.3253 (8.7971/2.2461) mem 68106MB [2022-12-20 15:44:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][940/1519] eta 0:09:43 lr 0.000003 time 0.9285 (1.0071) model_time 0.9284 (1.0062) loss 0.6698 (0.8029) grad_norm 12.3270 (8.7757/2.1758) mem 68106MB [2022-12-20 15:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][950/1519] eta 0:09:33 lr 0.000003 time 0.9344 (1.0072) model_time 0.9342 (1.0063) loss 0.6964 (0.8036) grad_norm 9.5263 (8.7442/2.1617) mem 68106MB [2022-12-20 15:44:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][960/1519] eta 0:09:23 lr 0.000003 time 0.9298 (1.0072) model_time 0.9297 (1.0063) loss 0.9620 (0.8034) grad_norm 6.9686 (8.7516/2.1622) mem 68106MB [2022-12-20 15:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][970/1519] eta 0:09:12 lr 0.000003 time 0.9338 (1.0073) model_time 0.9336 (1.0064) loss 0.9224 (0.8037) grad_norm 7.3976 (8.7250/2.1392) mem 68106MB [2022-12-20 15:45:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][980/1519] eta 0:09:02 lr 0.000003 time 0.9306 (1.0072) model_time 0.9305 (1.0063) loss 0.7381 (0.8037) grad_norm 6.7445 (8.7007/2.1311) mem 68106MB [2022-12-20 15:45:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][990/1519] eta 0:08:52 lr 0.000003 time 0.9313 (1.0071) model_time 0.9312 (1.0062) loss 0.6980 (0.8028) grad_norm 7.8374 (8.7062/2.1309) mem 68106MB [2022-12-20 15:45:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1000/1519] eta 0:08:42 lr 0.000003 time 0.9256 (1.0070) model_time 0.9254 (1.0062) loss 0.7426 (0.8022) grad_norm 9.5777 (8.7291/2.1340) mem 68106MB [2022-12-20 15:45:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1010/1519] eta 0:08:32 lr 0.000003 time 0.9254 (1.0070) model_time 0.9253 (1.0062) loss 0.7281 (0.8025) grad_norm 6.8989 (8.7359/2.1422) mem 68106MB [2022-12-20 15:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1020/1519] eta 0:08:22 lr 0.000003 time 0.9227 (1.0070) model_time 0.9225 (1.0061) loss 0.7556 (0.8020) grad_norm 11.5019 (8.7183/2.1197) mem 68106MB [2022-12-20 15:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1030/1519] eta 0:08:12 lr 0.000003 time 0.9218 (1.0069) model_time 0.9216 (1.0061) loss 0.6779 (0.8016) grad_norm 10.7094 (8.7124/2.1344) mem 68106MB [2022-12-20 15:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1040/1519] eta 0:08:02 lr 0.000003 time 0.9059 (1.0069) model_time 0.9057 (1.0061) loss 0.9276 (0.8021) grad_norm 6.9091 (8.7101/2.1254) mem 68106MB [2022-12-20 15:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1050/1519] eta 0:07:52 lr 0.000003 time 0.9489 (1.0072) model_time 0.9487 (1.0063) loss 0.6754 (0.8024) grad_norm 8.9688 (8.7083/2.1250) mem 68106MB [2022-12-20 15:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1060/1519] eta 0:07:42 lr 0.000003 time 0.9342 (1.0071) model_time 0.9340 (1.0062) loss 0.7938 (0.8023) grad_norm 10.0225 (8.6996/2.1282) mem 68106MB [2022-12-20 15:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1070/1519] eta 0:07:32 lr 0.000003 time 0.9383 (1.0070) model_time 0.9381 (1.0062) loss 0.7493 (0.8019) grad_norm 6.4271 (8.6902/2.1354) mem 68106MB [2022-12-20 15:46:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1080/1519] eta 0:07:22 lr 0.000003 time 0.9253 (1.0070) model_time 0.9252 (1.0062) loss 1.0046 (0.8020) grad_norm 8.2949 (8.7085/2.1409) mem 68106MB [2022-12-20 15:47:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1090/1519] eta 0:07:11 lr 0.000003 time 0.9244 (1.0069) model_time 0.9243 (1.0061) loss 0.6850 (0.8022) grad_norm 7.9172 (8.7050/2.1327) mem 68106MB [2022-12-20 15:47:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1100/1519] eta 0:07:01 lr 0.000003 time 0.9285 (1.0070) model_time 0.9283 (1.0061) loss 1.2629 (0.8028) grad_norm 7.9325 (8.6726/2.0889) mem 68106MB [2022-12-20 15:47:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1110/1519] eta 0:06:51 lr 0.000003 time 0.9274 (1.0069) model_time 0.9273 (1.0061) loss 0.8335 (0.8030) grad_norm 9.7141 (8.6631/2.0855) mem 68106MB [2022-12-20 15:47:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1120/1519] eta 0:06:41 lr 0.000003 time 0.9256 (1.0069) model_time 0.9254 (1.0061) loss 0.6892 (0.8034) grad_norm 7.9399 (8.6508/2.0649) mem 68106MB [2022-12-20 15:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1130/1519] eta 0:06:31 lr 0.000003 time 0.9362 (1.0070) model_time 0.9361 (1.0061) loss 1.0785 (0.8038) grad_norm 7.5135 (8.6040/2.0513) mem 68106MB [2022-12-20 15:47:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1140/1519] eta 0:06:21 lr 0.000003 time 0.9245 (1.0071) model_time 0.9244 (1.0063) loss 1.0193 (0.8043) grad_norm 7.3945 (8.6012/2.0497) mem 68106MB [2022-12-20 15:48:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1150/1519] eta 0:06:11 lr 0.000003 time 0.9301 (1.0071) model_time 0.9300 (1.0063) loss 0.8208 (0.8050) grad_norm 6.7844 (8.5574/2.0465) mem 68106MB [2022-12-20 15:48:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1160/1519] eta 0:06:01 lr 0.000003 time 0.9346 (1.0071) model_time 0.9345 (1.0063) loss 0.7025 (0.8049) grad_norm 6.7338 (8.5812/2.0442) mem 68106MB [2022-12-20 15:48:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1170/1519] eta 0:05:51 lr 0.000003 time 0.9366 (1.0070) model_time 0.9365 (1.0062) loss 0.6872 (0.8047) grad_norm 17.5071 (8.5828/2.0466) mem 68106MB [2022-12-20 15:48:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1180/1519] eta 0:05:41 lr 0.000003 time 0.9267 (1.0070) model_time 0.9265 (1.0062) loss 0.7185 (0.8050) grad_norm 6.2389 (8.5512/2.0236) mem 68106MB [2022-12-20 15:48:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1190/1519] eta 0:05:31 lr 0.000003 time 0.9303 (1.0069) model_time 0.9302 (1.0061) loss 0.7539 (0.8049) grad_norm 8.1163 (8.5154/1.9865) mem 68106MB [2022-12-20 15:48:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1200/1519] eta 0:05:21 lr 0.000003 time 0.9464 (1.0069) model_time 0.9462 (1.0061) loss 1.0477 (0.8056) grad_norm 11.1764 (8.5281/1.9914) mem 68106MB [2022-12-20 15:49:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1210/1519] eta 0:05:11 lr 0.000003 time 0.9275 (1.0069) model_time 0.9274 (1.0061) loss 0.7586 (0.8054) grad_norm 7.3966 (8.5625/2.0243) mem 68106MB [2022-12-20 15:49:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1220/1519] eta 0:05:01 lr 0.000003 time 0.9190 (1.0071) model_time 0.9185 (1.0063) loss 1.0474 (0.8055) grad_norm 9.4428 (8.5356/2.0163) mem 68106MB [2022-12-20 15:49:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1230/1519] eta 0:04:51 lr 0.000003 time 0.8994 (1.0071) model_time 0.8993 (1.0063) loss 0.7903 (0.8054) grad_norm 8.2013 (8.5243/2.0065) mem 68106MB [2022-12-20 15:49:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1240/1519] eta 0:04:40 lr 0.000003 time 0.9192 (1.0070) model_time 0.9191 (1.0063) loss 0.8493 (0.8054) grad_norm 8.1850 (8.5032/1.9872) mem 68106MB [2022-12-20 15:49:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1250/1519] eta 0:04:30 lr 0.000003 time 0.9276 (1.0070) model_time 0.9274 (1.0062) loss 0.6763 (0.8049) grad_norm 6.9942 (8.4930/1.9882) mem 68106MB [2022-12-20 15:49:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1260/1519] eta 0:04:20 lr 0.000003 time 0.9256 (1.0070) model_time 0.9255 (1.0063) loss 0.7571 (0.8050) grad_norm 13.6462 (8.5099/2.0127) mem 68106MB [2022-12-20 15:50:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1270/1519] eta 0:04:10 lr 0.000003 time 0.9291 (1.0070) model_time 0.9283 (1.0062) loss 0.8057 (0.8043) grad_norm 7.4706 (8.4840/2.0085) mem 68106MB [2022-12-20 15:50:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1280/1519] eta 0:04:00 lr 0.000003 time 0.9258 (1.0070) model_time 0.9256 (1.0063) loss 0.6644 (0.8041) grad_norm 10.8496 (8.4917/2.0169) mem 68106MB [2022-12-20 15:50:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1290/1519] eta 0:03:50 lr 0.000003 time 0.9219 (1.0070) model_time 0.9212 (1.0062) loss 0.8788 (0.8043) grad_norm 8.3014 (8.4865/2.0094) mem 68106MB [2022-12-20 15:50:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1300/1519] eta 0:03:40 lr 0.000003 time 0.9342 (1.0069) model_time 0.9340 (1.0062) loss 0.9060 (0.8047) grad_norm 6.6640 (8.4883/2.0064) mem 68106MB [2022-12-20 15:50:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1310/1519] eta 0:03:30 lr 0.000003 time 0.9372 (1.0069) model_time 0.9371 (1.0061) loss 0.6742 (0.8045) grad_norm 7.0709 (8.4759/2.0012) mem 68106MB [2022-12-20 15:50:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1320/1519] eta 0:03:20 lr 0.000003 time 0.9180 (1.0069) model_time 0.9178 (1.0061) loss 0.6663 (0.8047) grad_norm 7.0494 (8.4570/1.9920) mem 68106MB [2022-12-20 15:51:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1330/1519] eta 0:03:10 lr 0.000003 time 0.9367 (1.0068) model_time 0.9364 (1.0061) loss 0.8178 (0.8047) grad_norm 6.7930 (8.4252/1.9985) mem 68106MB [2022-12-20 15:51:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1340/1519] eta 0:03:00 lr 0.000003 time 0.9335 (1.0068) model_time 0.9333 (1.0060) loss 0.8426 (0.8051) grad_norm 9.3232 (8.4284/1.9838) mem 68106MB [2022-12-20 15:51:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1350/1519] eta 0:02:50 lr 0.000003 time 0.9300 (1.0067) model_time 0.9298 (1.0060) loss 0.8568 (0.8052) grad_norm 10.2263 (8.4239/1.9780) mem 68106MB [2022-12-20 15:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1360/1519] eta 0:02:40 lr 0.000003 time 0.9306 (1.0067) model_time 0.9305 (1.0059) loss 0.7314 (0.8052) grad_norm 11.0769 (8.4245/1.9750) mem 68106MB [2022-12-20 15:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1370/1519] eta 0:02:29 lr 0.000003 time 0.9311 (1.0067) model_time 0.9310 (1.0059) loss 0.6649 (0.8051) grad_norm 12.8983 (8.4736/1.9816) mem 68106MB [2022-12-20 15:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1380/1519] eta 0:02:19 lr 0.000003 time 0.9322 (1.0066) model_time 0.9320 (1.0059) loss 0.7338 (0.8048) grad_norm 10.3677 (8.4610/2.0010) mem 68106MB [2022-12-20 15:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1390/1519] eta 0:02:09 lr 0.000003 time 0.9257 (1.0066) model_time 0.9255 (1.0059) loss 0.6695 (0.8044) grad_norm 6.0549 (8.4540/1.9940) mem 68106MB [2022-12-20 15:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1400/1519] eta 0:01:59 lr 0.000003 time 0.9980 (1.0066) model_time 0.9979 (1.0059) loss 0.7471 (0.8041) grad_norm 10.4799 (8.4449/1.9769) mem 68106MB [2022-12-20 15:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1410/1519] eta 0:01:49 lr 0.000003 time 0.9319 (1.0066) model_time 0.9318 (1.0059) loss 1.0252 (0.8037) grad_norm 9.2303 (8.4392/1.9524) mem 68106MB [2022-12-20 15:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1420/1519] eta 0:01:39 lr 0.000003 time 0.9262 (1.0066) model_time 0.9260 (1.0058) loss 0.7505 (0.8036) grad_norm 9.8688 (8.4844/1.9514) mem 68106MB [2022-12-20 15:52:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1430/1519] eta 0:01:29 lr 0.000003 time 0.9232 (1.0067) model_time 0.9230 (1.0060) loss 0.6732 (0.8036) grad_norm 7.5404 (8.4803/1.9786) mem 68106MB [2022-12-20 15:52:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1440/1519] eta 0:01:19 lr 0.000003 time 0.9329 (1.0067) model_time 0.9326 (1.0060) loss 0.6636 (0.8032) grad_norm 7.3000 (8.4843/1.9758) mem 68106MB [2022-12-20 15:53:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1450/1519] eta 0:01:09 lr 0.000003 time 0.9818 (1.0068) model_time 0.9816 (1.0060) loss 0.9253 (0.8038) grad_norm 8.6624 (8.4957/1.9669) mem 68106MB [2022-12-20 15:53:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1460/1519] eta 0:00:59 lr 0.000003 time 0.9306 (1.0068) model_time 0.9305 (1.0061) loss 0.7316 (0.8042) grad_norm 15.5189 (8.5130/1.9744) mem 68106MB [2022-12-20 15:53:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1470/1519] eta 0:00:49 lr 0.000003 time 0.9215 (1.0067) model_time 0.9213 (1.0060) loss 0.6570 (0.8044) grad_norm 8.9290 (8.5291/1.9697) mem 68106MB [2022-12-20 15:53:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1480/1519] eta 0:00:39 lr 0.000003 time 0.9248 (1.0067) model_time 0.9246 (1.0060) loss 0.9466 (0.8044) grad_norm 12.8247 (8.5542/1.9804) mem 68106MB [2022-12-20 15:53:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1490/1519] eta 0:00:29 lr 0.000003 time 0.9294 (1.0067) model_time 0.9293 (1.0059) loss 0.8697 (0.8047) grad_norm 9.9032 (8.5452/1.9449) mem 68106MB [2022-12-20 15:53:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1500/1519] eta 0:00:19 lr 0.000003 time 0.9306 (1.0066) model_time 0.9305 (1.0059) loss 0.6955 (0.8046) grad_norm 6.0137 (8.5459/1.9487) mem 68106MB [2022-12-20 15:54:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [80/100][1510/1519] eta 0:00:09 lr 0.000003 time 0.9216 (1.0066) model_time 0.9215 (1.0059) loss 0.8227 (0.8046) grad_norm 6.2645 (8.5129/1.8495) mem 68106MB [2022-12-20 15:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 80 training takes 0:25:28 [2022-12-20 15:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_80.pth saving...... [2022-12-20 15:54:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_80.pth saved !!! [2022-12-20 15:54:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.5374 (0.5374) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 15:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.333) Loss 0.5354 (0.5102) Acc@1 92.014 (92.803) Acc@5 98.264 (98.516) Mem 68106MB [2022-12-20 15:54:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.317) Loss 0.4976 (0.5044) Acc@1 90.625 (92.824) Acc@5 98.958 (98.528) Mem 68106MB [2022-12-20 15:54:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.302 (0.311) Loss 0.6392 (0.5110) Acc@1 90.278 (92.451) Acc@5 97.569 (98.443) Mem 68106MB [2022-12-20 15:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.293 (0.308) Loss 0.4664 (0.5020) Acc@1 93.403 (92.514) Acc@5 98.958 (98.526) Mem 68106MB [2022-12-20 15:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.306) Loss 0.4919 (0.4999) Acc@1 91.319 (92.538) Acc@5 99.653 (98.570) Mem 68106MB [2022-12-20 15:54:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.306 (0.305) Loss 0.5136 (0.4997) Acc@1 91.667 (92.498) Acc@5 98.264 (98.537) Mem 68106MB [2022-12-20 15:55:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.5473 (0.5013) Acc@1 93.403 (92.425) Acc@5 97.917 (98.523) Mem 68106MB [2022-12-20 15:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.301 (0.303) Loss 0.4401 (0.4998) Acc@1 93.403 (92.477) Acc@5 98.264 (98.555) Mem 68106MB [2022-12-20 15:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:80] * Acc@1 92.440 Acc@5 98.559 [2022-12-20 15:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.4% [2022-12-20 15:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.53% [2022-12-20 15:55:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][0/1519] eta 0:47:38 lr 0.000003 time 1.8819 (1.8819) model_time 1.1132 (1.1132) loss 0.6796 (0.6796) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 15:55:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][10/1519] eta 0:27:32 lr 0.000003 time 1.0338 (1.0948) model_time 1.0336 (1.0245) loss 0.6730 (0.7460) grad_norm 7.3904 (10.5943/6.0198) mem 68106MB [2022-12-20 15:55:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][20/1519] eta 0:26:19 lr 0.000003 time 0.9294 (1.0537) model_time 0.9293 (1.0167) loss 0.9576 (0.7945) grad_norm 9.6715 (9.9719/4.4105) mem 68106MB [2022-12-20 15:55:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][30/1519] eta 0:25:44 lr 0.000003 time 0.9495 (1.0370) model_time 0.9493 (1.0119) loss 0.7093 (0.7855) grad_norm 8.6874 (9.6172/3.7225) mem 68106MB [2022-12-20 15:55:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][40/1519] eta 0:25:30 lr 0.000003 time 0.9776 (1.0351) model_time 0.9774 (1.0160) loss 0.6733 (0.7950) grad_norm 6.4319 (9.2336/3.3362) mem 68106MB [2022-12-20 15:55:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][50/1519] eta 0:25:17 lr 0.000003 time 1.0249 (1.0333) model_time 1.0248 (1.0178) loss 0.8147 (0.7997) grad_norm 6.5974 (8.8771/3.0877) mem 68106MB [2022-12-20 15:56:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][60/1519] eta 0:25:00 lr 0.000003 time 0.9284 (1.0282) model_time 0.9283 (1.0152) loss 0.9549 (0.7998) grad_norm 7.5136 (8.9672/2.9978) mem 68106MB [2022-12-20 15:56:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][70/1519] eta 0:24:43 lr 0.000003 time 0.9237 (1.0238) model_time 0.9236 (1.0127) loss 0.8232 (0.7977) grad_norm 5.6722 (8.7621/2.9367) mem 68106MB [2022-12-20 15:56:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][80/1519] eta 0:24:31 lr 0.000003 time 0.9696 (1.0223) model_time 0.9694 (1.0125) loss 0.9445 (0.7988) grad_norm 7.0749 (8.7949/2.8523) mem 68106MB [2022-12-20 15:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][90/1519] eta 0:24:18 lr 0.000003 time 0.9229 (1.0208) model_time 0.9227 (1.0120) loss 0.9362 (0.8089) grad_norm 10.2040 (8.7616/2.7324) mem 68106MB [2022-12-20 15:56:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][100/1519] eta 0:24:06 lr 0.000003 time 0.9577 (1.0193) model_time 0.9575 (1.0114) loss 1.1798 (0.8156) grad_norm 8.1434 (8.6478/2.6378) mem 68106MB [2022-12-20 15:56:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][110/1519] eta 0:23:54 lr 0.000003 time 0.9290 (1.0181) model_time 0.9289 (1.0108) loss 0.7042 (0.8143) grad_norm 11.1723 (8.6953/2.5557) mem 68106MB [2022-12-20 15:57:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][120/1519] eta 0:23:44 lr 0.000003 time 0.9337 (1.0179) model_time 0.9336 (1.0112) loss 0.8460 (0.8106) grad_norm 6.1061 (8.5658/2.4908) mem 68106MB [2022-12-20 15:57:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][130/1519] eta 0:23:31 lr 0.000003 time 0.9227 (1.0165) model_time 0.9226 (1.0102) loss 1.0496 (0.8121) grad_norm 8.9881 (8.6099/2.4150) mem 68106MB [2022-12-20 15:57:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][140/1519] eta 0:23:20 lr 0.000003 time 0.9154 (1.0154) model_time 0.9152 (1.0096) loss 0.6955 (0.8065) grad_norm 7.8944 (8.5679/2.3423) mem 68106MB [2022-12-20 15:57:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][150/1519] eta 0:23:08 lr 0.000003 time 0.9214 (1.0144) model_time 0.9212 (1.0089) loss 1.0377 (0.8139) grad_norm 8.7118 (8.5378/2.2929) mem 68106MB [2022-12-20 15:57:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][160/1519] eta 0:22:57 lr 0.000003 time 0.9141 (1.0133) model_time 0.9139 (1.0082) loss 0.6668 (0.8147) grad_norm 7.1801 (8.5704/2.2519) mem 68106MB [2022-12-20 15:57:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][170/1519] eta 0:22:45 lr 0.000003 time 0.9214 (1.0125) model_time 0.9212 (1.0077) loss 0.7663 (0.8125) grad_norm 11.3732 (8.5908/2.2291) mem 68106MB [2022-12-20 15:58:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][180/1519] eta 0:22:34 lr 0.000003 time 0.9292 (1.0117) model_time 0.9291 (1.0071) loss 0.9123 (0.8107) grad_norm 6.4723 (8.5477/2.1926) mem 68106MB [2022-12-20 15:58:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][190/1519] eta 0:22:23 lr 0.000003 time 0.9261 (1.0111) model_time 0.9259 (1.0068) loss 0.7921 (0.8131) grad_norm 8.3746 (8.5753/2.1874) mem 68106MB [2022-12-20 15:58:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][200/1519] eta 0:22:13 lr 0.000003 time 0.9249 (1.0108) model_time 0.9248 (1.0066) loss 0.8691 (0.8137) grad_norm 8.4006 (8.5977/2.1694) mem 68106MB [2022-12-20 15:58:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][210/1519] eta 0:22:04 lr 0.000003 time 0.9231 (1.0115) model_time 0.9229 (1.0075) loss 0.8640 (0.8137) grad_norm 9.6271 (8.6147/2.1287) mem 68106MB [2022-12-20 15:58:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][220/1519] eta 0:21:53 lr 0.000003 time 1.0127 (1.0112) model_time 1.0126 (1.0074) loss 0.9545 (0.8146) grad_norm 7.9485 (8.5970/2.0889) mem 68106MB [2022-12-20 15:58:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][230/1519] eta 0:21:44 lr 0.000003 time 0.9964 (1.0121) model_time 0.9963 (1.0084) loss 0.6663 (0.8133) grad_norm 13.2135 (8.5891/2.1101) mem 68106MB [2022-12-20 15:59:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][240/1519] eta 0:21:33 lr 0.000003 time 0.9213 (1.0117) model_time 0.9212 (1.0082) loss 0.8933 (0.8143) grad_norm 10.6491 (8.5689/2.0924) mem 68106MB [2022-12-20 15:59:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][250/1519] eta 0:21:23 lr 0.000003 time 0.9347 (1.0112) model_time 0.9345 (1.0078) loss 0.6673 (0.8160) grad_norm 7.8560 (8.5608/2.0665) mem 68106MB [2022-12-20 15:59:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][260/1519] eta 0:21:12 lr 0.000003 time 0.9451 (1.0108) model_time 0.9448 (1.0075) loss 0.6949 (0.8147) grad_norm 8.0446 (8.5426/2.0410) mem 68106MB [2022-12-20 15:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][270/1519] eta 0:21:01 lr 0.000003 time 0.9191 (1.0104) model_time 0.9190 (1.0072) loss 0.8161 (0.8150) grad_norm 6.0510 (8.5261/2.0180) mem 68106MB [2022-12-20 15:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][280/1519] eta 0:20:51 lr 0.000003 time 0.9333 (1.0100) model_time 0.9332 (1.0070) loss 0.7152 (0.8148) grad_norm 12.0194 (8.5554/2.0098) mem 68106MB [2022-12-20 15:59:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][290/1519] eta 0:20:40 lr 0.000003 time 0.9332 (1.0098) model_time 0.9330 (1.0068) loss 0.7610 (0.8122) grad_norm 8.7586 (8.6079/2.0137) mem 68106MB [2022-12-20 16:00:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][300/1519] eta 0:20:32 lr 0.000003 time 0.9441 (1.0108) model_time 0.9440 (1.0079) loss 0.9786 (0.8121) grad_norm 9.1317 (8.6390/2.0340) mem 68106MB [2022-12-20 16:00:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][310/1519] eta 0:20:21 lr 0.000003 time 0.9220 (1.0105) model_time 0.9218 (1.0077) loss 0.6824 (0.8100) grad_norm 7.5688 (8.6691/2.0487) mem 68106MB [2022-12-20 16:00:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][320/1519] eta 0:20:11 lr 0.000003 time 0.9341 (1.0105) model_time 0.9339 (1.0077) loss 0.7170 (0.8112) grad_norm 9.1820 (8.6723/2.0411) mem 68106MB [2022-12-20 16:00:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][330/1519] eta 0:20:02 lr 0.000003 time 0.9263 (1.0114) model_time 0.9261 (1.0087) loss 0.6988 (0.8100) grad_norm 7.5527 (8.6747/2.0250) mem 68106MB [2022-12-20 16:00:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][340/1519] eta 0:19:52 lr 0.000003 time 0.9328 (1.0111) model_time 0.9326 (1.0085) loss 1.1557 (0.8093) grad_norm 7.5785 (8.6659/2.0039) mem 68106MB [2022-12-20 16:00:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][350/1519] eta 0:19:41 lr 0.000003 time 0.9411 (1.0110) model_time 0.9409 (1.0084) loss 0.7779 (0.8073) grad_norm 8.2633 (8.6446/1.9895) mem 68106MB [2022-12-20 16:01:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][360/1519] eta 0:19:31 lr 0.000003 time 0.9366 (1.0112) model_time 0.9364 (1.0087) loss 0.7670 (0.8065) grad_norm 8.4166 (8.6329/1.9757) mem 68106MB [2022-12-20 16:01:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][370/1519] eta 0:19:21 lr 0.000003 time 0.9328 (1.0108) model_time 0.9326 (1.0084) loss 0.7674 (0.8067) grad_norm 7.8412 (8.6349/1.9559) mem 68106MB [2022-12-20 16:01:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][380/1519] eta 0:19:11 lr 0.000003 time 0.9270 (1.0106) model_time 0.9268 (1.0082) loss 0.8352 (0.8077) grad_norm 7.2641 (8.6207/1.9483) mem 68106MB [2022-12-20 16:01:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][390/1519] eta 0:19:00 lr 0.000003 time 0.9103 (1.0105) model_time 0.9101 (1.0082) loss 0.7725 (0.8082) grad_norm 7.6648 (8.6063/1.9285) mem 68106MB [2022-12-20 16:01:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][400/1519] eta 0:18:51 lr 0.000003 time 1.0706 (1.0108) model_time 1.0705 (1.0085) loss 0.8511 (0.8064) grad_norm 10.6161 (8.5919/1.9194) mem 68106MB [2022-12-20 16:02:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][410/1519] eta 0:18:40 lr 0.000003 time 0.9186 (1.0106) model_time 0.9184 (1.0084) loss 0.6555 (0.8049) grad_norm 7.6585 (8.5966/1.9221) mem 68106MB [2022-12-20 16:02:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][420/1519] eta 0:18:30 lr 0.000003 time 0.9311 (1.0104) model_time 0.9309 (1.0082) loss 0.6526 (0.8030) grad_norm 8.6434 (8.5669/1.9422) mem 68106MB [2022-12-20 16:02:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][430/1519] eta 0:18:20 lr 0.000003 time 0.9217 (1.0103) model_time 0.9216 (1.0082) loss 0.7450 (0.8035) grad_norm 9.6269 (8.5589/1.9261) mem 68106MB [2022-12-20 16:02:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][440/1519] eta 0:18:09 lr 0.000003 time 0.9274 (1.0102) model_time 0.9273 (1.0080) loss 0.8994 (0.8055) grad_norm 9.1189 (8.5491/1.9084) mem 68106MB [2022-12-20 16:02:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][450/1519] eta 0:17:59 lr 0.000003 time 0.9171 (1.0099) model_time 0.9169 (1.0078) loss 0.7152 (0.8066) grad_norm 9.8831 (8.5487/1.8925) mem 68106MB [2022-12-20 16:02:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][460/1519] eta 0:17:49 lr 0.000003 time 0.9361 (1.0097) model_time 0.9360 (1.0076) loss 0.9127 (0.8065) grad_norm 9.8795 (8.5685/1.8948) mem 68106MB [2022-12-20 16:03:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][470/1519] eta 0:17:38 lr 0.000003 time 0.9286 (1.0095) model_time 0.9284 (1.0075) loss 1.0865 (0.8065) grad_norm 10.6611 (8.5992/1.9669) mem 68106MB [2022-12-20 16:03:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][480/1519] eta 0:17:28 lr 0.000003 time 0.9248 (1.0093) model_time 0.9247 (1.0073) loss 0.9802 (0.8069) grad_norm 9.6505 (8.6588/2.1423) mem 68106MB [2022-12-20 16:03:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][490/1519] eta 0:17:18 lr 0.000003 time 0.9370 (1.0091) model_time 0.9369 (1.0072) loss 0.6832 (0.8062) grad_norm 6.2951 (8.6567/2.1364) mem 68106MB [2022-12-20 16:03:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][500/1519] eta 0:17:08 lr 0.000003 time 0.9307 (1.0089) model_time 0.9305 (1.0070) loss 0.9286 (0.8086) grad_norm 7.0494 (8.6780/2.1707) mem 68106MB [2022-12-20 16:03:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][510/1519] eta 0:16:57 lr 0.000003 time 0.9344 (1.0088) model_time 0.9343 (1.0069) loss 0.9006 (0.8086) grad_norm 7.9237 (8.6572/2.1561) mem 68106MB [2022-12-20 16:03:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][520/1519] eta 0:16:47 lr 0.000003 time 0.9348 (1.0087) model_time 0.9347 (1.0069) loss 0.7259 (0.8097) grad_norm 12.2672 (8.6662/2.1549) mem 68106MB [2022-12-20 16:04:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][530/1519] eta 0:16:37 lr 0.000003 time 0.9441 (1.0089) model_time 0.9439 (1.0071) loss 1.0261 (0.8100) grad_norm 6.2171 (8.6525/2.1539) mem 68106MB [2022-12-20 16:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][540/1519] eta 0:16:27 lr 0.000003 time 0.9363 (1.0091) model_time 0.9362 (1.0073) loss 0.7785 (0.8088) grad_norm 6.5788 (8.6834/2.2146) mem 68106MB [2022-12-20 16:04:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][550/1519] eta 0:16:17 lr 0.000003 time 0.9337 (1.0091) model_time 0.9335 (1.0074) loss 0.7512 (0.8097) grad_norm 8.4922 (8.6764/2.2094) mem 68106MB [2022-12-20 16:04:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][560/1519] eta 0:16:07 lr 0.000003 time 0.9287 (1.0091) model_time 0.9285 (1.0073) loss 0.7366 (0.8095) grad_norm 9.0513 (8.6761/2.1979) mem 68106MB [2022-12-20 16:04:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][570/1519] eta 0:15:57 lr 0.000003 time 0.9383 (1.0090) model_time 0.9381 (1.0073) loss 0.6766 (0.8086) grad_norm 8.6705 (8.6770/2.2001) mem 68106MB [2022-12-20 16:04:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][580/1519] eta 0:15:47 lr 0.000003 time 0.9349 (1.0090) model_time 0.9348 (1.0073) loss 0.6890 (0.8085) grad_norm 9.1413 (8.6950/2.2239) mem 68106MB [2022-12-20 16:05:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][590/1519] eta 0:15:37 lr 0.000003 time 0.9321 (1.0089) model_time 0.9319 (1.0072) loss 0.6743 (0.8080) grad_norm 6.3859 (8.6906/2.2198) mem 68106MB [2022-12-20 16:05:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][600/1519] eta 0:15:27 lr 0.000003 time 0.9289 (1.0088) model_time 0.9287 (1.0071) loss 0.6671 (0.8081) grad_norm 16.2279 (8.7118/2.2466) mem 68106MB [2022-12-20 16:05:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][610/1519] eta 0:15:17 lr 0.000003 time 0.9495 (1.0090) model_time 0.9493 (1.0074) loss 0.8222 (0.8101) grad_norm 11.2617 (8.6865/2.1120) mem 68106MB [2022-12-20 16:05:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][620/1519] eta 0:15:06 lr 0.000003 time 0.9290 (1.0089) model_time 0.9288 (1.0073) loss 0.8897 (0.8101) grad_norm 6.0166 (8.6931/2.2141) mem 68106MB [2022-12-20 16:05:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][630/1519] eta 0:14:56 lr 0.000003 time 0.9266 (1.0089) model_time 0.9264 (1.0073) loss 1.1633 (0.8107) grad_norm 7.5472 (8.6775/2.2175) mem 68106MB [2022-12-20 16:05:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][640/1519] eta 0:14:46 lr 0.000003 time 0.9219 (1.0090) model_time 0.9215 (1.0074) loss 0.8739 (0.8112) grad_norm 7.1714 (8.6750/2.2158) mem 68106MB [2022-12-20 16:06:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][650/1519] eta 0:14:36 lr 0.000003 time 0.9270 (1.0088) model_time 0.9268 (1.0073) loss 0.7825 (0.8113) grad_norm 8.4434 (8.6881/2.2119) mem 68106MB [2022-12-20 16:06:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][660/1519] eta 0:14:26 lr 0.000003 time 0.9340 (1.0087) model_time 0.9339 (1.0072) loss 0.9509 (0.8118) grad_norm 10.6892 (8.7115/2.2614) mem 68106MB [2022-12-20 16:06:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][670/1519] eta 0:14:16 lr 0.000003 time 0.9216 (1.0088) model_time 0.9214 (1.0073) loss 0.7415 (0.8107) grad_norm 8.9857 (8.7424/2.2545) mem 68106MB [2022-12-20 16:06:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][680/1519] eta 0:14:06 lr 0.000003 time 0.9273 (1.0087) model_time 0.9270 (1.0072) loss 0.6730 (0.8097) grad_norm 10.6244 (8.7273/2.2485) mem 68106MB [2022-12-20 16:06:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][690/1519] eta 0:13:56 lr 0.000003 time 0.9279 (1.0086) model_time 0.9277 (1.0071) loss 0.8339 (0.8099) grad_norm 8.2369 (8.7262/2.2585) mem 68106MB [2022-12-20 16:06:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][700/1519] eta 0:13:45 lr 0.000003 time 0.9314 (1.0085) model_time 0.9313 (1.0070) loss 1.0824 (0.8109) grad_norm 10.7800 (8.7475/2.2593) mem 68106MB [2022-12-20 16:07:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][710/1519] eta 0:13:35 lr 0.000003 time 0.9619 (1.0085) model_time 0.9618 (1.0070) loss 0.8694 (0.8101) grad_norm 8.2891 (8.7436/2.2532) mem 68106MB [2022-12-20 16:07:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][720/1519] eta 0:13:26 lr 0.000003 time 0.9296 (1.0089) model_time 0.9295 (1.0074) loss 0.8113 (0.8097) grad_norm 6.3497 (8.7581/2.2531) mem 68106MB [2022-12-20 16:07:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][730/1519] eta 0:13:15 lr 0.000003 time 0.9168 (1.0088) model_time 0.9167 (1.0074) loss 0.8342 (0.8094) grad_norm 7.9748 (8.7378/2.2560) mem 68106MB [2022-12-20 16:07:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][740/1519] eta 0:13:05 lr 0.000003 time 0.9212 (1.0087) model_time 0.9210 (1.0073) loss 0.8699 (0.8097) grad_norm 7.4382 (8.7536/2.2618) mem 68106MB [2022-12-20 16:07:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][750/1519] eta 0:12:55 lr 0.000003 time 0.9320 (1.0086) model_time 0.9318 (1.0072) loss 0.6839 (0.8094) grad_norm 10.7349 (8.7704/2.2722) mem 68106MB [2022-12-20 16:07:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][760/1519] eta 0:12:45 lr 0.000003 time 0.9251 (1.0085) model_time 0.9250 (1.0071) loss 1.0221 (0.8083) grad_norm 7.2223 (8.7755/2.2740) mem 68106MB [2022-12-20 16:08:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][770/1519] eta 0:12:35 lr 0.000003 time 0.9304 (1.0084) model_time 0.9302 (1.0070) loss 0.6819 (0.8086) grad_norm 9.1654 (8.7782/2.2703) mem 68106MB [2022-12-20 16:08:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][780/1519] eta 0:12:25 lr 0.000003 time 0.9317 (1.0083) model_time 0.9316 (1.0069) loss 0.8728 (0.8080) grad_norm 7.1515 (8.7716/2.2703) mem 68106MB [2022-12-20 16:08:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][790/1519] eta 0:12:15 lr 0.000003 time 0.9229 (1.0083) model_time 0.9228 (1.0069) loss 0.7197 (0.8089) grad_norm 7.6722 (8.7417/2.2663) mem 68106MB [2022-12-20 16:08:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][800/1519] eta 0:12:04 lr 0.000003 time 0.9181 (1.0081) model_time 0.9180 (1.0068) loss 0.7488 (0.8082) grad_norm 13.0543 (8.7574/2.2769) mem 68106MB [2022-12-20 16:08:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][810/1519] eta 0:11:54 lr 0.000003 time 0.9237 (1.0080) model_time 0.9236 (1.0067) loss 0.7056 (0.8085) grad_norm 6.7317 (8.7393/2.2786) mem 68106MB [2022-12-20 16:08:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][820/1519] eta 0:11:44 lr 0.000003 time 0.9680 (1.0080) model_time 0.9678 (1.0066) loss 0.7157 (0.8089) grad_norm 7.1736 (8.7365/2.2786) mem 68106MB [2022-12-20 16:09:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][830/1519] eta 0:11:34 lr 0.000003 time 0.9198 (1.0080) model_time 0.9197 (1.0067) loss 1.0680 (0.8088) grad_norm 5.9785 (8.7303/2.2616) mem 68106MB [2022-12-20 16:09:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][840/1519] eta 0:11:24 lr 0.000003 time 0.9283 (1.0079) model_time 0.9281 (1.0066) loss 0.7184 (0.8080) grad_norm 10.2512 (8.7327/2.2593) mem 68106MB [2022-12-20 16:09:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][850/1519] eta 0:11:14 lr 0.000003 time 1.1797 (1.0083) model_time 1.1795 (1.0070) loss 0.8732 (0.8087) grad_norm 6.4996 (8.7235/2.2698) mem 68106MB [2022-12-20 16:09:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][860/1519] eta 0:11:04 lr 0.000003 time 0.9257 (1.0082) model_time 0.9256 (1.0070) loss 0.7196 (0.8089) grad_norm 10.4334 (8.7351/2.2806) mem 68106MB [2022-12-20 16:09:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][870/1519] eta 0:10:54 lr 0.000003 time 0.9361 (1.0081) model_time 0.9359 (1.0069) loss 0.6996 (0.8091) grad_norm 7.7694 (8.7238/2.2805) mem 68106MB [2022-12-20 16:09:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][880/1519] eta 0:10:44 lr 0.000003 time 0.9218 (1.0081) model_time 0.9217 (1.0068) loss 0.9744 (0.8095) grad_norm 9.5663 (8.7013/2.2771) mem 68106MB [2022-12-20 16:10:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][890/1519] eta 0:10:34 lr 0.000003 time 0.9251 (1.0081) model_time 0.9249 (1.0068) loss 0.7839 (0.8096) grad_norm 7.2969 (8.6722/2.2673) mem 68106MB [2022-12-20 16:10:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][900/1519] eta 0:10:23 lr 0.000003 time 0.9205 (1.0080) model_time 0.9203 (1.0068) loss 0.8632 (0.8090) grad_norm 6.4341 (8.6528/2.2483) mem 68106MB [2022-12-20 16:10:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][910/1519] eta 0:10:13 lr 0.000003 time 0.9282 (1.0079) model_time 0.9281 (1.0067) loss 0.9575 (0.8090) grad_norm 8.4074 (8.6395/2.2350) mem 68106MB [2022-12-20 16:10:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][920/1519] eta 0:10:03 lr 0.000003 time 0.9728 (1.0082) model_time 0.9727 (1.0070) loss 0.8510 (0.8089) grad_norm 8.4713 (8.6153/2.2360) mem 68106MB [2022-12-20 16:10:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][930/1519] eta 0:09:53 lr 0.000003 time 0.9309 (1.0081) model_time 0.9307 (1.0069) loss 0.6715 (0.8084) grad_norm 9.3949 (8.6119/2.2325) mem 68106MB [2022-12-20 16:10:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][940/1519] eta 0:09:43 lr 0.000003 time 1.0113 (1.0081) model_time 1.0111 (1.0069) loss 0.8320 (0.8081) grad_norm 6.7161 (8.6026/2.2330) mem 68106MB [2022-12-20 16:11:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][950/1519] eta 0:09:33 lr 0.000003 time 0.9181 (1.0081) model_time 0.9180 (1.0069) loss 0.7410 (0.8078) grad_norm 7.2501 (8.6023/2.2283) mem 68106MB [2022-12-20 16:11:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][960/1519] eta 0:09:23 lr 0.000003 time 0.9287 (1.0080) model_time 0.9286 (1.0069) loss 0.7196 (0.8075) grad_norm 6.7515 (8.5904/2.2295) mem 68106MB [2022-12-20 16:11:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][970/1519] eta 0:09:13 lr 0.000003 time 0.9308 (1.0079) model_time 0.9307 (1.0068) loss 0.7342 (0.8076) grad_norm 9.0665 (8.5918/2.2310) mem 68106MB [2022-12-20 16:11:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][980/1519] eta 0:09:03 lr 0.000003 time 0.9929 (1.0080) model_time 0.9927 (1.0068) loss 0.6733 (0.8075) grad_norm 7.8892 (8.5922/2.2220) mem 68106MB [2022-12-20 16:11:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][990/1519] eta 0:08:53 lr 0.000003 time 0.9181 (1.0079) model_time 0.9180 (1.0067) loss 0.6978 (0.8068) grad_norm 6.7060 (8.6152/2.2340) mem 68106MB [2022-12-20 16:11:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1000/1519] eta 0:08:43 lr 0.000003 time 0.9770 (1.0078) model_time 0.9768 (1.0067) loss 0.6668 (0.8067) grad_norm 7.9908 (8.6121/2.2292) mem 68106MB [2022-12-20 16:12:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1010/1519] eta 0:08:32 lr 0.000003 time 0.9280 (1.0078) model_time 0.9279 (1.0067) loss 1.0370 (0.8074) grad_norm 5.6560 (8.5847/2.2254) mem 68106MB [2022-12-20 16:12:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1020/1519] eta 0:08:22 lr 0.000003 time 0.9357 (1.0078) model_time 0.9355 (1.0067) loss 0.6925 (0.8076) grad_norm 9.5673 (8.6003/2.2038) mem 68106MB [2022-12-20 16:12:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1030/1519] eta 0:08:12 lr 0.000003 time 0.9473 (1.0079) model_time 0.9472 (1.0068) loss 0.6678 (0.8074) grad_norm 7.2931 (8.5785/2.2127) mem 68106MB [2022-12-20 16:12:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1040/1519] eta 0:08:02 lr 0.000003 time 0.9586 (1.0078) model_time 0.9584 (1.0067) loss 0.7504 (0.8071) grad_norm 12.3880 (8.5835/2.2257) mem 68106MB [2022-12-20 16:12:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1050/1519] eta 0:07:52 lr 0.000003 time 0.9280 (1.0078) model_time 0.9278 (1.0067) loss 0.7738 (0.8072) grad_norm 9.1685 (8.5753/2.2279) mem 68106MB [2022-12-20 16:12:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1060/1519] eta 0:07:42 lr 0.000003 time 0.9292 (1.0077) model_time 0.9291 (1.0066) loss 0.7727 (0.8072) grad_norm 8.3394 (8.5404/2.2197) mem 68106MB [2022-12-20 16:13:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1070/1519] eta 0:07:32 lr 0.000003 time 0.9212 (1.0076) model_time 0.9210 (1.0066) loss 0.6935 (0.8073) grad_norm 7.6904 (8.5072/2.1590) mem 68106MB [2022-12-20 16:13:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1080/1519] eta 0:07:22 lr 0.000003 time 0.9240 (1.0076) model_time 0.9238 (1.0065) loss 0.6620 (0.8075) grad_norm 6.1413 (8.4344/2.0117) mem 68106MB [2022-12-20 16:13:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1090/1519] eta 0:07:12 lr 0.000003 time 0.9313 (1.0075) model_time 0.9311 (1.0064) loss 0.8668 (0.8078) grad_norm 7.1037 (8.4135/2.0036) mem 68106MB [2022-12-20 16:13:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1100/1519] eta 0:07:02 lr 0.000003 time 0.9694 (1.0075) model_time 0.9693 (1.0065) loss 0.8556 (0.8077) grad_norm 7.9984 (8.4113/1.9711) mem 68106MB [2022-12-20 16:13:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1110/1519] eta 0:06:52 lr 0.000003 time 0.9305 (1.0075) model_time 0.9304 (1.0064) loss 0.6856 (0.8078) grad_norm 6.6907 (8.4172/1.9739) mem 68106MB [2022-12-20 16:13:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1120/1519] eta 0:06:41 lr 0.000003 time 0.9912 (1.0074) model_time 0.9911 (1.0064) loss 0.9409 (0.8075) grad_norm 9.5794 (8.4066/1.9629) mem 68106MB [2022-12-20 16:14:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1130/1519] eta 0:06:31 lr 0.000003 time 0.9289 (1.0075) model_time 0.9288 (1.0064) loss 0.7721 (0.8069) grad_norm 7.4639 (8.4227/1.9502) mem 68106MB [2022-12-20 16:14:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1140/1519] eta 0:06:21 lr 0.000003 time 0.9291 (1.0074) model_time 0.9289 (1.0064) loss 0.6989 (0.8059) grad_norm 5.1323 (8.3737/1.8799) mem 68106MB [2022-12-20 16:14:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1150/1519] eta 0:06:11 lr 0.000003 time 0.9168 (1.0074) model_time 0.9167 (1.0063) loss 0.8643 (0.8055) grad_norm 11.1930 (8.3918/1.8827) mem 68106MB [2022-12-20 16:14:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1160/1519] eta 0:06:01 lr 0.000003 time 0.9911 (1.0075) model_time 0.9907 (1.0064) loss 0.7006 (0.8053) grad_norm 10.2779 (8.3885/1.8870) mem 68106MB [2022-12-20 16:14:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1170/1519] eta 0:05:51 lr 0.000003 time 0.9265 (1.0075) model_time 0.9264 (1.0065) loss 0.7731 (0.8053) grad_norm 7.9233 (8.3824/1.8740) mem 68106MB [2022-12-20 16:14:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1180/1519] eta 0:05:41 lr 0.000003 time 0.9175 (1.0074) model_time 0.9174 (1.0064) loss 0.7664 (0.8054) grad_norm 7.6330 (8.3445/1.8282) mem 68106MB [2022-12-20 16:15:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1190/1519] eta 0:05:31 lr 0.000003 time 0.9566 (1.0074) model_time 0.9564 (1.0064) loss 0.7188 (0.8053) grad_norm 6.3497 (8.3476/1.8282) mem 68106MB [2022-12-20 16:15:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1200/1519] eta 0:05:21 lr 0.000003 time 0.8866 (1.0074) model_time 0.8865 (1.0064) loss 0.9146 (0.8053) grad_norm 9.1371 (8.3116/1.7738) mem 68106MB [2022-12-20 16:15:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1210/1519] eta 0:05:11 lr 0.000003 time 0.9233 (1.0074) model_time 0.9231 (1.0064) loss 0.7142 (0.8052) grad_norm 6.5641 (8.2890/1.7576) mem 68106MB [2022-12-20 16:15:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1220/1519] eta 0:05:01 lr 0.000003 time 0.9238 (1.0073) model_time 0.9236 (1.0064) loss 0.7805 (0.8049) grad_norm 8.5501 (8.2518/1.6169) mem 68106MB [2022-12-20 16:15:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1230/1519] eta 0:04:51 lr 0.000003 time 1.0070 (1.0074) model_time 1.0069 (1.0064) loss 1.1416 (0.8050) grad_norm 5.7686 (8.2288/1.6242) mem 68106MB [2022-12-20 16:15:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1240/1519] eta 0:04:41 lr 0.000003 time 0.9250 (1.0074) model_time 0.9248 (1.0064) loss 0.7279 (0.8049) grad_norm 9.9757 (8.2323/1.6318) mem 68106MB [2022-12-20 16:16:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1250/1519] eta 0:04:30 lr 0.000003 time 0.9179 (1.0073) model_time 0.9178 (1.0063) loss 0.8424 (0.8047) grad_norm 9.0781 (8.2849/1.7928) mem 68106MB [2022-12-20 16:16:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1260/1519] eta 0:04:20 lr 0.000003 time 0.9203 (1.0072) model_time 0.9201 (1.0063) loss 0.7108 (0.8046) grad_norm 9.0507 (8.2768/1.7366) mem 68106MB [2022-12-20 16:16:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1270/1519] eta 0:04:10 lr 0.000003 time 0.9465 (1.0072) model_time 0.9464 (1.0062) loss 0.8093 (0.8046) grad_norm 5.8104 (8.2563/1.7192) mem 68106MB [2022-12-20 16:16:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1280/1519] eta 0:04:00 lr 0.000003 time 0.9196 (1.0071) model_time 0.9194 (1.0062) loss 0.6934 (0.8046) grad_norm 5.7162 (8.2742/1.7463) mem 68106MB [2022-12-20 16:16:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1290/1519] eta 0:03:50 lr 0.000003 time 0.9295 (1.0071) model_time 0.9293 (1.0061) loss 0.9027 (0.8050) grad_norm 6.1459 (8.2480/1.7379) mem 68106MB [2022-12-20 16:16:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1300/1519] eta 0:03:40 lr 0.000003 time 0.9244 (1.0070) model_time 0.9242 (1.0061) loss 0.8546 (0.8057) grad_norm 7.5209 (8.2270/1.7310) mem 68106MB [2022-12-20 16:17:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1310/1519] eta 0:03:30 lr 0.000003 time 0.9296 (1.0070) model_time 0.9294 (1.0060) loss 0.7204 (0.8054) grad_norm 8.5791 (8.2462/1.7663) mem 68106MB [2022-12-20 16:17:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1320/1519] eta 0:03:20 lr 0.000003 time 0.9308 (1.0069) model_time 0.9306 (1.0060) loss 0.6812 (0.8053) grad_norm 8.7100 (8.2598/1.7595) mem 68106MB [2022-12-20 16:17:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1330/1519] eta 0:03:10 lr 0.000003 time 0.9323 (1.0071) model_time 0.9321 (1.0061) loss 1.0244 (0.8055) grad_norm 8.9011 (8.2707/1.7563) mem 68106MB [2022-12-20 16:17:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1340/1519] eta 0:03:00 lr 0.000003 time 0.9273 (1.0071) model_time 0.9271 (1.0061) loss 0.6860 (0.8052) grad_norm 6.5599 (8.2481/1.7457) mem 68106MB [2022-12-20 16:17:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1350/1519] eta 0:02:50 lr 0.000003 time 0.9363 (1.0071) model_time 0.9361 (1.0062) loss 0.7289 (0.8050) grad_norm 7.4611 (8.2387/1.7322) mem 68106MB [2022-12-20 16:17:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1360/1519] eta 0:02:40 lr 0.000003 time 0.9333 (1.0071) model_time 0.9330 (1.0062) loss 0.6687 (0.8052) grad_norm 10.2008 (8.2315/1.7276) mem 68106MB [2022-12-20 16:18:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1370/1519] eta 0:02:30 lr 0.000003 time 0.9330 (1.0071) model_time 0.9328 (1.0062) loss 0.6594 (0.8051) grad_norm 11.3877 (8.2401/1.7413) mem 68106MB [2022-12-20 16:18:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1380/1519] eta 0:02:19 lr 0.000003 time 0.9313 (1.0071) model_time 0.9310 (1.0062) loss 0.7084 (0.8048) grad_norm 8.6986 (8.2717/1.7775) mem 68106MB [2022-12-20 16:18:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1390/1519] eta 0:02:09 lr 0.000003 time 0.9240 (1.0070) model_time 0.9238 (1.0061) loss 0.6999 (0.8046) grad_norm 6.8858 (8.2875/1.7910) mem 68106MB [2022-12-20 16:18:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1400/1519] eta 0:01:59 lr 0.000003 time 0.9212 (1.0070) model_time 0.9210 (1.0061) loss 0.8124 (0.8047) grad_norm 9.5581 (8.2737/1.7887) mem 68106MB [2022-12-20 16:18:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1410/1519] eta 0:01:49 lr 0.000003 time 1.0127 (1.0070) model_time 1.0125 (1.0061) loss 0.6716 (0.8044) grad_norm 5.7485 (8.2628/1.7951) mem 68106MB [2022-12-20 16:18:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1420/1519] eta 0:01:39 lr 0.000003 time 0.9306 (1.0070) model_time 0.9304 (1.0061) loss 0.9905 (0.8043) grad_norm 5.4286 (8.2688/1.8176) mem 68106MB [2022-12-20 16:19:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1430/1519] eta 0:01:29 lr 0.000003 time 0.9395 (1.0069) model_time 0.9393 (1.0060) loss 0.9724 (0.8044) grad_norm 15.3655 (8.2962/1.8622) mem 68106MB [2022-12-20 16:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1440/1519] eta 0:01:19 lr 0.000003 time 0.9176 (1.0069) model_time 0.9174 (1.0060) loss 0.6732 (0.8041) grad_norm 10.5767 (8.3054/1.8631) mem 68106MB [2022-12-20 16:19:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1450/1519] eta 0:01:09 lr 0.000003 time 0.9280 (1.0070) model_time 0.9278 (1.0060) loss 0.7328 (0.8038) grad_norm 5.8603 (8.3106/1.8833) mem 68106MB [2022-12-20 16:19:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1460/1519] eta 0:00:59 lr 0.000003 time 0.9308 (1.0069) model_time 0.9305 (1.0060) loss 0.8240 (0.8039) grad_norm 4.8305 (8.3004/1.8809) mem 68106MB [2022-12-20 16:19:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1470/1519] eta 0:00:49 lr 0.000003 time 0.9202 (1.0070) model_time 0.9200 (1.0061) loss 0.8378 (0.8039) grad_norm 8.5812 (8.3091/1.8814) mem 68106MB [2022-12-20 16:19:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1480/1519] eta 0:00:39 lr 0.000003 time 0.9401 (1.0070) model_time 0.9399 (1.0061) loss 0.7778 (0.8040) grad_norm 8.7156 (8.3091/1.8774) mem 68106MB [2022-12-20 16:20:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1490/1519] eta 0:00:29 lr 0.000003 time 0.9364 (1.0069) model_time 0.9362 (1.0060) loss 1.0248 (0.8039) grad_norm 5.6552 (8.2994/1.8803) mem 68106MB [2022-12-20 16:20:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1500/1519] eta 0:00:19 lr 0.000003 time 0.9400 (1.0069) model_time 0.9397 (1.0060) loss 0.6629 (0.8035) grad_norm 8.7269 (8.3406/1.9201) mem 68106MB [2022-12-20 16:20:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [81/100][1510/1519] eta 0:00:09 lr 0.000003 time 0.9249 (1.0069) model_time 0.9247 (1.0060) loss 0.9245 (0.8037) grad_norm 7.9668 (8.3324/1.9167) mem 68106MB [2022-12-20 16:20:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 81 training takes 0:25:29 [2022-12-20 16:20:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_81.pth saving...... [2022-12-20 16:20:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_81.pth saved !!! [2022-12-20 16:20:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.632 (0.632) Loss 0.5402 (0.5402) Acc@1 92.014 (92.014) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 16:21:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.329) Loss 0.5359 (0.5075) Acc@1 92.361 (92.645) Acc@5 97.917 (98.453) Mem 68106MB [2022-12-20 16:21:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4982 (0.5024) Acc@1 90.972 (92.708) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 16:21:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.295 (0.309) Loss 0.6396 (0.5095) Acc@1 90.278 (92.462) Acc@5 98.264 (98.432) Mem 68106MB [2022-12-20 16:21:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.304 (0.307) Loss 0.4627 (0.5004) Acc@1 93.750 (92.556) Acc@5 99.306 (98.518) Mem 68106MB [2022-12-20 16:21:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.4929 (0.4979) Acc@1 91.667 (92.627) Acc@5 99.306 (98.584) Mem 68106MB [2022-12-20 16:21:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.303 (0.304) Loss 0.5107 (0.4974) Acc@1 90.625 (92.555) Acc@5 98.264 (98.554) Mem 68106MB [2022-12-20 16:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.303) Loss 0.5421 (0.4986) Acc@1 93.056 (92.459) Acc@5 97.917 (98.543) Mem 68106MB [2022-12-20 16:21:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.302) Loss 0.4322 (0.4971) Acc@1 93.403 (92.511) Acc@5 98.264 (98.564) Mem 68106MB [2022-12-20 16:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:81] * Acc@1 92.469 Acc@5 98.567 [2022-12-20 16:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 16:21:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.53% [2022-12-20 16:21:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][0/1519] eta 0:45:44 lr 0.000003 time 1.8066 (1.8066) model_time 1.1318 (1.1318) loss 0.9608 (0.9608) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 16:21:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][10/1519] eta 0:27:01 lr 0.000003 time 0.9383 (1.0748) model_time 0.9382 (1.0131) loss 0.8557 (0.8965) grad_norm 8.5675 (9.6980/2.2844) mem 68106MB [2022-12-20 16:21:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][20/1519] eta 0:25:59 lr 0.000003 time 0.9665 (1.0404) model_time 0.9663 (1.0079) loss 1.1683 (0.8726) grad_norm 11.0602 (9.2078/2.1827) mem 68106MB [2022-12-20 16:21:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][30/1519] eta 0:25:35 lr 0.000003 time 1.0014 (1.0313) model_time 1.0013 (1.0092) loss 0.7274 (0.8628) grad_norm 8.4487 (8.8457/1.9133) mem 68106MB [2022-12-20 16:22:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][40/1519] eta 0:25:13 lr 0.000003 time 0.9239 (1.0235) model_time 0.9237 (1.0066) loss 0.6772 (0.8405) grad_norm 7.9909 (9.0220/2.1034) mem 68106MB [2022-12-20 16:22:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][50/1519] eta 0:24:56 lr 0.000003 time 0.9316 (1.0188) model_time 0.9315 (1.0051) loss 0.7045 (0.8318) grad_norm 8.2673 (8.8941/2.0569) mem 68106MB [2022-12-20 16:22:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][60/1519] eta 0:24:42 lr 0.000003 time 0.9248 (1.0164) model_time 0.9247 (1.0050) loss 0.8326 (0.8216) grad_norm 8.2512 (8.8165/1.9829) mem 68106MB [2022-12-20 16:22:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][70/1519] eta 0:24:29 lr 0.000003 time 0.9306 (1.0145) model_time 0.9305 (1.0046) loss 0.6801 (0.8154) grad_norm 7.9108 (8.7335/1.8946) mem 68106MB [2022-12-20 16:22:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][80/1519] eta 0:24:17 lr 0.000003 time 0.9282 (1.0131) model_time 0.9280 (1.0044) loss 0.6880 (0.8119) grad_norm 10.7425 (8.6757/1.8417) mem 68106MB [2022-12-20 16:22:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][90/1519] eta 0:24:05 lr 0.000003 time 0.9208 (1.0118) model_time 0.9206 (1.0040) loss 0.7268 (0.8046) grad_norm 8.3714 (8.6233/1.7495) mem 68106MB [2022-12-20 16:23:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][100/1519] eta 0:23:54 lr 0.000003 time 0.9470 (1.0109) model_time 0.9467 (1.0038) loss 0.8642 (0.8049) grad_norm 6.7009 (8.6244/1.7129) mem 68106MB [2022-12-20 16:23:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][110/1519] eta 0:23:42 lr 0.000003 time 0.9254 (1.0097) model_time 0.9253 (1.0032) loss 0.7447 (0.8049) grad_norm 6.4521 (8.5753/1.7005) mem 68106MB [2022-12-20 16:23:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][120/1519] eta 0:23:32 lr 0.000003 time 0.9349 (1.0093) model_time 0.9348 (1.0033) loss 0.8685 (0.8043) grad_norm 8.2851 (8.5459/1.6500) mem 68106MB [2022-12-20 16:23:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][130/1519] eta 0:23:21 lr 0.000003 time 0.9847 (1.0089) model_time 0.9845 (1.0034) loss 0.9255 (0.8066) grad_norm 8.8394 (8.4811/1.6243) mem 68106MB [2022-12-20 16:23:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][140/1519] eta 0:23:12 lr 0.000003 time 0.9215 (1.0100) model_time 0.9212 (1.0048) loss 0.6689 (0.8047) grad_norm 9.0866 (8.5063/1.6166) mem 68106MB [2022-12-20 16:23:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][150/1519] eta 0:23:03 lr 0.000003 time 0.9257 (1.0102) model_time 0.9255 (1.0054) loss 0.6589 (0.8013) grad_norm 6.3505 (8.4875/1.5900) mem 68106MB [2022-12-20 16:24:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][160/1519] eta 0:22:54 lr 0.000003 time 0.9238 (1.0114) model_time 0.9236 (1.0068) loss 0.6568 (0.8065) grad_norm 6.2443 (8.5759/1.9353) mem 68106MB [2022-12-20 16:24:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][170/1519] eta 0:22:43 lr 0.000003 time 0.9221 (1.0107) model_time 0.9219 (1.0064) loss 0.9096 (0.8078) grad_norm 6.6158 (8.6185/1.9318) mem 68106MB [2022-12-20 16:24:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][180/1519] eta 0:22:32 lr 0.000003 time 0.9337 (1.0102) model_time 0.9335 (1.0061) loss 0.7096 (0.8051) grad_norm 5.4397 (8.6210/1.9617) mem 68106MB [2022-12-20 16:24:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][190/1519] eta 0:22:21 lr 0.000003 time 0.9285 (1.0097) model_time 0.9284 (1.0058) loss 1.0588 (0.8063) grad_norm 8.6835 (8.6627/1.9705) mem 68106MB [2022-12-20 16:24:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][200/1519] eta 0:22:11 lr 0.000003 time 0.9766 (1.0094) model_time 0.9763 (1.0057) loss 0.7962 (0.8073) grad_norm 11.2444 (8.7305/1.9908) mem 68106MB [2022-12-20 16:24:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][210/1519] eta 0:22:02 lr 0.000003 time 1.1803 (1.0105) model_time 1.1801 (1.0070) loss 0.6953 (0.8060) grad_norm 7.4929 (8.7053/1.9586) mem 68106MB [2022-12-20 16:25:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][220/1519] eta 0:21:51 lr 0.000003 time 0.9245 (1.0099) model_time 0.9243 (1.0065) loss 0.9103 (0.8042) grad_norm 6.3651 (8.6108/1.9628) mem 68106MB [2022-12-20 16:25:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][230/1519] eta 0:21:42 lr 0.000003 time 0.9362 (1.0106) model_time 0.9361 (1.0073) loss 0.6838 (0.8036) grad_norm 9.5739 (8.5833/1.9506) mem 68106MB [2022-12-20 16:25:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][240/1519] eta 0:21:32 lr 0.000003 time 0.9353 (1.0102) model_time 0.9350 (1.0070) loss 0.7923 (0.8033) grad_norm 12.6802 (8.6437/1.9962) mem 68106MB [2022-12-20 16:25:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][250/1519] eta 0:21:21 lr 0.000003 time 0.9308 (1.0098) model_time 0.9306 (1.0068) loss 0.8911 (0.8072) grad_norm 9.4559 (8.6100/1.9842) mem 68106MB [2022-12-20 16:25:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][260/1519] eta 0:21:10 lr 0.000003 time 0.9200 (1.0094) model_time 0.9198 (1.0064) loss 0.8142 (0.8062) grad_norm 5.8888 (8.5534/1.9774) mem 68106MB [2022-12-20 16:25:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][270/1519] eta 0:21:00 lr 0.000003 time 0.9319 (1.0092) model_time 0.9317 (1.0063) loss 0.7413 (0.8061) grad_norm 7.7229 (8.5621/1.9491) mem 68106MB [2022-12-20 16:26:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][280/1519] eta 0:20:50 lr 0.000003 time 0.9290 (1.0089) model_time 0.9288 (1.0061) loss 0.7545 (0.8065) grad_norm 8.0407 (8.5759/1.9191) mem 68106MB [2022-12-20 16:26:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][290/1519] eta 0:20:39 lr 0.000003 time 0.9406 (1.0088) model_time 0.9405 (1.0061) loss 0.7792 (0.8071) grad_norm 6.9204 (8.5311/1.9023) mem 68106MB [2022-12-20 16:26:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][300/1519] eta 0:20:29 lr 0.000003 time 0.9736 (1.0086) model_time 0.9734 (1.0060) loss 0.7709 (0.8070) grad_norm 8.2651 (8.5180/1.8778) mem 68106MB [2022-12-20 16:26:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][310/1519] eta 0:20:19 lr 0.000003 time 0.9235 (1.0087) model_time 0.9231 (1.0062) loss 0.9467 (0.8084) grad_norm 7.4862 (8.5012/1.8546) mem 68106MB [2022-12-20 16:26:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][320/1519] eta 0:20:09 lr 0.000003 time 0.9295 (1.0088) model_time 0.9293 (1.0063) loss 0.7013 (0.8087) grad_norm 7.8543 (8.5181/1.9636) mem 68106MB [2022-12-20 16:26:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][330/1519] eta 0:19:59 lr 0.000003 time 0.9207 (1.0091) model_time 0.9206 (1.0067) loss 0.7275 (0.8085) grad_norm 8.1823 (8.5289/1.9816) mem 68106MB [2022-12-20 16:27:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][340/1519] eta 0:19:49 lr 0.000003 time 0.9226 (1.0089) model_time 0.9221 (1.0066) loss 0.8728 (0.8081) grad_norm 7.7089 (8.5220/1.9718) mem 68106MB [2022-12-20 16:27:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][350/1519] eta 0:19:39 lr 0.000003 time 0.9218 (1.0086) model_time 0.9216 (1.0063) loss 0.6846 (0.8071) grad_norm 11.2724 (8.5096/1.9651) mem 68106MB [2022-12-20 16:27:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][360/1519] eta 0:19:28 lr 0.000003 time 0.9333 (1.0084) model_time 0.9332 (1.0062) loss 0.9974 (0.8079) grad_norm 10.0257 (8.5072/1.9523) mem 68106MB [2022-12-20 16:27:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][370/1519] eta 0:19:18 lr 0.000003 time 0.9216 (1.0082) model_time 0.9215 (1.0060) loss 0.7760 (0.8087) grad_norm 5.6989 (8.4971/1.9415) mem 68106MB [2022-12-20 16:27:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][380/1519] eta 0:19:08 lr 0.000003 time 0.9057 (1.0083) model_time 0.9055 (1.0061) loss 0.6603 (0.8080) grad_norm 8.7188 (8.4735/1.9274) mem 68106MB [2022-12-20 16:27:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][390/1519] eta 0:18:58 lr 0.000003 time 0.9249 (1.0085) model_time 0.9247 (1.0064) loss 0.7678 (0.8068) grad_norm 8.3245 (8.4491/1.9230) mem 68106MB [2022-12-20 16:28:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][400/1519] eta 0:18:48 lr 0.000003 time 0.9240 (1.0082) model_time 0.9239 (1.0062) loss 0.7250 (0.8058) grad_norm 9.5111 (8.4948/1.9762) mem 68106MB [2022-12-20 16:28:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][410/1519] eta 0:18:37 lr 0.000003 time 0.9250 (1.0080) model_time 0.9248 (1.0060) loss 0.9108 (0.8054) grad_norm 8.6292 (8.4870/1.9685) mem 68106MB [2022-12-20 16:28:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][420/1519] eta 0:18:27 lr 0.000003 time 0.9230 (1.0078) model_time 0.9228 (1.0058) loss 0.6723 (0.8058) grad_norm 6.3240 (8.4762/1.9626) mem 68106MB [2022-12-20 16:28:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][430/1519] eta 0:18:17 lr 0.000003 time 0.9113 (1.0075) model_time 0.9112 (1.0056) loss 0.8254 (0.8038) grad_norm 8.4944 (8.5061/1.9760) mem 68106MB [2022-12-20 16:28:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][440/1519] eta 0:18:06 lr 0.000003 time 0.9315 (1.0073) model_time 0.9314 (1.0054) loss 0.8467 (0.8037) grad_norm 8.4657 (8.5021/1.9710) mem 68106MB [2022-12-20 16:28:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][450/1519] eta 0:17:56 lr 0.000003 time 0.9311 (1.0074) model_time 0.9309 (1.0055) loss 0.7100 (0.8026) grad_norm 7.0167 (8.4772/1.9574) mem 68106MB [2022-12-20 16:29:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][460/1519] eta 0:17:46 lr 0.000003 time 0.9202 (1.0075) model_time 0.9200 (1.0057) loss 0.8619 (0.8036) grad_norm 9.6231 (8.4789/1.9462) mem 68106MB [2022-12-20 16:29:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][470/1519] eta 0:17:37 lr 0.000003 time 0.9261 (1.0077) model_time 0.9258 (1.0059) loss 0.9448 (0.8027) grad_norm 5.6119 (8.4729/1.9545) mem 68106MB [2022-12-20 16:29:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][480/1519] eta 0:17:26 lr 0.000003 time 1.0120 (1.0077) model_time 1.0118 (1.0059) loss 0.8183 (0.8025) grad_norm 8.3443 (8.4460/1.9474) mem 68106MB [2022-12-20 16:29:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][490/1519] eta 0:17:16 lr 0.000003 time 0.9349 (1.0075) model_time 0.9347 (1.0058) loss 1.3286 (0.8029) grad_norm 7.5041 (8.4405/1.9344) mem 68106MB [2022-12-20 16:29:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][500/1519] eta 0:17:06 lr 0.000003 time 0.9345 (1.0075) model_time 0.9344 (1.0058) loss 0.6768 (0.8023) grad_norm 6.2648 (8.4133/1.9316) mem 68106MB [2022-12-20 16:29:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][510/1519] eta 0:16:56 lr 0.000003 time 0.9195 (1.0074) model_time 0.9194 (1.0057) loss 0.6856 (0.8034) grad_norm 7.7030 (8.4585/1.9557) mem 68106MB [2022-12-20 16:30:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][520/1519] eta 0:16:46 lr 0.000003 time 0.9317 (1.0075) model_time 0.9314 (1.0058) loss 0.6558 (0.8049) grad_norm 10.6999 (8.4858/1.9664) mem 68106MB [2022-12-20 16:30:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][530/1519] eta 0:16:36 lr 0.000003 time 0.9228 (1.0074) model_time 0.9226 (1.0058) loss 0.7151 (0.8054) grad_norm 11.4436 (8.4977/1.9680) mem 68106MB [2022-12-20 16:30:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][540/1519] eta 0:16:26 lr 0.000003 time 0.9227 (1.0074) model_time 0.9225 (1.0058) loss 1.1838 (0.8066) grad_norm 5.9382 (8.4976/1.9659) mem 68106MB [2022-12-20 16:30:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][550/1519] eta 0:16:16 lr 0.000003 time 0.9273 (1.0073) model_time 0.9271 (1.0057) loss 0.8023 (0.8071) grad_norm 10.8886 (8.5092/1.9633) mem 68106MB [2022-12-20 16:30:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][560/1519] eta 0:16:05 lr 0.000003 time 0.9212 (1.0071) model_time 0.9210 (1.0056) loss 0.7125 (0.8077) grad_norm 7.5467 (8.5132/1.9612) mem 68106MB [2022-12-20 16:30:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][570/1519] eta 0:15:55 lr 0.000003 time 0.9303 (1.0070) model_time 0.9302 (1.0055) loss 0.6640 (0.8073) grad_norm 8.5315 (8.5038/1.9519) mem 68106MB [2022-12-20 16:31:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][580/1519] eta 0:15:45 lr 0.000003 time 0.9271 (1.0069) model_time 0.9270 (1.0054) loss 0.7072 (0.8065) grad_norm 7.8476 (8.4865/1.9404) mem 68106MB [2022-12-20 16:31:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][590/1519] eta 0:15:35 lr 0.000003 time 0.9148 (1.0067) model_time 0.9147 (1.0052) loss 0.8408 (0.8069) grad_norm 10.1622 (8.5037/1.9361) mem 68106MB [2022-12-20 16:31:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][600/1519] eta 0:15:25 lr 0.000003 time 0.9306 (1.0067) model_time 0.9304 (1.0052) loss 1.1679 (0.8075) grad_norm 8.8204 (8.4991/1.9261) mem 68106MB [2022-12-20 16:31:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][610/1519] eta 0:15:14 lr 0.000003 time 0.9263 (1.0065) model_time 0.9262 (1.0051) loss 1.0485 (0.8085) grad_norm 8.1799 (8.4758/1.9038) mem 68106MB [2022-12-20 16:31:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][620/1519] eta 0:15:05 lr 0.000003 time 0.9336 (1.0067) model_time 0.9334 (1.0052) loss 0.6888 (0.8092) grad_norm 7.2788 (8.4883/1.9046) mem 68106MB [2022-12-20 16:32:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][630/1519] eta 0:14:55 lr 0.000003 time 0.9274 (1.0068) model_time 0.9272 (1.0054) loss 0.7317 (0.8095) grad_norm 8.6159 (8.4934/1.9073) mem 68106MB [2022-12-20 16:32:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][640/1519] eta 0:14:45 lr 0.000003 time 0.9326 (1.0069) model_time 0.9324 (1.0055) loss 0.8897 (0.8100) grad_norm 12.5902 (8.4963/1.9002) mem 68106MB [2022-12-20 16:32:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][650/1519] eta 0:14:34 lr 0.000003 time 0.9306 (1.0068) model_time 0.9303 (1.0054) loss 0.9728 (0.8101) grad_norm 7.9668 (8.4834/1.8941) mem 68106MB [2022-12-20 16:32:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][660/1519] eta 0:14:24 lr 0.000003 time 0.9294 (1.0066) model_time 0.9293 (1.0052) loss 0.6841 (0.8100) grad_norm 9.5996 (8.4807/1.8976) mem 68106MB [2022-12-20 16:32:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][670/1519] eta 0:14:14 lr 0.000003 time 0.9300 (1.0065) model_time 0.9298 (1.0052) loss 0.6686 (0.8098) grad_norm 9.4469 (8.4854/1.8994) mem 68106MB [2022-12-20 16:32:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][680/1519] eta 0:14:04 lr 0.000003 time 0.9315 (1.0065) model_time 0.9313 (1.0051) loss 0.9503 (0.8089) grad_norm 8.5119 (8.5230/1.9489) mem 68106MB [2022-12-20 16:33:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][690/1519] eta 0:13:54 lr 0.000003 time 0.9166 (1.0064) model_time 0.9164 (1.0051) loss 0.7364 (0.8087) grad_norm 9.0183 (8.5413/1.9601) mem 68106MB [2022-12-20 16:33:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][700/1519] eta 0:13:44 lr 0.000003 time 0.9350 (1.0066) model_time 0.9348 (1.0052) loss 0.6681 (0.8084) grad_norm 13.2962 (8.5673/1.9899) mem 68106MB [2022-12-20 16:33:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][710/1519] eta 0:13:34 lr 0.000003 time 0.9238 (1.0065) model_time 0.9234 (1.0051) loss 0.6620 (0.8075) grad_norm 9.5849 (8.5618/1.9928) mem 68106MB [2022-12-20 16:33:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][720/1519] eta 0:13:24 lr 0.000003 time 0.9339 (1.0067) model_time 0.9338 (1.0054) loss 0.6686 (0.8081) grad_norm 6.4134 (8.5526/2.0024) mem 68106MB [2022-12-20 16:33:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][730/1519] eta 0:13:14 lr 0.000003 time 0.9199 (1.0067) model_time 0.9197 (1.0054) loss 0.9943 (0.8085) grad_norm 6.8597 (8.5886/2.0151) mem 68106MB [2022-12-20 16:33:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][740/1519] eta 0:13:04 lr 0.000003 time 0.9349 (1.0066) model_time 0.9347 (1.0053) loss 0.7270 (0.8077) grad_norm 7.9715 (8.6020/2.0392) mem 68106MB [2022-12-20 16:34:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][750/1519] eta 0:12:53 lr 0.000003 time 0.9212 (1.0065) model_time 0.9210 (1.0052) loss 0.6728 (0.8068) grad_norm 5.9741 (8.5761/2.0524) mem 68106MB [2022-12-20 16:34:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][760/1519] eta 0:12:43 lr 0.000003 time 0.9110 (1.0065) model_time 0.9108 (1.0052) loss 0.9348 (0.8068) grad_norm 8.7364 (8.5603/1.9660) mem 68106MB [2022-12-20 16:34:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][770/1519] eta 0:12:33 lr 0.000003 time 0.9345 (1.0065) model_time 0.9343 (1.0053) loss 0.7804 (0.8065) grad_norm 11.0484 (8.5508/1.9682) mem 68106MB [2022-12-20 16:34:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][780/1519] eta 0:12:24 lr 0.000003 time 0.9347 (1.0068) model_time 0.9345 (1.0056) loss 0.6943 (0.8064) grad_norm 6.9033 (8.5283/1.9577) mem 68106MB [2022-12-20 16:34:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][790/1519] eta 0:12:13 lr 0.000003 time 0.9293 (1.0067) model_time 0.9287 (1.0055) loss 0.7131 (0.8053) grad_norm 8.0607 (8.5134/1.9433) mem 68106MB [2022-12-20 16:34:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][800/1519] eta 0:12:03 lr 0.000003 time 0.9127 (1.0069) model_time 0.9124 (1.0057) loss 0.7302 (0.8052) grad_norm 10.1477 (8.5062/1.9450) mem 68106MB [2022-12-20 16:35:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][810/1519] eta 0:11:53 lr 0.000003 time 0.9277 (1.0069) model_time 0.9275 (1.0057) loss 0.7833 (0.8054) grad_norm 8.6044 (8.4888/1.9425) mem 68106MB [2022-12-20 16:35:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][820/1519] eta 0:11:43 lr 0.000003 time 0.9209 (1.0068) model_time 0.9208 (1.0056) loss 0.7685 (0.8044) grad_norm 5.9811 (8.4854/1.9468) mem 68106MB [2022-12-20 16:35:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][830/1519] eta 0:11:33 lr 0.000003 time 0.9301 (1.0068) model_time 0.9299 (1.0056) loss 0.6513 (0.8046) grad_norm 6.5874 (8.5059/1.9541) mem 68106MB [2022-12-20 16:35:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][840/1519] eta 0:11:23 lr 0.000003 time 0.9271 (1.0070) model_time 0.9269 (1.0058) loss 0.7165 (0.8041) grad_norm 10.4645 (8.4968/1.9367) mem 68106MB [2022-12-20 16:35:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][850/1519] eta 0:11:13 lr 0.000003 time 0.9273 (1.0069) model_time 0.9271 (1.0057) loss 1.2624 (0.8051) grad_norm 9.4619 (8.4992/1.9210) mem 68106MB [2022-12-20 16:35:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][860/1519] eta 0:11:03 lr 0.000003 time 0.9262 (1.0068) model_time 0.9260 (1.0056) loss 0.9061 (0.8049) grad_norm 11.9366 (8.5052/1.9323) mem 68106MB [2022-12-20 16:36:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][870/1519] eta 0:10:53 lr 0.000003 time 0.9272 (1.0067) model_time 0.9271 (1.0056) loss 0.6648 (0.8058) grad_norm 8.2492 (8.5285/1.9740) mem 68106MB [2022-12-20 16:36:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][880/1519] eta 0:10:43 lr 0.000003 time 0.9283 (1.0066) model_time 0.9281 (1.0055) loss 0.6761 (0.8062) grad_norm 11.4192 (8.5216/1.9807) mem 68106MB [2022-12-20 16:36:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][890/1519] eta 0:10:33 lr 0.000003 time 0.9286 (1.0066) model_time 0.9284 (1.0054) loss 1.0020 (0.8065) grad_norm 7.5150 (8.5326/1.9772) mem 68106MB [2022-12-20 16:36:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][900/1519] eta 0:10:23 lr 0.000003 time 0.9307 (1.0065) model_time 0.9304 (1.0053) loss 0.6931 (0.8070) grad_norm 7.4615 (8.5450/1.9788) mem 68106MB [2022-12-20 16:36:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][910/1519] eta 0:10:12 lr 0.000003 time 0.9190 (1.0064) model_time 0.9189 (1.0052) loss 0.6735 (0.8062) grad_norm 9.7481 (8.5323/1.9894) mem 68106MB [2022-12-20 16:36:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][920/1519] eta 0:10:02 lr 0.000003 time 0.9219 (1.0063) model_time 0.9217 (1.0052) loss 0.6915 (0.8059) grad_norm 9.2863 (8.5401/1.9220) mem 68106MB [2022-12-20 16:37:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][930/1519] eta 0:09:52 lr 0.000003 time 0.9207 (1.0062) model_time 0.9206 (1.0051) loss 0.7609 (0.8054) grad_norm 8.1274 (8.5099/1.9086) mem 68106MB [2022-12-20 16:37:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][940/1519] eta 0:09:42 lr 0.000003 time 0.9350 (1.0062) model_time 0.9348 (1.0051) loss 0.9841 (0.8055) grad_norm 7.4755 (8.4991/1.9022) mem 68106MB [2022-12-20 16:37:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][950/1519] eta 0:09:32 lr 0.000003 time 0.9114 (1.0063) model_time 0.9113 (1.0052) loss 0.6987 (0.8056) grad_norm 11.6414 (8.5274/1.9064) mem 68106MB [2022-12-20 16:37:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][960/1519] eta 0:09:22 lr 0.000003 time 0.9207 (1.0062) model_time 0.9206 (1.0051) loss 1.2139 (0.8057) grad_norm 7.4551 (8.5183/1.8954) mem 68106MB [2022-12-20 16:37:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][970/1519] eta 0:09:12 lr 0.000003 time 0.9225 (1.0061) model_time 0.9224 (1.0051) loss 0.7541 (0.8058) grad_norm 11.7760 (8.5089/1.9038) mem 68106MB [2022-12-20 16:37:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][980/1519] eta 0:09:02 lr 0.000003 time 0.9246 (1.0061) model_time 0.9244 (1.0050) loss 0.7302 (0.8067) grad_norm 8.4669 (8.5408/1.8945) mem 68106MB [2022-12-20 16:38:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][990/1519] eta 0:08:52 lr 0.000003 time 0.9275 (1.0060) model_time 0.9273 (1.0049) loss 0.6690 (0.8064) grad_norm 8.6700 (8.5602/1.8836) mem 68106MB [2022-12-20 16:38:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1000/1519] eta 0:08:42 lr 0.000003 time 0.9274 (1.0061) model_time 0.9273 (1.0050) loss 0.8143 (0.8063) grad_norm 8.6856 (8.5452/1.8423) mem 68106MB [2022-12-20 16:38:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1010/1519] eta 0:08:32 lr 0.000003 time 0.9005 (1.0061) model_time 0.9004 (1.0051) loss 0.6558 (0.8053) grad_norm 6.8980 (8.5276/1.8491) mem 68106MB [2022-12-20 16:38:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1020/1519] eta 0:08:22 lr 0.000003 time 0.9283 (1.0060) model_time 0.9281 (1.0050) loss 0.8525 (0.8051) grad_norm 6.9147 (8.5141/1.8486) mem 68106MB [2022-12-20 16:38:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1030/1519] eta 0:08:11 lr 0.000003 time 0.9166 (1.0060) model_time 0.9164 (1.0050) loss 1.0763 (0.8052) grad_norm 9.2718 (8.5292/1.8972) mem 68106MB [2022-12-20 16:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1040/1519] eta 0:08:01 lr 0.000003 time 0.9255 (1.0060) model_time 0.9253 (1.0049) loss 0.8478 (0.8050) grad_norm 12.4613 (8.5208/1.9186) mem 68106MB [2022-12-20 16:39:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1050/1519] eta 0:07:51 lr 0.000003 time 0.9256 (1.0059) model_time 0.9254 (1.0049) loss 0.9197 (0.8048) grad_norm 11.7663 (8.5532/1.9233) mem 68106MB [2022-12-20 16:39:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1060/1519] eta 0:07:41 lr 0.000003 time 0.9264 (1.0059) model_time 0.9263 (1.0049) loss 0.8474 (0.8055) grad_norm 10.5720 (8.5582/1.9215) mem 68106MB [2022-12-20 16:39:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1070/1519] eta 0:07:31 lr 0.000003 time 0.9224 (1.0059) model_time 0.9223 (1.0049) loss 0.7606 (0.8053) grad_norm 12.7729 (8.6157/1.9793) mem 68106MB [2022-12-20 16:39:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1080/1519] eta 0:07:21 lr 0.000003 time 0.9216 (1.0059) model_time 0.9213 (1.0049) loss 0.8036 (0.8056) grad_norm 11.7237 (8.6584/1.9800) mem 68106MB [2022-12-20 16:39:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1090/1519] eta 0:07:11 lr 0.000003 time 0.9980 (1.0060) model_time 0.9978 (1.0050) loss 0.9281 (0.8062) grad_norm 9.9770 (8.6703/1.9786) mem 68106MB [2022-12-20 16:39:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1100/1519] eta 0:07:01 lr 0.000003 time 0.9324 (1.0060) model_time 0.9323 (1.0050) loss 0.6966 (0.8059) grad_norm 10.6989 (8.7181/1.9706) mem 68106MB [2022-12-20 16:40:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1110/1519] eta 0:06:51 lr 0.000003 time 0.9213 (1.0060) model_time 0.9212 (1.0050) loss 0.6743 (0.8053) grad_norm 6.9932 (8.6573/1.9547) mem 68106MB [2022-12-20 16:40:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1120/1519] eta 0:06:41 lr 0.000003 time 0.9289 (1.0061) model_time 0.9287 (1.0051) loss 0.9407 (0.8052) grad_norm 7.2741 (8.6164/1.9393) mem 68106MB [2022-12-20 16:40:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1130/1519] eta 0:06:31 lr 0.000003 time 0.9287 (1.0060) model_time 0.9286 (1.0050) loss 0.6938 (0.8049) grad_norm 8.6941 (8.6185/1.9464) mem 68106MB [2022-12-20 16:40:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1140/1519] eta 0:06:21 lr 0.000003 time 0.9258 (1.0060) model_time 0.9257 (1.0050) loss 0.7001 (0.8044) grad_norm 7.7614 (8.5943/1.9461) mem 68106MB [2022-12-20 16:40:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1150/1519] eta 0:06:11 lr 0.000003 time 0.9271 (1.0060) model_time 0.9270 (1.0050) loss 0.9740 (0.8041) grad_norm 8.7896 (8.5752/1.9455) mem 68106MB [2022-12-20 16:40:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1160/1519] eta 0:06:01 lr 0.000003 time 0.9240 (1.0060) model_time 0.9239 (1.0050) loss 0.6752 (0.8038) grad_norm 9.7343 (8.5787/1.9327) mem 68106MB [2022-12-20 16:41:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1170/1519] eta 0:05:51 lr 0.000003 time 0.9409 (1.0060) model_time 0.9408 (1.0050) loss 0.7999 (0.8041) grad_norm 9.2918 (8.5941/1.9292) mem 68106MB [2022-12-20 16:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1180/1519] eta 0:05:41 lr 0.000003 time 0.9323 (1.0060) model_time 0.9322 (1.0050) loss 0.8210 (0.8040) grad_norm 7.3697 (8.6066/1.9249) mem 68106MB [2022-12-20 16:41:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1190/1519] eta 0:05:30 lr 0.000003 time 0.9960 (1.0060) model_time 0.9959 (1.0051) loss 0.9990 (0.8040) grad_norm 9.1331 (8.6157/1.9414) mem 68106MB [2022-12-20 16:41:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1200/1519] eta 0:05:20 lr 0.000003 time 0.9249 (1.0059) model_time 0.9248 (1.0050) loss 0.6781 (0.8041) grad_norm 6.9112 (8.5981/1.9456) mem 68106MB [2022-12-20 16:41:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1210/1519] eta 0:05:10 lr 0.000003 time 0.9615 (1.0059) model_time 0.9613 (1.0050) loss 0.8071 (0.8041) grad_norm 9.0728 (8.5954/1.9406) mem 68106MB [2022-12-20 16:41:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1220/1519] eta 0:05:00 lr 0.000003 time 0.9212 (1.0059) model_time 0.9211 (1.0049) loss 0.7322 (0.8044) grad_norm 8.0022 (8.5666/1.9395) mem 68106MB [2022-12-20 16:42:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1230/1519] eta 0:04:50 lr 0.000003 time 0.9307 (1.0058) model_time 0.9305 (1.0049) loss 1.0734 (0.8050) grad_norm 8.8027 (8.5680/1.9489) mem 68106MB [2022-12-20 16:42:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1240/1519] eta 0:04:40 lr 0.000003 time 0.9302 (1.0058) model_time 0.9301 (1.0049) loss 0.8237 (0.8050) grad_norm 9.2370 (8.5869/1.9590) mem 68106MB [2022-12-20 16:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1250/1519] eta 0:04:30 lr 0.000003 time 1.0141 (1.0059) model_time 1.0140 (1.0050) loss 0.8770 (0.8048) grad_norm 7.2018 (8.6006/1.9694) mem 68106MB [2022-12-20 16:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1260/1519] eta 0:04:20 lr 0.000003 time 1.0038 (1.0061) model_time 1.0036 (1.0052) loss 0.8483 (0.8053) grad_norm 10.9568 (8.6089/1.9692) mem 68106MB [2022-12-20 16:42:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1270/1519] eta 0:04:10 lr 0.000003 time 1.1279 (1.0062) model_time 1.1277 (1.0053) loss 0.6902 (0.8057) grad_norm 7.2188 (8.6040/1.9704) mem 68106MB [2022-12-20 16:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1280/1519] eta 0:04:00 lr 0.000003 time 0.9309 (1.0062) model_time 0.9307 (1.0053) loss 0.6615 (0.8056) grad_norm 8.8451 (8.5720/1.9278) mem 68106MB [2022-12-20 16:43:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1290/1519] eta 0:03:50 lr 0.000003 time 0.9220 (1.0062) model_time 0.9218 (1.0053) loss 0.6871 (0.8048) grad_norm 7.6698 (8.5686/1.9271) mem 68106MB [2022-12-20 16:43:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1300/1519] eta 0:03:40 lr 0.000003 time 0.9251 (1.0061) model_time 0.9250 (1.0053) loss 0.6889 (0.8048) grad_norm 11.9964 (8.5555/1.9192) mem 68106MB [2022-12-20 16:43:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1310/1519] eta 0:03:30 lr 0.000003 time 0.9266 (1.0061) model_time 0.9265 (1.0053) loss 0.7143 (0.8050) grad_norm 7.9269 (8.5667/1.9072) mem 68106MB [2022-12-20 16:43:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1320/1519] eta 0:03:20 lr 0.000003 time 0.9251 (1.0061) model_time 0.9250 (1.0053) loss 0.7587 (0.8052) grad_norm 8.8389 (8.5942/1.8928) mem 68106MB [2022-12-20 16:43:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1330/1519] eta 0:03:10 lr 0.000003 time 0.9332 (1.0062) model_time 0.9330 (1.0053) loss 1.0123 (0.8048) grad_norm 5.8763 (8.5538/1.8825) mem 68106MB [2022-12-20 16:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1340/1519] eta 0:03:00 lr 0.000003 time 0.9274 (1.0062) model_time 0.9273 (1.0053) loss 0.7094 (0.8048) grad_norm 10.1654 (8.5264/1.8545) mem 68106MB [2022-12-20 16:44:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1350/1519] eta 0:02:50 lr 0.000003 time 0.9231 (1.0061) model_time 0.9230 (1.0053) loss 1.1801 (0.8054) grad_norm 8.0017 (8.5667/1.8603) mem 68106MB [2022-12-20 16:44:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1360/1519] eta 0:02:39 lr 0.000003 time 0.9427 (1.0061) model_time 0.9426 (1.0053) loss 0.7122 (0.8051) grad_norm 9.2767 (8.5628/1.8726) mem 68106MB [2022-12-20 16:44:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1370/1519] eta 0:02:29 lr 0.000003 time 0.9296 (1.0061) model_time 0.9295 (1.0052) loss 0.6957 (0.8051) grad_norm 7.9836 (8.5354/1.8650) mem 68106MB [2022-12-20 16:44:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1380/1519] eta 0:02:19 lr 0.000003 time 0.9199 (1.0060) model_time 0.9198 (1.0052) loss 0.9603 (0.8053) grad_norm 7.3086 (8.5555/1.8592) mem 68106MB [2022-12-20 16:44:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1390/1519] eta 0:02:09 lr 0.000003 time 0.9318 (1.0060) model_time 0.9317 (1.0052) loss 0.7749 (0.8048) grad_norm 5.7062 (8.5854/1.9334) mem 68106MB [2022-12-20 16:44:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1400/1519] eta 0:01:59 lr 0.000003 time 0.9336 (1.0060) model_time 0.9335 (1.0051) loss 0.7971 (0.8049) grad_norm 12.7769 (8.5898/1.9385) mem 68106MB [2022-12-20 16:45:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1410/1519] eta 0:01:49 lr 0.000003 time 0.9252 (1.0060) model_time 0.9251 (1.0051) loss 0.7200 (0.8048) grad_norm 6.1129 (8.5867/1.9362) mem 68106MB [2022-12-20 16:45:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1420/1519] eta 0:01:39 lr 0.000003 time 0.9744 (1.0060) model_time 0.9743 (1.0051) loss 0.7014 (0.8046) grad_norm 9.3045 (8.6391/1.9252) mem 68106MB [2022-12-20 16:45:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1430/1519] eta 0:01:29 lr 0.000003 time 0.9881 (1.0060) model_time 0.9880 (1.0052) loss 0.8979 (0.8045) grad_norm 8.6000 (8.6410/1.9061) mem 68106MB [2022-12-20 16:45:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1440/1519] eta 0:01:19 lr 0.000003 time 1.0161 (1.0061) model_time 1.0160 (1.0053) loss 0.7735 (0.8049) grad_norm 13.7851 (8.6554/1.9257) mem 68106MB [2022-12-20 16:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1450/1519] eta 0:01:09 lr 0.000003 time 0.9310 (1.0061) model_time 0.9309 (1.0053) loss 0.7011 (0.8049) grad_norm 5.1507 (8.6318/1.9412) mem 68106MB [2022-12-20 16:45:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1460/1519] eta 0:00:59 lr 0.000003 time 0.9263 (1.0061) model_time 0.9262 (1.0052) loss 1.6766 (0.8053) grad_norm 6.6264 (8.6212/1.9354) mem 68106MB [2022-12-20 16:46:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1470/1519] eta 0:00:49 lr 0.000003 time 0.9549 (1.0060) model_time 0.9548 (1.0052) loss 0.7340 (0.8056) grad_norm 7.6306 (8.5856/1.8890) mem 68106MB [2022-12-20 16:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1480/1519] eta 0:00:39 lr 0.000003 time 0.9204 (1.0060) model_time 0.9203 (1.0052) loss 0.7877 (0.8056) grad_norm 6.5591 (8.5802/1.8908) mem 68106MB [2022-12-20 16:46:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1490/1519] eta 0:00:29 lr 0.000003 time 0.9013 (1.0060) model_time 0.9011 (1.0052) loss 0.7632 (0.8055) grad_norm 9.0795 (8.5812/1.9043) mem 68106MB [2022-12-20 16:46:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1500/1519] eta 0:00:19 lr 0.000003 time 0.9293 (1.0060) model_time 0.9291 (1.0052) loss 0.9553 (0.8051) grad_norm 5.9660 (8.5717/1.9107) mem 68106MB [2022-12-20 16:46:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [82/100][1510/1519] eta 0:00:09 lr 0.000003 time 0.9212 (1.0060) model_time 0.9211 (1.0052) loss 0.6712 (0.8050) grad_norm 6.6411 (8.5804/1.9076) mem 68106MB [2022-12-20 16:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 82 training takes 0:25:28 [2022-12-20 16:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_82.pth saving...... [2022-12-20 16:47:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_82.pth saved !!! [2022-12-20 16:47:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.644 (0.644) Loss 0.5399 (0.5399) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 16:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.329) Loss 0.5338 (0.5092) Acc@1 92.014 (92.771) Acc@5 97.917 (98.453) Mem 68106MB [2022-12-20 16:47:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4905 (0.5034) Acc@1 90.972 (92.824) Acc@5 98.958 (98.413) Mem 68106MB [2022-12-20 16:47:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.307 (0.310) Loss 0.6404 (0.5104) Acc@1 90.625 (92.552) Acc@5 97.917 (98.398) Mem 68106MB [2022-12-20 16:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4626 (0.5012) Acc@1 93.750 (92.607) Acc@5 98.958 (98.535) Mem 68106MB [2022-12-20 16:47:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.306) Loss 0.4899 (0.4986) Acc@1 92.014 (92.674) Acc@5 99.653 (98.597) Mem 68106MB [2022-12-20 16:47:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.301 (0.305) Loss 0.5144 (0.4984) Acc@1 90.972 (92.612) Acc@5 98.264 (98.560) Mem 68106MB [2022-12-20 16:47:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.5453 (0.4996) Acc@1 93.750 (92.552) Acc@5 97.917 (98.552) Mem 68106MB [2022-12-20 16:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4356 (0.4981) Acc@1 92.708 (92.610) Acc@5 98.264 (98.577) Mem 68106MB [2022-12-20 16:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:82] * Acc@1 92.559 Acc@5 98.580 [2022-12-20 16:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 16:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 16:48:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 16:48:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.56% [2022-12-20 16:48:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][0/1519] eta 0:33:25 lr 0.000003 time 1.3206 (1.3206) model_time 0.9173 (0.9173) loss 0.7854 (0.7854) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 16:48:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][10/1519] eta 0:25:50 lr 0.000003 time 0.9246 (1.0276) model_time 0.9245 (0.9907) loss 0.7489 (0.8483) grad_norm 7.1959 (8.5442/2.8137) mem 68106MB [2022-12-20 16:48:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][20/1519] eta 0:25:21 lr 0.000003 time 0.9443 (1.0151) model_time 0.9441 (0.9956) loss 0.9250 (0.8124) grad_norm 9.8412 (9.0275/2.1600) mem 68106MB [2022-12-20 16:48:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][30/1519] eta 0:25:04 lr 0.000003 time 0.9216 (1.0105) model_time 0.9215 (0.9972) loss 0.8974 (0.7870) grad_norm 9.1791 (9.0841/2.5463) mem 68106MB [2022-12-20 16:48:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][40/1519] eta 0:24:50 lr 0.000003 time 0.9233 (1.0076) model_time 0.9232 (0.9975) loss 0.8790 (0.7945) grad_norm 16.3650 (9.2652/2.7809) mem 68106MB [2022-12-20 16:49:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][50/1519] eta 0:24:43 lr 0.000003 time 1.0313 (1.0097) model_time 1.0312 (1.0015) loss 0.6614 (0.7849) grad_norm 6.6787 (9.0392/2.6021) mem 68106MB [2022-12-20 16:49:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][60/1519] eta 0:24:32 lr 0.000003 time 0.9925 (1.0095) model_time 0.9924 (1.0026) loss 0.7442 (0.7870) grad_norm 6.6711 (8.7822/2.4784) mem 68106MB [2022-12-20 16:49:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][70/1519] eta 0:24:22 lr 0.000003 time 0.9294 (1.0091) model_time 0.9293 (1.0031) loss 0.7320 (0.7882) grad_norm 9.4635 (8.8456/2.3426) mem 68106MB [2022-12-20 16:49:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][80/1519] eta 0:24:10 lr 0.000003 time 0.9213 (1.0080) model_time 0.9211 (1.0027) loss 0.6684 (0.7827) grad_norm 7.2639 (8.8119/2.2140) mem 68106MB [2022-12-20 16:49:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][90/1519] eta 0:23:59 lr 0.000003 time 0.9345 (1.0071) model_time 0.9343 (1.0024) loss 0.7757 (0.7833) grad_norm 7.8707 (8.9702/2.4389) mem 68106MB [2022-12-20 16:49:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][100/1519] eta 0:23:48 lr 0.000003 time 0.9200 (1.0066) model_time 0.9199 (1.0023) loss 0.8388 (0.7862) grad_norm 9.1742 (8.8172/2.3849) mem 68106MB [2022-12-20 16:50:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][110/1519] eta 0:23:37 lr 0.000003 time 0.9232 (1.0060) model_time 0.9231 (1.0020) loss 0.6869 (0.7868) grad_norm 10.6687 (8.9920/2.4820) mem 68106MB [2022-12-20 16:50:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][120/1519] eta 0:23:27 lr 0.000003 time 0.9369 (1.0059) model_time 0.9367 (1.0023) loss 0.7943 (0.7901) grad_norm 8.5766 (8.9592/2.4020) mem 68106MB [2022-12-20 16:50:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][130/1519] eta 0:23:17 lr 0.000003 time 0.9251 (1.0063) model_time 0.9250 (1.0029) loss 0.6875 (0.7864) grad_norm 6.7392 (8.8667/2.3357) mem 68106MB [2022-12-20 16:50:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][140/1519] eta 0:23:07 lr 0.000003 time 0.9570 (1.0064) model_time 0.9568 (1.0032) loss 0.6836 (0.7876) grad_norm 7.2815 (8.7914/2.2767) mem 68106MB [2022-12-20 16:50:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][150/1519] eta 0:22:56 lr 0.000003 time 0.9300 (1.0058) model_time 0.9298 (1.0029) loss 0.7135 (0.7856) grad_norm 10.0166 (8.8541/2.2191) mem 68106MB [2022-12-20 16:50:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][160/1519] eta 0:22:46 lr 0.000003 time 0.9298 (1.0056) model_time 0.9297 (1.0028) loss 0.8642 (0.7848) grad_norm 6.2511 (8.8008/2.2211) mem 68106MB [2022-12-20 16:51:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][170/1519] eta 0:22:36 lr 0.000003 time 0.9241 (1.0058) model_time 0.9240 (1.0031) loss 0.8677 (0.7909) grad_norm 9.0733 (8.8490/2.2093) mem 68106MB [2022-12-20 16:51:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][180/1519] eta 0:22:26 lr 0.000003 time 0.9251 (1.0054) model_time 0.9249 (1.0029) loss 1.1537 (0.7910) grad_norm 8.6120 (8.7850/2.1932) mem 68106MB [2022-12-20 16:51:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][190/1519] eta 0:22:15 lr 0.000003 time 0.9267 (1.0051) model_time 0.9266 (1.0027) loss 0.7415 (0.7910) grad_norm 7.3731 (8.7203/2.1603) mem 68106MB [2022-12-20 16:51:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][200/1519] eta 0:22:06 lr 0.000003 time 0.9237 (1.0057) model_time 0.9236 (1.0034) loss 0.6951 (0.7920) grad_norm 14.5644 (8.7278/2.2005) mem 68106MB [2022-12-20 16:51:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][210/1519] eta 0:21:56 lr 0.000003 time 0.9220 (1.0056) model_time 0.9219 (1.0034) loss 0.7413 (0.7892) grad_norm 8.8201 (8.6952/2.1620) mem 68106MB [2022-12-20 16:51:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][220/1519] eta 0:21:46 lr 0.000003 time 0.9331 (1.0061) model_time 0.9330 (1.0039) loss 0.9988 (0.7905) grad_norm 7.2059 (8.7045/2.1453) mem 68106MB [2022-12-20 16:52:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][230/1519] eta 0:21:38 lr 0.000003 time 1.0384 (1.0070) model_time 1.0382 (1.0050) loss 0.9704 (0.7957) grad_norm 10.4490 (8.7104/2.1148) mem 68106MB [2022-12-20 16:52:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][240/1519] eta 0:21:28 lr 0.000003 time 1.0150 (1.0077) model_time 1.0148 (1.0057) loss 0.7925 (0.7940) grad_norm 7.8089 (8.7323/2.0807) mem 68106MB [2022-12-20 16:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][250/1519] eta 0:21:18 lr 0.000003 time 0.9268 (1.0076) model_time 0.9267 (1.0056) loss 0.8209 (0.7948) grad_norm 7.4024 (8.7172/2.0519) mem 68106MB [2022-12-20 16:52:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][260/1519] eta 0:21:08 lr 0.000003 time 0.9302 (1.0072) model_time 0.9300 (1.0053) loss 0.7374 (0.7979) grad_norm 7.3673 (8.7056/2.0277) mem 68106MB [2022-12-20 16:52:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][270/1519] eta 0:20:57 lr 0.000003 time 0.9234 (1.0069) model_time 0.9233 (1.0051) loss 0.6631 (0.7981) grad_norm 6.4277 (8.7096/2.0404) mem 68106MB [2022-12-20 16:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][280/1519] eta 0:20:47 lr 0.000003 time 0.9253 (1.0066) model_time 0.9252 (1.0049) loss 0.9689 (0.7969) grad_norm 7.3913 (8.7261/2.0239) mem 68106MB [2022-12-20 16:53:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][290/1519] eta 0:20:36 lr 0.000003 time 0.9302 (1.0063) model_time 0.9300 (1.0047) loss 0.6759 (0.7972) grad_norm 10.4255 (8.7224/2.0247) mem 68106MB [2022-12-20 16:53:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][300/1519] eta 0:20:27 lr 0.000003 time 0.9273 (1.0066) model_time 0.9272 (1.0049) loss 0.8662 (0.7975) grad_norm 11.3396 (8.7155/2.0168) mem 68106MB [2022-12-20 16:53:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][310/1519] eta 0:20:16 lr 0.000003 time 0.9238 (1.0065) model_time 0.9237 (1.0049) loss 0.8837 (0.8000) grad_norm 8.0703 (8.6894/1.9946) mem 68106MB [2022-12-20 16:53:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][320/1519] eta 0:20:06 lr 0.000003 time 0.9241 (1.0065) model_time 0.9239 (1.0049) loss 0.6649 (0.7986) grad_norm 7.9052 (8.6623/1.9831) mem 68106MB [2022-12-20 16:53:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][330/1519] eta 0:19:56 lr 0.000003 time 0.9207 (1.0063) model_time 0.9206 (1.0047) loss 0.6920 (0.7975) grad_norm 6.9119 (8.6123/1.9748) mem 68106MB [2022-12-20 16:53:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][340/1519] eta 0:19:46 lr 0.000003 time 0.9244 (1.0061) model_time 0.9243 (1.0046) loss 0.6904 (0.7981) grad_norm 11.1741 (8.6186/1.9599) mem 68106MB [2022-12-20 16:54:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][350/1519] eta 0:19:35 lr 0.000003 time 0.9244 (1.0059) model_time 0.9243 (1.0044) loss 0.8027 (0.7982) grad_norm 8.1131 (8.6140/1.9433) mem 68106MB [2022-12-20 16:54:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][360/1519] eta 0:19:25 lr 0.000003 time 0.9223 (1.0057) model_time 0.9221 (1.0043) loss 0.7471 (0.7999) grad_norm 8.2938 (8.5995/1.9205) mem 68106MB [2022-12-20 16:54:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][370/1519] eta 0:19:15 lr 0.000003 time 0.9307 (1.0056) model_time 0.9306 (1.0042) loss 0.7512 (0.7994) grad_norm 7.3770 (8.5952/1.9016) mem 68106MB [2022-12-20 16:54:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][380/1519] eta 0:19:05 lr 0.000003 time 0.9263 (1.0054) model_time 0.9262 (1.0041) loss 0.8889 (0.7990) grad_norm 9.8496 (8.5956/1.8827) mem 68106MB [2022-12-20 16:54:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][390/1519] eta 0:18:55 lr 0.000003 time 0.9347 (1.0055) model_time 0.9345 (1.0042) loss 0.7685 (0.8000) grad_norm 7.0758 (8.5809/1.8676) mem 68106MB [2022-12-20 16:54:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][400/1519] eta 0:18:45 lr 0.000003 time 0.9301 (1.0054) model_time 0.9300 (1.0041) loss 0.6898 (0.7996) grad_norm 8.2994 (8.5876/1.8669) mem 68106MB [2022-12-20 16:55:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][410/1519] eta 0:18:35 lr 0.000003 time 0.9599 (1.0056) model_time 0.9598 (1.0043) loss 1.0160 (0.8010) grad_norm 12.1804 (8.6023/1.8898) mem 68106MB [2022-12-20 16:55:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][420/1519] eta 0:18:25 lr 0.000003 time 0.9197 (1.0056) model_time 0.9195 (1.0043) loss 0.6652 (0.8014) grad_norm 9.9699 (8.6459/1.9266) mem 68106MB [2022-12-20 16:55:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][430/1519] eta 0:18:14 lr 0.000003 time 0.9241 (1.0054) model_time 0.9239 (1.0041) loss 0.7272 (0.8008) grad_norm 8.3506 (8.6295/1.9170) mem 68106MB [2022-12-20 16:55:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][440/1519] eta 0:18:04 lr 0.000003 time 0.9265 (1.0053) model_time 0.9264 (1.0040) loss 0.7551 (0.7993) grad_norm 13.4513 (8.6396/1.9276) mem 68106MB [2022-12-20 16:55:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][450/1519] eta 0:17:54 lr 0.000003 time 0.9201 (1.0052) model_time 0.9199 (1.0040) loss 1.0484 (0.8010) grad_norm 11.6657 (8.6697/1.9785) mem 68106MB [2022-12-20 16:55:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][460/1519] eta 0:17:44 lr 0.000003 time 0.9241 (1.0051) model_time 0.9239 (1.0039) loss 0.7348 (0.8009) grad_norm 8.9970 (8.6705/1.9599) mem 68106MB [2022-12-20 16:56:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][470/1519] eta 0:17:34 lr 0.000003 time 0.9217 (1.0050) model_time 0.9215 (1.0038) loss 0.7963 (0.8001) grad_norm 8.7368 (8.6479/1.9570) mem 68106MB [2022-12-20 16:56:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][480/1519] eta 0:17:24 lr 0.000003 time 0.9310 (1.0053) model_time 0.9308 (1.0041) loss 0.7969 (0.7997) grad_norm 7.0116 (8.6665/1.9589) mem 68106MB [2022-12-20 16:56:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][490/1519] eta 0:17:14 lr 0.000003 time 0.9268 (1.0053) model_time 0.9266 (1.0041) loss 0.9071 (0.8001) grad_norm 7.0435 (8.6420/1.9512) mem 68106MB [2022-12-20 16:56:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][500/1519] eta 0:17:04 lr 0.000003 time 0.9225 (1.0051) model_time 0.9224 (1.0040) loss 0.6650 (0.7997) grad_norm 8.5146 (8.6378/1.9366) mem 68106MB [2022-12-20 16:56:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][510/1519] eta 0:16:54 lr 0.000003 time 0.8876 (1.0056) model_time 0.8874 (1.0045) loss 0.8049 (0.8007) grad_norm 8.1918 (8.6109/1.9299) mem 68106MB [2022-12-20 16:56:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][520/1519] eta 0:16:44 lr 0.000003 time 0.9270 (1.0056) model_time 0.9269 (1.0045) loss 0.7046 (0.8013) grad_norm 9.2209 (8.6182/1.9156) mem 68106MB [2022-12-20 16:57:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][530/1519] eta 0:16:34 lr 0.000003 time 1.0126 (1.0058) model_time 1.0125 (1.0048) loss 0.6723 (0.8016) grad_norm 6.2200 (8.6131/1.9147) mem 68106MB [2022-12-20 16:57:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][540/1519] eta 0:16:24 lr 0.000003 time 0.9230 (1.0058) model_time 0.9229 (1.0048) loss 0.7145 (0.8034) grad_norm 6.1935 (8.5851/1.9146) mem 68106MB [2022-12-20 16:57:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][550/1519] eta 0:16:14 lr 0.000003 time 0.9252 (1.0060) model_time 0.9250 (1.0049) loss 0.8560 (0.8023) grad_norm 8.6180 (8.5994/1.9061) mem 68106MB [2022-12-20 16:57:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][560/1519] eta 0:16:04 lr 0.000003 time 0.9273 (1.0061) model_time 0.9271 (1.0051) loss 0.6846 (0.8017) grad_norm 5.9684 (8.5855/1.8982) mem 68106MB [2022-12-20 16:57:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][570/1519] eta 0:15:54 lr 0.000003 time 0.9267 (1.0061) model_time 0.9266 (1.0050) loss 0.6731 (0.8029) grad_norm 7.3397 (8.5678/1.8895) mem 68106MB [2022-12-20 16:57:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][580/1519] eta 0:15:44 lr 0.000003 time 0.9307 (1.0059) model_time 0.9306 (1.0049) loss 0.7507 (0.8042) grad_norm 8.2717 (8.5546/1.8777) mem 68106MB [2022-12-20 16:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][590/1519] eta 0:15:34 lr 0.000003 time 0.9208 (1.0058) model_time 0.9206 (1.0048) loss 0.6855 (0.8046) grad_norm 11.9734 (8.5751/1.8759) mem 68106MB [2022-12-20 16:58:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][600/1519] eta 0:15:24 lr 0.000003 time 0.9235 (1.0057) model_time 0.9233 (1.0047) loss 0.6680 (0.8040) grad_norm 7.4936 (8.5798/1.8825) mem 68106MB [2022-12-20 16:58:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][610/1519] eta 0:15:14 lr 0.000003 time 0.9251 (1.0057) model_time 0.9250 (1.0047) loss 0.6757 (0.8037) grad_norm 6.2287 (8.5624/1.8567) mem 68106MB [2022-12-20 16:58:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][620/1519] eta 0:15:04 lr 0.000003 time 0.9280 (1.0057) model_time 0.9277 (1.0047) loss 0.9815 (0.8042) grad_norm 9.1525 (8.5677/1.8848) mem 68106MB [2022-12-20 16:58:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][630/1519] eta 0:14:53 lr 0.000003 time 0.9358 (1.0056) model_time 0.9356 (1.0046) loss 0.7382 (0.8051) grad_norm 6.0427 (8.5475/1.8468) mem 68106MB [2022-12-20 16:58:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][640/1519] eta 0:14:43 lr 0.000003 time 0.9200 (1.0056) model_time 0.9198 (1.0046) loss 0.7767 (0.8055) grad_norm 7.5076 (8.5364/1.7992) mem 68106MB [2022-12-20 16:59:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][650/1519] eta 0:14:33 lr 0.000003 time 0.9251 (1.0055) model_time 0.9250 (1.0046) loss 0.7122 (0.8057) grad_norm 7.5558 (8.5236/1.7973) mem 68106MB [2022-12-20 16:59:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][660/1519] eta 0:14:23 lr 0.000003 time 0.9295 (1.0054) model_time 0.9293 (1.0045) loss 0.7437 (0.8061) grad_norm 8.9592 (8.5710/1.8868) mem 68106MB [2022-12-20 16:59:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][670/1519] eta 0:14:13 lr 0.000003 time 0.9196 (1.0053) model_time 0.9195 (1.0044) loss 0.8305 (0.8058) grad_norm 8.2854 (8.5430/1.8890) mem 68106MB [2022-12-20 16:59:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][680/1519] eta 0:14:03 lr 0.000003 time 0.9255 (1.0052) model_time 0.9254 (1.0043) loss 0.8141 (0.8054) grad_norm 6.9677 (8.5303/1.8989) mem 68106MB [2022-12-20 16:59:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][690/1519] eta 0:13:53 lr 0.000003 time 0.9244 (1.0052) model_time 0.9243 (1.0043) loss 0.8976 (0.8049) grad_norm 7.0576 (8.4959/1.8352) mem 68106MB [2022-12-20 16:59:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][700/1519] eta 0:13:43 lr 0.000003 time 0.9256 (1.0052) model_time 0.9254 (1.0043) loss 0.7041 (0.8046) grad_norm 9.4307 (8.5181/1.8263) mem 68106MB [2022-12-20 17:00:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][710/1519] eta 0:13:33 lr 0.000003 time 0.9217 (1.0051) model_time 0.9216 (1.0042) loss 0.8299 (0.8042) grad_norm 7.2234 (8.5193/1.8767) mem 68106MB [2022-12-20 17:00:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][720/1519] eta 0:13:23 lr 0.000003 time 0.9256 (1.0052) model_time 0.9255 (1.0043) loss 0.9994 (0.8034) grad_norm 10.0942 (8.5141/1.8877) mem 68106MB [2022-12-20 17:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][730/1519] eta 0:13:13 lr 0.000003 time 0.9374 (1.0055) model_time 0.9372 (1.0046) loss 0.6761 (0.8031) grad_norm 9.3603 (8.5374/1.8956) mem 68106MB [2022-12-20 17:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][740/1519] eta 0:13:03 lr 0.000003 time 0.9184 (1.0054) model_time 0.9183 (1.0045) loss 0.7141 (0.8032) grad_norm 7.0914 (8.5531/1.9031) mem 68106MB [2022-12-20 17:00:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][750/1519] eta 0:12:53 lr 0.000003 time 0.9247 (1.0054) model_time 0.9246 (1.0045) loss 0.6951 (0.8043) grad_norm 7.8976 (8.5262/1.9041) mem 68106MB [2022-12-20 17:00:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][760/1519] eta 0:12:43 lr 0.000003 time 0.9605 (1.0053) model_time 0.9604 (1.0045) loss 0.6539 (0.8040) grad_norm 7.8859 (8.5232/1.8917) mem 68106MB [2022-12-20 17:01:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][770/1519] eta 0:12:32 lr 0.000003 time 0.9234 (1.0053) model_time 0.9233 (1.0045) loss 0.7924 (0.8037) grad_norm 9.5355 (8.5095/1.8765) mem 68106MB [2022-12-20 17:01:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][780/1519] eta 0:12:22 lr 0.000003 time 0.9194 (1.0053) model_time 0.9193 (1.0044) loss 0.8286 (0.8038) grad_norm 7.9096 (8.5164/1.8769) mem 68106MB [2022-12-20 17:01:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][790/1519] eta 0:12:12 lr 0.000003 time 0.9458 (1.0054) model_time 0.9457 (1.0045) loss 0.7050 (0.8039) grad_norm 8.1021 (8.5515/1.8792) mem 68106MB [2022-12-20 17:01:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][800/1519] eta 0:12:02 lr 0.000003 time 0.9214 (1.0053) model_time 0.9213 (1.0045) loss 0.7615 (0.8031) grad_norm 8.3848 (8.5390/1.8604) mem 68106MB [2022-12-20 17:01:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][810/1519] eta 0:11:52 lr 0.000003 time 0.9215 (1.0052) model_time 0.9213 (1.0044) loss 0.6838 (0.8040) grad_norm 9.0706 (8.5736/1.8926) mem 68106MB [2022-12-20 17:01:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][820/1519] eta 0:11:42 lr 0.000003 time 0.9708 (1.0052) model_time 0.9707 (1.0044) loss 0.6938 (0.8040) grad_norm 6.2609 (8.5457/1.8938) mem 68106MB [2022-12-20 17:02:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][830/1519] eta 0:11:32 lr 0.000003 time 0.9237 (1.0051) model_time 0.9236 (1.0043) loss 0.8454 (0.8033) grad_norm 7.6368 (8.5406/1.8902) mem 68106MB [2022-12-20 17:02:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][840/1519] eta 0:11:22 lr 0.000003 time 0.9268 (1.0054) model_time 0.9267 (1.0046) loss 0.8928 (0.8032) grad_norm 9.6168 (8.5300/1.8870) mem 68106MB [2022-12-20 17:02:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][850/1519] eta 0:11:12 lr 0.000003 time 0.9203 (1.0055) model_time 0.9202 (1.0047) loss 0.8805 (0.8029) grad_norm 6.0154 (8.5426/1.9100) mem 68106MB [2022-12-20 17:02:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][860/1519] eta 0:11:02 lr 0.000003 time 0.9200 (1.0055) model_time 0.9199 (1.0047) loss 0.6968 (0.8029) grad_norm 9.5764 (8.5543/1.9093) mem 68106MB [2022-12-20 17:02:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][870/1519] eta 0:10:52 lr 0.000003 time 0.9808 (1.0055) model_time 0.9806 (1.0048) loss 0.8742 (0.8029) grad_norm 5.1752 (8.5331/1.8968) mem 68106MB [2022-12-20 17:02:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][880/1519] eta 0:10:42 lr 0.000003 time 0.9245 (1.0055) model_time 0.9244 (1.0047) loss 0.6732 (0.8021) grad_norm 7.2663 (8.5384/1.9085) mem 68106MB [2022-12-20 17:03:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][890/1519] eta 0:10:32 lr 0.000003 time 0.9234 (1.0054) model_time 0.9233 (1.0047) loss 0.8152 (0.8027) grad_norm 7.6491 (8.5274/1.8931) mem 68106MB [2022-12-20 17:03:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][900/1519] eta 0:10:22 lr 0.000003 time 0.9240 (1.0053) model_time 0.9239 (1.0046) loss 0.7297 (0.8026) grad_norm 11.1988 (8.5336/1.8935) mem 68106MB [2022-12-20 17:03:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][910/1519] eta 0:10:12 lr 0.000003 time 0.9295 (1.0056) model_time 0.9293 (1.0049) loss 0.7397 (0.8030) grad_norm 9.6299 (8.5571/1.9092) mem 68106MB [2022-12-20 17:03:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][920/1519] eta 0:10:02 lr 0.000003 time 0.9258 (1.0056) model_time 0.9257 (1.0049) loss 0.8156 (0.8026) grad_norm 8.1002 (8.5703/1.9061) mem 68106MB [2022-12-20 17:03:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][930/1519] eta 0:09:52 lr 0.000003 time 0.9333 (1.0056) model_time 0.9331 (1.0048) loss 0.7764 (0.8024) grad_norm 6.6828 (8.5782/1.9080) mem 68106MB [2022-12-20 17:03:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][940/1519] eta 0:09:42 lr 0.000003 time 0.9755 (1.0056) model_time 0.9753 (1.0048) loss 0.6713 (0.8021) grad_norm 6.3261 (8.5577/1.9103) mem 68106MB [2022-12-20 17:04:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][950/1519] eta 0:09:32 lr 0.000002 time 0.9274 (1.0055) model_time 0.9273 (1.0048) loss 0.7062 (0.8019) grad_norm 7.2195 (8.5401/1.9161) mem 68106MB [2022-12-20 17:04:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][960/1519] eta 0:09:22 lr 0.000002 time 0.9532 (1.0055) model_time 0.9530 (1.0047) loss 0.7599 (0.8016) grad_norm 11.3406 (8.5588/1.9235) mem 68106MB [2022-12-20 17:04:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][970/1519] eta 0:09:11 lr 0.000002 time 0.9211 (1.0054) model_time 0.9209 (1.0046) loss 0.8139 (0.8020) grad_norm 7.2638 (8.5467/1.9275) mem 68106MB [2022-12-20 17:04:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][980/1519] eta 0:09:01 lr 0.000002 time 0.9235 (1.0054) model_time 0.9234 (1.0046) loss 0.6806 (0.8017) grad_norm 11.8251 (8.5527/1.9354) mem 68106MB [2022-12-20 17:04:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][990/1519] eta 0:08:51 lr 0.000002 time 0.9345 (1.0053) model_time 0.9344 (1.0045) loss 1.0000 (0.8027) grad_norm 8.3962 (8.5632/1.9344) mem 68106MB [2022-12-20 17:04:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1000/1519] eta 0:08:41 lr 0.000002 time 0.9780 (1.0053) model_time 0.9778 (1.0045) loss 0.6692 (0.8026) grad_norm 10.4516 (8.5666/1.9328) mem 68106MB [2022-12-20 17:05:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1010/1519] eta 0:08:31 lr 0.000002 time 0.9272 (1.0052) model_time 0.9270 (1.0045) loss 0.9545 (0.8027) grad_norm 9.7893 (8.5558/1.9102) mem 68106MB [2022-12-20 17:05:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1020/1519] eta 0:08:21 lr 0.000002 time 0.9268 (1.0052) model_time 0.9267 (1.0045) loss 0.6787 (0.8030) grad_norm 10.4686 (8.5296/1.8784) mem 68106MB [2022-12-20 17:05:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1030/1519] eta 0:08:11 lr 0.000002 time 0.9634 (1.0053) model_time 0.9633 (1.0046) loss 0.6783 (0.8032) grad_norm 6.8255 (8.5469/1.9016) mem 68106MB [2022-12-20 17:05:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1040/1519] eta 0:08:01 lr 0.000002 time 1.0317 (1.0054) model_time 1.0316 (1.0046) loss 0.7762 (0.8032) grad_norm 6.7169 (8.5272/1.8838) mem 68106MB [2022-12-20 17:05:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1050/1519] eta 0:07:51 lr 0.000002 time 0.9322 (1.0053) model_time 0.9320 (1.0046) loss 0.6583 (0.8029) grad_norm 9.5707 (8.5038/1.8426) mem 68106MB [2022-12-20 17:05:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1060/1519] eta 0:07:41 lr 0.000002 time 0.9235 (1.0052) model_time 0.9234 (1.0045) loss 0.9330 (0.8038) grad_norm 8.5195 (8.4778/1.8511) mem 68106MB [2022-12-20 17:06:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1070/1519] eta 0:07:31 lr 0.000002 time 0.9210 (1.0051) model_time 0.9209 (1.0044) loss 0.8382 (0.8040) grad_norm 9.3302 (8.4921/1.8403) mem 68106MB [2022-12-20 17:06:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1080/1519] eta 0:07:21 lr 0.000002 time 0.9228 (1.0051) model_time 0.9227 (1.0044) loss 0.9771 (0.8040) grad_norm 10.4924 (8.4755/1.8310) mem 68106MB [2022-12-20 17:06:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1090/1519] eta 0:07:11 lr 0.000002 time 0.9225 (1.0050) model_time 0.9224 (1.0043) loss 0.9685 (0.8042) grad_norm 13.2083 (8.5001/1.8500) mem 68106MB [2022-12-20 17:06:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1100/1519] eta 0:07:01 lr 0.000002 time 0.9706 (1.0050) model_time 0.9705 (1.0043) loss 0.8838 (0.8048) grad_norm 7.0682 (8.5064/1.8517) mem 68106MB [2022-12-20 17:06:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1110/1519] eta 0:06:51 lr 0.000002 time 0.9232 (1.0050) model_time 0.9231 (1.0043) loss 0.6883 (0.8044) grad_norm 10.1163 (8.5422/1.8520) mem 68106MB [2022-12-20 17:06:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1120/1519] eta 0:06:40 lr 0.000002 time 0.9326 (1.0050) model_time 0.9324 (1.0043) loss 0.6940 (0.8044) grad_norm 9.5740 (8.5365/1.8658) mem 68106MB [2022-12-20 17:07:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1130/1519] eta 0:06:30 lr 0.000002 time 0.9312 (1.0049) model_time 0.9310 (1.0042) loss 0.7076 (0.8042) grad_norm 6.5737 (8.5438/1.8607) mem 68106MB [2022-12-20 17:07:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1140/1519] eta 0:06:20 lr 0.000002 time 0.9306 (1.0049) model_time 0.9305 (1.0042) loss 1.1301 (0.8043) grad_norm 6.1341 (8.5701/1.8704) mem 68106MB [2022-12-20 17:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1150/1519] eta 0:06:10 lr 0.000002 time 0.9202 (1.0049) model_time 0.9201 (1.0042) loss 0.9960 (0.8047) grad_norm 11.2714 (8.5485/1.8780) mem 68106MB [2022-12-20 17:07:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1160/1519] eta 0:06:00 lr 0.000002 time 0.9857 (1.0049) model_time 0.9856 (1.0042) loss 0.6623 (0.8043) grad_norm 9.4077 (8.5552/1.8788) mem 68106MB [2022-12-20 17:07:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1170/1519] eta 0:05:50 lr 0.000002 time 0.9259 (1.0051) model_time 0.9257 (1.0044) loss 0.7108 (0.8039) grad_norm 9.8878 (8.6094/1.9267) mem 68106MB [2022-12-20 17:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1180/1519] eta 0:05:40 lr 0.000002 time 0.9245 (1.0051) model_time 0.9243 (1.0044) loss 0.9309 (0.8046) grad_norm 12.0880 (8.6286/1.9368) mem 68106MB [2022-12-20 17:08:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1190/1519] eta 0:05:30 lr 0.000002 time 0.9222 (1.0052) model_time 0.9221 (1.0045) loss 0.8272 (0.8046) grad_norm 18.4517 (8.6585/2.0153) mem 68106MB [2022-12-20 17:08:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1200/1519] eta 0:05:20 lr 0.000002 time 0.9317 (1.0052) model_time 0.9316 (1.0045) loss 1.1513 (0.8049) grad_norm 6.3556 (8.6356/2.0035) mem 68106MB [2022-12-20 17:08:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1210/1519] eta 0:05:10 lr 0.000002 time 0.9259 (1.0052) model_time 0.9257 (1.0045) loss 1.0144 (0.8047) grad_norm 7.9883 (8.6378/2.0038) mem 68106MB [2022-12-20 17:08:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1220/1519] eta 0:05:00 lr 0.000002 time 0.9296 (1.0052) model_time 0.9295 (1.0046) loss 0.8879 (0.8047) grad_norm 7.6007 (8.6532/2.0867) mem 68106MB [2022-12-20 17:08:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1230/1519] eta 0:04:50 lr 0.000002 time 0.9685 (1.0052) model_time 0.9684 (1.0045) loss 0.7458 (0.8046) grad_norm 10.3947 (8.6641/2.0894) mem 68106MB [2022-12-20 17:08:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1240/1519] eta 0:04:40 lr 0.000002 time 0.9957 (1.0052) model_time 0.9956 (1.0045) loss 0.7033 (0.8043) grad_norm 5.7730 (8.6265/2.0976) mem 68106MB [2022-12-20 17:09:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1250/1519] eta 0:04:30 lr 0.000002 time 0.9362 (1.0051) model_time 0.9361 (1.0045) loss 0.7006 (0.8042) grad_norm 8.6981 (8.6784/2.1135) mem 68106MB [2022-12-20 17:09:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1260/1519] eta 0:04:20 lr 0.000002 time 0.9305 (1.0052) model_time 0.9303 (1.0045) loss 0.6853 (0.8049) grad_norm 7.8742 (8.6455/2.0320) mem 68106MB [2022-12-20 17:09:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1270/1519] eta 0:04:10 lr 0.000002 time 0.9186 (1.0051) model_time 0.9184 (1.0044) loss 0.7616 (0.8054) grad_norm 7.3299 (8.6634/2.0312) mem 68106MB [2022-12-20 17:09:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1280/1519] eta 0:04:00 lr 0.000002 time 0.9676 (1.0051) model_time 0.9673 (1.0044) loss 0.6817 (0.8053) grad_norm 8.2413 (8.6724/2.0346) mem 68106MB [2022-12-20 17:09:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1290/1519] eta 0:03:50 lr 0.000002 time 0.9201 (1.0052) model_time 0.9199 (1.0045) loss 0.8128 (0.8049) grad_norm 11.9351 (8.6750/2.0434) mem 68106MB [2022-12-20 17:09:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1300/1519] eta 0:03:40 lr 0.000002 time 0.9197 (1.0051) model_time 0.9195 (1.0045) loss 0.7589 (0.8047) grad_norm 6.9530 (8.6700/2.0526) mem 68106MB [2022-12-20 17:10:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1310/1519] eta 0:03:30 lr 0.000002 time 0.9291 (1.0051) model_time 0.9290 (1.0044) loss 0.7022 (0.8048) grad_norm 11.0162 (8.6507/1.9907) mem 68106MB [2022-12-20 17:10:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1320/1519] eta 0:03:20 lr 0.000002 time 0.9549 (1.0051) model_time 0.9548 (1.0045) loss 0.8839 (0.8051) grad_norm 7.1770 (8.6566/1.9977) mem 68106MB [2022-12-20 17:10:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1330/1519] eta 0:03:09 lr 0.000002 time 0.9327 (1.0051) model_time 0.9325 (1.0045) loss 0.7521 (0.8046) grad_norm 10.0494 (8.6470/2.0008) mem 68106MB [2022-12-20 17:10:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1340/1519] eta 0:02:59 lr 0.000002 time 0.9810 (1.0052) model_time 0.9808 (1.0046) loss 0.9078 (0.8046) grad_norm 7.8750 (8.6401/1.9952) mem 68106MB [2022-12-20 17:10:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1350/1519] eta 0:02:49 lr 0.000002 time 0.9207 (1.0053) model_time 0.9205 (1.0047) loss 0.8356 (0.8049) grad_norm 7.6774 (8.6331/1.9929) mem 68106MB [2022-12-20 17:10:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1360/1519] eta 0:02:39 lr 0.000002 time 0.9227 (1.0054) model_time 0.9226 (1.0048) loss 0.6677 (0.8043) grad_norm 7.7311 (8.6766/2.0785) mem 68106MB [2022-12-20 17:11:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1370/1519] eta 0:02:29 lr 0.000002 time 0.9202 (1.0054) model_time 0.9201 (1.0048) loss 0.8293 (0.8043) grad_norm 9.5203 (8.6765/2.0770) mem 68106MB [2022-12-20 17:11:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1380/1519] eta 0:02:19 lr 0.000002 time 0.9204 (1.0053) model_time 0.9203 (1.0047) loss 0.7755 (0.8041) grad_norm 7.2518 (8.7047/2.1157) mem 68106MB [2022-12-20 17:11:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1390/1519] eta 0:02:09 lr 0.000002 time 0.9261 (1.0053) model_time 0.9259 (1.0047) loss 0.9139 (0.8045) grad_norm 9.6495 (8.6875/2.1149) mem 68106MB [2022-12-20 17:11:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1400/1519] eta 0:01:59 lr 0.000002 time 0.9205 (1.0053) model_time 0.9203 (1.0046) loss 0.7341 (0.8040) grad_norm 6.9429 (8.7058/2.1151) mem 68106MB [2022-12-20 17:11:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1410/1519] eta 0:01:49 lr 0.000002 time 0.9804 (1.0052) model_time 0.9802 (1.0046) loss 0.7814 (0.8040) grad_norm 9.8039 (8.6920/2.1077) mem 68106MB [2022-12-20 17:11:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1420/1519] eta 0:01:39 lr 0.000002 time 0.9925 (1.0053) model_time 0.9924 (1.0047) loss 0.9289 (0.8044) grad_norm 8.6639 (8.7176/2.0926) mem 68106MB [2022-12-20 17:12:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1430/1519] eta 0:01:29 lr 0.000002 time 0.9235 (1.0052) model_time 0.9234 (1.0046) loss 0.9225 (0.8043) grad_norm 6.4135 (8.7377/2.1728) mem 68106MB [2022-12-20 17:12:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1440/1519] eta 0:01:19 lr 0.000002 time 0.9356 (1.0052) model_time 0.9355 (1.0046) loss 0.7070 (0.8044) grad_norm 12.9802 (8.7510/2.1865) mem 68106MB [2022-12-20 17:12:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1450/1519] eta 0:01:09 lr 0.000002 time 0.9261 (1.0052) model_time 0.9260 (1.0045) loss 0.6537 (0.8045) grad_norm 6.9284 (8.7269/2.1723) mem 68106MB [2022-12-20 17:12:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1460/1519] eta 0:00:59 lr 0.000002 time 0.9283 (1.0051) model_time 0.9282 (1.0045) loss 0.7422 (0.8045) grad_norm 8.8526 (8.7043/2.1743) mem 68106MB [2022-12-20 17:12:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1470/1519] eta 0:00:49 lr 0.000002 time 0.9267 (1.0051) model_time 0.9265 (1.0045) loss 0.8651 (0.8046) grad_norm 11.6812 (8.7171/2.1760) mem 68106MB [2022-12-20 17:12:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1480/1519] eta 0:00:39 lr 0.000002 time 0.9247 (1.0050) model_time 0.9246 (1.0044) loss 0.7342 (0.8043) grad_norm 6.0739 (8.7223/2.2144) mem 68106MB [2022-12-20 17:13:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1490/1519] eta 0:00:29 lr 0.000002 time 0.9261 (1.0050) model_time 0.9260 (1.0044) loss 0.6947 (0.8042) grad_norm 9.8354 (8.7318/2.2199) mem 68106MB [2022-12-20 17:13:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1500/1519] eta 0:00:19 lr 0.000002 time 0.9322 (1.0049) model_time 0.9320 (1.0043) loss 0.8924 (0.8042) grad_norm 8.7727 (8.7182/2.2113) mem 68106MB [2022-12-20 17:13:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [83/100][1510/1519] eta 0:00:09 lr 0.000002 time 0.9233 (1.0049) model_time 0.9232 (1.0043) loss 0.9334 (0.8049) grad_norm 6.5915 (8.6996/2.2150) mem 68106MB [2022-12-20 17:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 83 training takes 0:25:26 [2022-12-20 17:13:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_83.pth saving...... [2022-12-20 17:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_83.pth saved !!! [2022-12-20 17:14:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.628 (0.628) Loss 0.5403 (0.5403) Acc@1 91.667 (91.667) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 17:14:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.303 (0.328) Loss 0.5383 (0.5098) Acc@1 91.667 (92.519) Acc@5 98.264 (98.359) Mem 68106MB [2022-12-20 17:14:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4895 (0.5040) Acc@1 90.972 (92.708) Acc@5 98.958 (98.380) Mem 68106MB [2022-12-20 17:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.294 (0.309) Loss 0.6366 (0.5114) Acc@1 90.972 (92.440) Acc@5 97.569 (98.376) Mem 68106MB [2022-12-20 17:14:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.298 (0.307) Loss 0.4588 (0.5022) Acc@1 93.750 (92.505) Acc@5 99.306 (98.501) Mem 68106MB [2022-12-20 17:14:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.305) Loss 0.4917 (0.4995) Acc@1 92.361 (92.593) Acc@5 99.653 (98.563) Mem 68106MB [2022-12-20 17:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.5123 (0.4991) Acc@1 90.625 (92.515) Acc@5 98.264 (98.543) Mem 68106MB [2022-12-20 17:14:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.5440 (0.5001) Acc@1 93.056 (92.474) Acc@5 98.264 (98.538) Mem 68106MB [2022-12-20 17:14:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.302 (0.303) Loss 0.4338 (0.4986) Acc@1 93.403 (92.541) Acc@5 98.264 (98.573) Mem 68106MB [2022-12-20 17:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:83] * Acc@1 92.510 Acc@5 98.576 [2022-12-20 17:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 17:14:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.56% [2022-12-20 17:14:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][0/1519] eta 0:48:38 lr 0.000002 time 1.9211 (1.9211) model_time 1.2618 (1.2618) loss 0.7894 (0.7894) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 17:14:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][10/1519] eta 0:27:25 lr 0.000002 time 0.9315 (1.0903) model_time 0.9313 (1.0300) loss 0.6764 (0.8132) grad_norm 9.9464 (8.0724/0.9847) mem 68106MB [2022-12-20 17:14:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][20/1519] eta 0:26:11 lr 0.000002 time 0.9242 (1.0486) model_time 0.9241 (1.0169) loss 0.6901 (0.8111) grad_norm 13.0719 (8.5485/1.7167) mem 68106MB [2022-12-20 17:14:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][30/1519] eta 0:25:42 lr 0.000002 time 0.9335 (1.0360) model_time 0.9333 (1.0144) loss 0.6791 (0.8056) grad_norm 10.2561 (8.9433/1.7135) mem 68106MB [2022-12-20 17:15:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][40/1519] eta 0:25:19 lr 0.000002 time 0.9270 (1.0273) model_time 0.9269 (1.0109) loss 0.7043 (0.8038) grad_norm 11.8395 (8.8765/1.7931) mem 68106MB [2022-12-20 17:15:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][50/1519] eta 0:25:01 lr 0.000002 time 0.9190 (1.0219) model_time 0.9189 (1.0087) loss 0.7160 (0.7993) grad_norm 8.6244 (9.0998/1.9985) mem 68106MB [2022-12-20 17:15:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][60/1519] eta 0:24:48 lr 0.000002 time 0.9460 (1.0202) model_time 0.9457 (1.0090) loss 0.7286 (0.7930) grad_norm 8.3293 (8.9179/1.9126) mem 68106MB [2022-12-20 17:15:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][70/1519] eta 0:24:35 lr 0.000002 time 0.9867 (1.0182) model_time 0.9866 (1.0086) loss 0.7536 (0.8028) grad_norm 7.0485 (8.6726/1.8960) mem 68106MB [2022-12-20 17:15:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][80/1519] eta 0:24:25 lr 0.000002 time 0.9209 (1.0183) model_time 0.9207 (1.0099) loss 0.6627 (0.8082) grad_norm 5.8955 (8.6600/1.8680) mem 68106MB [2022-12-20 17:15:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][90/1519] eta 0:24:12 lr 0.000002 time 0.9256 (1.0164) model_time 0.9255 (1.0088) loss 0.9103 (0.8054) grad_norm 8.7634 (8.6434/1.7927) mem 68106MB [2022-12-20 17:16:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][100/1519] eta 0:23:59 lr 0.000002 time 0.9213 (1.0145) model_time 0.9212 (1.0077) loss 0.7891 (0.7987) grad_norm 7.8627 (8.6544/1.7313) mem 68106MB [2022-12-20 17:16:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][110/1519] eta 0:23:47 lr 0.000002 time 0.9295 (1.0133) model_time 0.9293 (1.0071) loss 0.7481 (0.8003) grad_norm 8.7906 (8.6964/1.7769) mem 68106MB [2022-12-20 17:16:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][120/1519] eta 0:23:36 lr 0.000002 time 0.9793 (1.0127) model_time 0.9792 (1.0070) loss 0.6993 (0.7977) grad_norm 7.7133 (8.6101/1.7362) mem 68106MB [2022-12-20 17:16:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][130/1519] eta 0:23:26 lr 0.000002 time 1.0075 (1.0124) model_time 1.0073 (1.0070) loss 0.8182 (0.7999) grad_norm 20.2079 (8.7175/2.2193) mem 68106MB [2022-12-20 17:16:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][140/1519] eta 0:23:15 lr 0.000002 time 0.9415 (1.0121) model_time 0.9412 (1.0071) loss 0.7244 (0.8029) grad_norm 7.0410 (8.7821/2.1869) mem 68106MB [2022-12-20 17:16:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][150/1519] eta 0:23:05 lr 0.000002 time 0.9219 (1.0124) model_time 0.9218 (1.0077) loss 0.9503 (0.8077) grad_norm 6.0036 (8.7528/2.1814) mem 68106MB [2022-12-20 17:17:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][160/1519] eta 0:22:56 lr 0.000002 time 0.9301 (1.0126) model_time 0.9299 (1.0081) loss 1.2220 (0.8123) grad_norm 8.3132 (8.7587/2.1656) mem 68106MB [2022-12-20 17:17:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][170/1519] eta 0:22:44 lr 0.000002 time 0.9224 (1.0119) model_time 0.9222 (1.0076) loss 0.9553 (0.8113) grad_norm 8.2581 (8.7859/2.1577) mem 68106MB [2022-12-20 17:17:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][180/1519] eta 0:22:34 lr 0.000002 time 0.9510 (1.0114) model_time 0.9507 (1.0074) loss 0.8928 (0.8099) grad_norm 6.8953 (8.7669/2.1248) mem 68106MB [2022-12-20 17:17:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][190/1519] eta 0:22:23 lr 0.000002 time 0.9379 (1.0108) model_time 0.9376 (1.0069) loss 0.9951 (0.8097) grad_norm 8.4188 (8.7588/2.0821) mem 68106MB [2022-12-20 17:17:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][200/1519] eta 0:22:12 lr 0.000002 time 0.9371 (1.0102) model_time 0.9369 (1.0066) loss 0.7733 (0.8118) grad_norm 7.5650 (8.7733/2.0678) mem 68106MB [2022-12-20 17:17:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][210/1519] eta 0:22:02 lr 0.000002 time 0.9342 (1.0099) model_time 0.9340 (1.0064) loss 0.7147 (0.8110) grad_norm 8.0953 (8.7998/2.0731) mem 68106MB [2022-12-20 17:18:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][220/1519] eta 0:21:51 lr 0.000002 time 0.9333 (1.0099) model_time 0.9331 (1.0065) loss 0.8803 (0.8124) grad_norm 8.0710 (8.7561/2.0493) mem 68106MB [2022-12-20 17:18:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][230/1519] eta 0:21:41 lr 0.000002 time 0.9349 (1.0094) model_time 0.9348 (1.0062) loss 1.0661 (0.8140) grad_norm 6.8189 (8.6977/2.0266) mem 68106MB [2022-12-20 17:18:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][240/1519] eta 0:21:30 lr 0.000002 time 0.9401 (1.0090) model_time 0.9400 (1.0059) loss 0.6747 (0.8109) grad_norm 8.4643 (8.6778/2.0082) mem 68106MB [2022-12-20 17:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][250/1519] eta 0:21:19 lr 0.000002 time 0.9324 (1.0087) model_time 0.9322 (1.0056) loss 1.3271 (0.8140) grad_norm 8.4412 (8.6612/1.9810) mem 68106MB [2022-12-20 17:18:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][260/1519] eta 0:21:10 lr 0.000002 time 0.9357 (1.0091) model_time 0.9356 (1.0062) loss 0.7699 (0.8140) grad_norm 7.6436 (8.6333/1.9593) mem 68106MB [2022-12-20 17:18:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][270/1519] eta 0:20:59 lr 0.000002 time 0.9319 (1.0087) model_time 0.9318 (1.0059) loss 0.7407 (0.8134) grad_norm 10.3233 (8.6722/1.9437) mem 68106MB [2022-12-20 17:19:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][280/1519] eta 0:20:49 lr 0.000002 time 0.9271 (1.0083) model_time 0.9270 (1.0056) loss 0.9469 (0.8114) grad_norm 12.4881 (8.7642/2.0524) mem 68106MB [2022-12-20 17:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][290/1519] eta 0:20:38 lr 0.000002 time 0.9448 (1.0081) model_time 0.9446 (1.0055) loss 0.6906 (0.8101) grad_norm 10.7473 (8.7885/2.0331) mem 68106MB [2022-12-20 17:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][300/1519] eta 0:20:28 lr 0.000002 time 0.9796 (1.0081) model_time 0.9795 (1.0056) loss 0.9352 (0.8097) grad_norm 6.6219 (8.7367/2.0217) mem 68106MB [2022-12-20 17:19:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][310/1519] eta 0:20:18 lr 0.000002 time 0.9390 (1.0078) model_time 0.9388 (1.0053) loss 1.2463 (0.8123) grad_norm 9.0373 (8.7879/2.0186) mem 68106MB [2022-12-20 17:19:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][320/1519] eta 0:20:08 lr 0.000002 time 0.9350 (1.0082) model_time 0.9348 (1.0057) loss 0.7339 (0.8124) grad_norm 5.7773 (8.7773/2.0185) mem 68106MB [2022-12-20 17:19:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][330/1519] eta 0:19:59 lr 0.000002 time 0.9324 (1.0088) model_time 0.9322 (1.0065) loss 0.7818 (0.8112) grad_norm 7.2865 (8.7550/1.9992) mem 68106MB [2022-12-20 17:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][340/1519] eta 0:19:50 lr 0.000002 time 0.9297 (1.0098) model_time 0.9295 (1.0075) loss 0.7015 (0.8119) grad_norm 9.0858 (8.7485/1.9917) mem 68106MB [2022-12-20 17:20:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][350/1519] eta 0:19:40 lr 0.000002 time 0.9294 (1.0099) model_time 0.9292 (1.0076) loss 0.8008 (0.8131) grad_norm 8.4756 (8.7560/1.9788) mem 68106MB [2022-12-20 17:20:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][360/1519] eta 0:19:30 lr 0.000002 time 0.9295 (1.0096) model_time 0.9293 (1.0074) loss 1.0054 (0.8115) grad_norm 7.0103 (8.7562/1.9711) mem 68106MB [2022-12-20 17:20:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][370/1519] eta 0:19:19 lr 0.000002 time 0.9426 (1.0095) model_time 0.9425 (1.0073) loss 0.8241 (0.8134) grad_norm 9.5153 (8.7377/1.9577) mem 68106MB [2022-12-20 17:20:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][380/1519] eta 0:19:09 lr 0.000002 time 0.9323 (1.0094) model_time 0.9321 (1.0072) loss 0.6980 (0.8133) grad_norm 6.4628 (8.7071/1.9508) mem 68106MB [2022-12-20 17:21:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][390/1519] eta 0:19:00 lr 0.000002 time 0.9173 (1.0102) model_time 0.9172 (1.0081) loss 0.6810 (0.8126) grad_norm 18.3194 (8.7526/2.0604) mem 68106MB [2022-12-20 17:21:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][400/1519] eta 0:18:50 lr 0.000002 time 0.9302 (1.0100) model_time 0.9300 (1.0080) loss 0.9733 (0.8125) grad_norm 6.1550 (8.7545/2.1361) mem 68106MB [2022-12-20 17:21:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][410/1519] eta 0:18:39 lr 0.000002 time 0.9261 (1.0098) model_time 0.9259 (1.0078) loss 0.7933 (0.8112) grad_norm 12.1925 (8.7682/2.1280) mem 68106MB [2022-12-20 17:21:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][420/1519] eta 0:18:29 lr 0.000002 time 0.9278 (1.0095) model_time 0.9276 (1.0076) loss 0.6655 (0.8099) grad_norm 9.3541 (8.7587/2.1051) mem 68106MB [2022-12-20 17:21:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][430/1519] eta 0:18:19 lr 0.000002 time 0.9310 (1.0093) model_time 0.9309 (1.0074) loss 0.7761 (0.8104) grad_norm 8.5248 (8.7366/2.0963) mem 68106MB [2022-12-20 17:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][440/1519] eta 0:18:08 lr 0.000002 time 0.9244 (1.0092) model_time 0.9237 (1.0073) loss 0.7846 (0.8115) grad_norm 7.6131 (8.7268/2.0818) mem 68106MB [2022-12-20 17:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][450/1519] eta 0:17:58 lr 0.000002 time 0.9203 (1.0092) model_time 0.9202 (1.0073) loss 0.6926 (0.8107) grad_norm 10.4496 (8.7285/2.0680) mem 68106MB [2022-12-20 17:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][460/1519] eta 0:17:48 lr 0.000002 time 0.9319 (1.0090) model_time 0.9316 (1.0071) loss 0.7390 (0.8116) grad_norm 7.9840 (8.7258/2.0573) mem 68106MB [2022-12-20 17:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][470/1519] eta 0:17:38 lr 0.000002 time 0.9294 (1.0088) model_time 0.9292 (1.0070) loss 0.8269 (0.8115) grad_norm 8.0833 (8.7181/2.0493) mem 68106MB [2022-12-20 17:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][480/1519] eta 0:17:28 lr 0.000002 time 0.9262 (1.0087) model_time 0.9260 (1.0069) loss 0.9050 (0.8115) grad_norm 6.3814 (8.7154/2.0978) mem 68106MB [2022-12-20 17:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][490/1519] eta 0:17:17 lr 0.000002 time 0.9322 (1.0085) model_time 0.9320 (1.0068) loss 0.6968 (0.8101) grad_norm 7.3339 (8.7046/2.0845) mem 68106MB [2022-12-20 17:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][500/1519] eta 0:17:08 lr 0.000002 time 0.9472 (1.0089) model_time 0.9470 (1.0072) loss 1.0551 (0.8109) grad_norm 8.8236 (8.6829/2.0810) mem 68106MB [2022-12-20 17:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][510/1519] eta 0:16:57 lr 0.000002 time 0.9342 (1.0088) model_time 0.9340 (1.0071) loss 0.7158 (0.8094) grad_norm 7.9195 (8.7049/2.1520) mem 68106MB [2022-12-20 17:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][520/1519] eta 0:16:47 lr 0.000002 time 0.9298 (1.0089) model_time 0.9296 (1.0072) loss 1.0714 (0.8102) grad_norm 8.9383 (8.6999/2.1390) mem 68106MB [2022-12-20 17:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][530/1519] eta 0:16:37 lr 0.000002 time 0.9335 (1.0088) model_time 0.9333 (1.0071) loss 0.9233 (0.8113) grad_norm 7.0781 (8.7428/2.2337) mem 68106MB [2022-12-20 17:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][540/1519] eta 0:16:27 lr 0.000002 time 0.9399 (1.0086) model_time 0.9397 (1.0070) loss 0.7449 (0.8116) grad_norm 14.5758 (8.7576/2.2493) mem 68106MB [2022-12-20 17:23:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][550/1519] eta 0:16:17 lr 0.000002 time 0.9406 (1.0085) model_time 0.9404 (1.0069) loss 0.6878 (0.8115) grad_norm 9.4582 (8.7467/2.2366) mem 68106MB [2022-12-20 17:23:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][560/1519] eta 0:16:06 lr 0.000002 time 0.9344 (1.0083) model_time 0.9342 (1.0067) loss 0.9120 (0.8120) grad_norm 8.8658 (8.7463/2.2238) mem 68106MB [2022-12-20 17:24:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][570/1519] eta 0:15:56 lr 0.000002 time 0.9199 (1.0084) model_time 0.9198 (1.0068) loss 0.8365 (0.8123) grad_norm 7.3801 (8.7269/2.2118) mem 68106MB [2022-12-20 17:24:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][580/1519] eta 0:15:46 lr 0.000002 time 0.9267 (1.0083) model_time 0.9266 (1.0067) loss 0.8199 (0.8114) grad_norm 8.4421 (8.7308/2.2013) mem 68106MB [2022-12-20 17:24:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][590/1519] eta 0:15:36 lr 0.000002 time 0.9081 (1.0081) model_time 0.9079 (1.0066) loss 0.6968 (0.8113) grad_norm 16.2417 (8.7492/2.2285) mem 68106MB [2022-12-20 17:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][600/1519] eta 0:15:26 lr 0.000002 time 0.9771 (1.0080) model_time 0.9768 (1.0065) loss 0.8741 (0.8117) grad_norm 10.1834 (8.7357/2.2171) mem 68106MB [2022-12-20 17:24:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][610/1519] eta 0:15:16 lr 0.000002 time 0.9319 (1.0082) model_time 0.9318 (1.0068) loss 1.0808 (0.8117) grad_norm 8.9059 (8.7435/2.2188) mem 68106MB [2022-12-20 17:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][620/1519] eta 0:15:06 lr 0.000002 time 0.9281 (1.0082) model_time 0.9280 (1.0067) loss 0.7816 (0.8109) grad_norm 8.5064 (8.7321/2.2095) mem 68106MB [2022-12-20 17:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][630/1519] eta 0:14:56 lr 0.000002 time 0.9335 (1.0084) model_time 0.9333 (1.0069) loss 0.6673 (0.8106) grad_norm 20.0469 (8.7383/2.2992) mem 68106MB [2022-12-20 17:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][640/1519] eta 0:14:46 lr 0.000002 time 0.9310 (1.0084) model_time 0.9308 (1.0070) loss 1.0940 (0.8112) grad_norm 11.5772 (8.7371/2.3045) mem 68106MB [2022-12-20 17:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][650/1519] eta 0:14:36 lr 0.000002 time 0.9111 (1.0085) model_time 0.9110 (1.0071) loss 0.7607 (0.8107) grad_norm 8.9306 (8.7099/2.2795) mem 68106MB [2022-12-20 17:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][660/1519] eta 0:14:26 lr 0.000002 time 0.9343 (1.0084) model_time 0.9341 (1.0070) loss 0.6655 (0.8105) grad_norm 8.6468 (8.7192/2.2817) mem 68106MB [2022-12-20 17:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][670/1519] eta 0:14:16 lr 0.000002 time 0.9299 (1.0083) model_time 0.9298 (1.0069) loss 0.7963 (0.8101) grad_norm 8.5257 (8.7183/2.2834) mem 68106MB [2022-12-20 17:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][680/1519] eta 0:14:05 lr 0.000002 time 0.9358 (1.0083) model_time 0.9356 (1.0069) loss 0.6670 (0.8098) grad_norm 10.2820 (8.7540/2.3102) mem 68106MB [2022-12-20 17:26:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][690/1519] eta 0:13:56 lr 0.000002 time 0.9232 (1.0085) model_time 0.9231 (1.0071) loss 0.7921 (0.8099) grad_norm 7.4025 (8.7452/2.3098) mem 68106MB [2022-12-20 17:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][700/1519] eta 0:13:46 lr 0.000002 time 1.0532 (1.0087) model_time 1.0531 (1.0073) loss 0.6826 (0.8098) grad_norm 8.0485 (8.7396/2.3127) mem 68106MB [2022-12-20 17:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][710/1519] eta 0:13:35 lr 0.000002 time 0.9362 (1.0086) model_time 0.9361 (1.0073) loss 0.8762 (0.8093) grad_norm 7.4198 (8.7237/2.2991) mem 68106MB [2022-12-20 17:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][720/1519] eta 0:13:25 lr 0.000002 time 0.9285 (1.0085) model_time 0.9283 (1.0072) loss 0.7155 (0.8088) grad_norm 7.7247 (8.7481/2.3651) mem 68106MB [2022-12-20 17:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][730/1519] eta 0:13:15 lr 0.000002 time 0.9366 (1.0084) model_time 0.9364 (1.0071) loss 0.7421 (0.8082) grad_norm 11.0819 (8.7782/2.3975) mem 68106MB [2022-12-20 17:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][740/1519] eta 0:13:05 lr 0.000002 time 0.9287 (1.0083) model_time 0.9285 (1.0070) loss 0.7297 (0.8097) grad_norm 16.2052 (8.8088/2.4394) mem 68106MB [2022-12-20 17:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][750/1519] eta 0:12:55 lr 0.000002 time 0.9363 (1.0082) model_time 0.9362 (1.0069) loss 0.9296 (0.8110) grad_norm 9.8134 (8.8510/2.4811) mem 68106MB [2022-12-20 17:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][760/1519] eta 0:12:45 lr 0.000002 time 0.9301 (1.0081) model_time 0.9300 (1.0068) loss 0.8424 (0.8111) grad_norm 8.5112 (8.8287/2.4769) mem 68106MB [2022-12-20 17:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][770/1519] eta 0:12:34 lr 0.000002 time 0.9367 (1.0080) model_time 0.9366 (1.0067) loss 0.9060 (0.8117) grad_norm 5.9078 (8.8359/2.5041) mem 68106MB [2022-12-20 17:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][780/1519] eta 0:12:24 lr 0.000002 time 0.9889 (1.0079) model_time 0.9888 (1.0067) loss 0.8133 (0.8116) grad_norm 10.2118 (8.8695/2.5204) mem 68106MB [2022-12-20 17:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][790/1519] eta 0:12:14 lr 0.000002 time 0.9323 (1.0079) model_time 0.9322 (1.0066) loss 0.9859 (0.8119) grad_norm 7.4164 (8.8590/2.5212) mem 68106MB [2022-12-20 17:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][800/1519] eta 0:12:04 lr 0.000002 time 0.9299 (1.0077) model_time 0.9298 (1.0065) loss 0.6656 (0.8116) grad_norm 14.4258 (8.8771/2.5332) mem 68106MB [2022-12-20 17:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][810/1519] eta 0:11:54 lr 0.000002 time 0.9236 (1.0077) model_time 0.9235 (1.0065) loss 0.7752 (0.8114) grad_norm 7.6193 (8.8533/2.5229) mem 68106MB [2022-12-20 17:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][820/1519] eta 0:11:44 lr 0.000002 time 0.9147 (1.0077) model_time 0.9146 (1.0065) loss 0.7840 (0.8126) grad_norm 8.3626 (8.8641/2.5189) mem 68106MB [2022-12-20 17:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][830/1519] eta 0:11:34 lr 0.000002 time 0.9334 (1.0076) model_time 0.9333 (1.0064) loss 0.7971 (0.8117) grad_norm 7.5565 (8.8722/2.5243) mem 68106MB [2022-12-20 17:28:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][840/1519] eta 0:11:24 lr 0.000002 time 0.9305 (1.0076) model_time 0.9304 (1.0064) loss 1.1686 (0.8115) grad_norm 11.2555 (8.8980/2.5408) mem 68106MB [2022-12-20 17:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][850/1519] eta 0:11:13 lr 0.000002 time 0.9308 (1.0075) model_time 0.9306 (1.0063) loss 0.6966 (0.8109) grad_norm 6.4635 (8.8970/2.5404) mem 68106MB [2022-12-20 17:28:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][860/1519] eta 0:11:03 lr 0.000002 time 0.9287 (1.0074) model_time 0.9286 (1.0063) loss 0.7284 (0.8109) grad_norm 7.9915 (8.8964/2.5409) mem 68106MB [2022-12-20 17:29:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][870/1519] eta 0:10:53 lr 0.000002 time 0.9275 (1.0074) model_time 0.9273 (1.0062) loss 0.7416 (0.8108) grad_norm 10.8299 (8.8815/2.5492) mem 68106MB [2022-12-20 17:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][880/1519] eta 0:10:43 lr 0.000002 time 0.9726 (1.0073) model_time 0.9724 (1.0062) loss 0.9971 (0.8109) grad_norm 7.9271 (8.8606/2.5338) mem 68106MB [2022-12-20 17:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][890/1519] eta 0:10:33 lr 0.000002 time 0.9248 (1.0075) model_time 0.9246 (1.0063) loss 0.8911 (0.8101) grad_norm 6.3664 (8.8358/2.5375) mem 68106MB [2022-12-20 17:29:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][900/1519] eta 0:10:23 lr 0.000002 time 0.9312 (1.0074) model_time 0.9310 (1.0063) loss 0.7472 (0.8096) grad_norm 7.1737 (8.8507/2.5376) mem 68106MB [2022-12-20 17:29:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][910/1519] eta 0:10:13 lr 0.000002 time 0.9407 (1.0074) model_time 0.9406 (1.0062) loss 0.6888 (0.8088) grad_norm 8.5343 (8.8111/2.5295) mem 68106MB [2022-12-20 17:29:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][920/1519] eta 0:10:03 lr 0.000002 time 0.9402 (1.0073) model_time 0.9400 (1.0062) loss 0.8630 (0.8089) grad_norm 11.3750 (8.8380/2.5478) mem 68106MB [2022-12-20 17:30:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][930/1519] eta 0:09:53 lr 0.000002 time 0.9300 (1.0073) model_time 0.9299 (1.0062) loss 0.7965 (0.8089) grad_norm 6.2910 (8.8459/2.5524) mem 68106MB [2022-12-20 17:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][940/1519] eta 0:09:43 lr 0.000002 time 1.1828 (1.0076) model_time 1.1825 (1.0065) loss 0.9325 (0.8088) grad_norm 9.6016 (8.8616/2.5573) mem 68106MB [2022-12-20 17:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][950/1519] eta 0:09:33 lr 0.000002 time 0.9349 (1.0077) model_time 0.9348 (1.0066) loss 0.7087 (0.8094) grad_norm 9.4714 (8.8512/2.5618) mem 68106MB [2022-12-20 17:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][960/1519] eta 0:09:23 lr 0.000002 time 0.9322 (1.0076) model_time 0.9320 (1.0066) loss 0.6968 (0.8091) grad_norm 6.2542 (8.8637/2.5861) mem 68106MB [2022-12-20 17:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][970/1519] eta 0:09:13 lr 0.000002 time 0.9424 (1.0077) model_time 0.9422 (1.0066) loss 0.6630 (0.8093) grad_norm 7.4419 (8.8584/2.5857) mem 68106MB [2022-12-20 17:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][980/1519] eta 0:09:03 lr 0.000002 time 0.9292 (1.0076) model_time 0.9290 (1.0065) loss 0.6933 (0.8089) grad_norm 7.5207 (8.8699/2.5792) mem 68106MB [2022-12-20 17:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][990/1519] eta 0:08:52 lr 0.000002 time 0.9346 (1.0075) model_time 0.9345 (1.0064) loss 0.8803 (0.8084) grad_norm 10.7835 (8.8512/2.5215) mem 68106MB [2022-12-20 17:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1000/1519] eta 0:08:42 lr 0.000002 time 0.9346 (1.0074) model_time 0.9344 (1.0064) loss 0.8286 (0.8085) grad_norm 13.7520 (8.8813/2.4948) mem 68106MB [2022-12-20 17:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1010/1519] eta 0:08:32 lr 0.000002 time 0.9769 (1.0074) model_time 0.9767 (1.0064) loss 1.0863 (0.8090) grad_norm 6.8899 (8.8916/2.5190) mem 68106MB [2022-12-20 17:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1020/1519] eta 0:08:22 lr 0.000002 time 0.9701 (1.0077) model_time 0.9700 (1.0066) loss 0.7978 (0.8091) grad_norm 9.4259 (8.9195/2.5384) mem 68106MB [2022-12-20 17:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1030/1519] eta 0:08:12 lr 0.000002 time 0.9901 (1.0076) model_time 0.9899 (1.0066) loss 0.7080 (0.8091) grad_norm 10.0134 (8.9351/2.5337) mem 68106MB [2022-12-20 17:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1040/1519] eta 0:08:02 lr 0.000002 time 0.9259 (1.0075) model_time 0.9257 (1.0065) loss 0.8464 (0.8089) grad_norm 7.6490 (8.9393/2.5359) mem 68106MB [2022-12-20 17:32:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1050/1519] eta 0:07:52 lr 0.000002 time 0.9327 (1.0075) model_time 0.9326 (1.0065) loss 0.6681 (0.8086) grad_norm 6.4094 (8.9285/2.5554) mem 68106MB [2022-12-20 17:32:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1060/1519] eta 0:07:42 lr 0.000002 time 0.9306 (1.0074) model_time 0.9304 (1.0064) loss 0.9225 (0.8092) grad_norm 9.5761 (8.9383/2.5564) mem 68106MB [2022-12-20 17:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1070/1519] eta 0:07:32 lr 0.000002 time 0.9252 (1.0073) model_time 0.9251 (1.0063) loss 0.8589 (0.8091) grad_norm 6.1086 (8.9477/2.5608) mem 68106MB [2022-12-20 17:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1080/1519] eta 0:07:22 lr 0.000002 time 0.9402 (1.0073) model_time 0.9401 (1.0063) loss 0.8201 (0.8085) grad_norm 6.4610 (8.9413/2.5197) mem 68106MB [2022-12-20 17:32:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1090/1519] eta 0:07:12 lr 0.000002 time 0.9340 (1.0072) model_time 0.9338 (1.0062) loss 0.8173 (0.8082) grad_norm 10.1286 (8.9602/2.5182) mem 68106MB [2022-12-20 17:32:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1100/1519] eta 0:07:02 lr 0.000002 time 0.9385 (1.0072) model_time 0.9384 (1.0062) loss 0.9050 (0.8079) grad_norm 6.0121 (8.9621/2.5139) mem 68106MB [2022-12-20 17:33:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1110/1519] eta 0:06:51 lr 0.000002 time 0.9377 (1.0072) model_time 0.9375 (1.0062) loss 0.7730 (0.8077) grad_norm 10.2302 (8.9442/2.4581) mem 68106MB [2022-12-20 17:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1120/1519] eta 0:06:41 lr 0.000002 time 0.9860 (1.0072) model_time 0.9858 (1.0062) loss 0.7170 (0.8074) grad_norm 9.1778 (8.9488/2.4520) mem 68106MB [2022-12-20 17:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1130/1519] eta 0:06:31 lr 0.000002 time 0.9338 (1.0072) model_time 0.9337 (1.0062) loss 1.1542 (0.8075) grad_norm 7.2089 (8.8937/2.3701) mem 68106MB [2022-12-20 17:33:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1140/1519] eta 0:06:21 lr 0.000002 time 0.9309 (1.0072) model_time 0.9307 (1.0062) loss 0.6896 (0.8073) grad_norm 9.1280 (8.8951/2.3447) mem 68106MB [2022-12-20 17:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1150/1519] eta 0:06:11 lr 0.000002 time 0.9247 (1.0072) model_time 0.9245 (1.0063) loss 0.8924 (0.8075) grad_norm 8.6211 (8.8990/2.3435) mem 68106MB [2022-12-20 17:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1160/1519] eta 0:06:01 lr 0.000002 time 0.9343 (1.0072) model_time 0.9342 (1.0062) loss 0.7196 (0.8071) grad_norm 9.3792 (8.9382/2.3615) mem 68106MB [2022-12-20 17:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1170/1519] eta 0:05:51 lr 0.000002 time 0.9114 (1.0071) model_time 0.9112 (1.0062) loss 0.6973 (0.8073) grad_norm 9.1603 (8.9670/2.3694) mem 68106MB [2022-12-20 17:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1180/1519] eta 0:05:41 lr 0.000002 time 0.9399 (1.0071) model_time 0.9398 (1.0062) loss 0.9490 (0.8079) grad_norm 6.1153 (8.9306/2.3797) mem 68106MB [2022-12-20 17:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1190/1519] eta 0:05:31 lr 0.000002 time 0.9332 (1.0071) model_time 0.9331 (1.0061) loss 0.6732 (0.8075) grad_norm 7.1390 (8.9024/2.3446) mem 68106MB [2022-12-20 17:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1200/1519] eta 0:05:21 lr 0.000002 time 0.9580 (1.0072) model_time 0.9577 (1.0062) loss 0.9793 (0.8074) grad_norm 7.4352 (8.9128/2.3432) mem 68106MB [2022-12-20 17:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1210/1519] eta 0:05:11 lr 0.000002 time 0.9297 (1.0071) model_time 0.9296 (1.0061) loss 0.6629 (0.8071) grad_norm 7.7666 (8.9082/2.3462) mem 68106MB [2022-12-20 17:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1220/1519] eta 0:05:01 lr 0.000002 time 0.9283 (1.0070) model_time 0.9280 (1.0061) loss 0.6711 (0.8067) grad_norm 7.3273 (8.9297/2.3746) mem 68106MB [2022-12-20 17:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1230/1519] eta 0:04:51 lr 0.000002 time 0.9292 (1.0069) model_time 0.9290 (1.0060) loss 0.7569 (0.8064) grad_norm 8.6100 (8.9296/2.3141) mem 68106MB [2022-12-20 17:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1240/1519] eta 0:04:40 lr 0.000002 time 0.9335 (1.0070) model_time 0.9334 (1.0061) loss 0.6656 (0.8063) grad_norm 7.0501 (8.9220/2.3000) mem 68106MB [2022-12-20 17:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1250/1519] eta 0:04:30 lr 0.000002 time 0.9294 (1.0070) model_time 0.9293 (1.0061) loss 0.8399 (0.8057) grad_norm 8.0751 (8.9247/2.3207) mem 68106MB [2022-12-20 17:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1260/1519] eta 0:04:20 lr 0.000002 time 1.0371 (1.0071) model_time 1.0370 (1.0062) loss 0.6926 (0.8060) grad_norm 7.4305 (8.9285/2.3241) mem 68106MB [2022-12-20 17:35:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1270/1519] eta 0:04:10 lr 0.000002 time 1.0024 (1.0071) model_time 1.0022 (1.0062) loss 0.7841 (0.8056) grad_norm 8.0558 (8.9421/2.3135) mem 68106MB [2022-12-20 17:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1280/1519] eta 0:04:00 lr 0.000002 time 0.9315 (1.0072) model_time 0.9314 (1.0063) loss 0.7036 (0.8062) grad_norm 5.3160 (8.8961/2.2970) mem 68106MB [2022-12-20 17:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1290/1519] eta 0:03:50 lr 0.000002 time 0.9346 (1.0072) model_time 0.9344 (1.0063) loss 0.6698 (0.8060) grad_norm 9.4832 (8.9039/2.3015) mem 68106MB [2022-12-20 17:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1300/1519] eta 0:03:40 lr 0.000002 time 0.9346 (1.0071) model_time 0.9344 (1.0062) loss 0.6635 (0.8062) grad_norm 8.3800 (8.9140/2.3046) mem 68106MB [2022-12-20 17:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1310/1519] eta 0:03:30 lr 0.000002 time 0.9284 (1.0070) model_time 0.9282 (1.0061) loss 0.7217 (0.8060) grad_norm 6.2900 (8.9045/2.3167) mem 68106MB [2022-12-20 17:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1320/1519] eta 0:03:20 lr 0.000002 time 0.9436 (1.0070) model_time 0.9435 (1.0061) loss 0.7233 (0.8061) grad_norm 10.2284 (8.8937/2.2474) mem 68106MB [2022-12-20 17:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1330/1519] eta 0:03:10 lr 0.000002 time 0.9224 (1.0069) model_time 0.9221 (1.0060) loss 0.6712 (0.8058) grad_norm 5.7276 (8.8622/2.1843) mem 68106MB [2022-12-20 17:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1340/1519] eta 0:03:00 lr 0.000002 time 0.9218 (1.0069) model_time 0.9216 (1.0060) loss 0.7336 (0.8058) grad_norm 10.1426 (8.8292/2.1408) mem 68106MB [2022-12-20 17:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1350/1519] eta 0:02:50 lr 0.000002 time 0.9379 (1.0069) model_time 0.9378 (1.0060) loss 0.6739 (0.8055) grad_norm 9.9199 (8.7848/2.0997) mem 68106MB [2022-12-20 17:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1360/1519] eta 0:02:40 lr 0.000002 time 0.9324 (1.0068) model_time 0.9323 (1.0059) loss 0.9298 (0.8058) grad_norm 8.1246 (8.7880/2.0971) mem 68106MB [2022-12-20 17:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1370/1519] eta 0:02:30 lr 0.000002 time 0.9297 (1.0067) model_time 0.9295 (1.0059) loss 0.8037 (0.8058) grad_norm 7.8543 (8.7647/2.0720) mem 68106MB [2022-12-20 17:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1380/1519] eta 0:02:19 lr 0.000002 time 0.9180 (1.0068) model_time 0.9179 (1.0059) loss 1.0593 (0.8056) grad_norm 7.3649 (8.7145/2.0506) mem 68106MB [2022-12-20 17:37:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1390/1519] eta 0:02:09 lr 0.000002 time 0.9315 (1.0068) model_time 0.9313 (1.0059) loss 0.6698 (0.8050) grad_norm 8.0149 (8.7107/2.0502) mem 68106MB [2022-12-20 17:37:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1400/1519] eta 0:01:59 lr 0.000002 time 0.9289 (1.0067) model_time 0.9287 (1.0059) loss 0.9670 (0.8047) grad_norm 14.6709 (8.6945/2.0686) mem 68106MB [2022-12-20 17:38:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1410/1519] eta 0:01:49 lr 0.000002 time 0.9353 (1.0067) model_time 0.9352 (1.0058) loss 0.7461 (0.8046) grad_norm 6.5748 (8.6919/2.0747) mem 68106MB [2022-12-20 17:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1420/1519] eta 0:01:39 lr 0.000002 time 0.9221 (1.0069) model_time 0.9219 (1.0060) loss 0.8547 (0.8051) grad_norm 8.7083 (8.7016/2.0752) mem 68106MB [2022-12-20 17:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1430/1519] eta 0:01:29 lr 0.000002 time 0.9276 (1.0068) model_time 0.9274 (1.0059) loss 0.6836 (0.8055) grad_norm 8.7951 (8.7326/2.0684) mem 68106MB [2022-12-20 17:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1440/1519] eta 0:01:19 lr 0.000002 time 1.0312 (1.0071) model_time 1.0311 (1.0063) loss 0.7495 (0.8053) grad_norm 6.4669 (8.7184/2.0445) mem 68106MB [2022-12-20 17:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1450/1519] eta 0:01:09 lr 0.000002 time 0.9947 (1.0071) model_time 0.9945 (1.0063) loss 0.6817 (0.8049) grad_norm 7.8914 (8.7172/2.0458) mem 68106MB [2022-12-20 17:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1460/1519] eta 0:00:59 lr 0.000002 time 0.9111 (1.0072) model_time 0.9110 (1.0064) loss 0.6815 (0.8047) grad_norm 10.2105 (8.7143/2.0488) mem 68106MB [2022-12-20 17:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1470/1519] eta 0:00:49 lr 0.000002 time 0.9222 (1.0072) model_time 0.9221 (1.0063) loss 1.0235 (0.8048) grad_norm 6.9429 (8.7054/2.0480) mem 68106MB [2022-12-20 17:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1480/1519] eta 0:00:39 lr 0.000002 time 0.9227 (1.0071) model_time 0.9226 (1.0063) loss 0.6714 (0.8048) grad_norm 9.9711 (8.6914/2.0091) mem 68106MB [2022-12-20 17:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1490/1519] eta 0:00:29 lr 0.000002 time 0.9168 (1.0071) model_time 0.9167 (1.0063) loss 0.7643 (0.8048) grad_norm 7.8897 (8.7194/2.0083) mem 68106MB [2022-12-20 17:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1500/1519] eta 0:00:19 lr 0.000002 time 0.9223 (1.0071) model_time 0.9222 (1.0062) loss 0.7222 (0.8047) grad_norm 7.9179 (8.7208/2.0034) mem 68106MB [2022-12-20 17:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [84/100][1510/1519] eta 0:00:09 lr 0.000002 time 0.9235 (1.0071) model_time 0.9234 (1.0063) loss 0.7121 (0.8046) grad_norm 6.4911 (8.7256/2.0151) mem 68106MB [2022-12-20 17:39:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 84 training takes 0:25:29 [2022-12-20 17:39:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_84.pth saving...... [2022-12-20 17:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_84.pth saved !!! [2022-12-20 17:40:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.635 (0.635) Loss 0.5353 (0.5353) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 17:40:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.328) Loss 0.5332 (0.5071) Acc@1 92.361 (92.866) Acc@5 98.264 (98.422) Mem 68106MB [2022-12-20 17:40:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.295 (0.313) Loss 0.4828 (0.5008) Acc@1 91.319 (92.890) Acc@5 98.958 (98.396) Mem 68106MB [2022-12-20 17:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.309) Loss 0.6301 (0.5078) Acc@1 90.972 (92.608) Acc@5 98.264 (98.376) Mem 68106MB [2022-12-20 17:40:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.299 (0.306) Loss 0.4561 (0.4982) Acc@1 93.750 (92.649) Acc@5 99.306 (98.501) Mem 68106MB [2022-12-20 17:40:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.296 (0.305) Loss 0.4893 (0.4955) Acc@1 91.667 (92.695) Acc@5 99.653 (98.563) Mem 68106MB [2022-12-20 17:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.5063 (0.4951) Acc@1 90.972 (92.612) Acc@5 98.264 (98.543) Mem 68106MB [2022-12-20 17:40:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5451 (0.4964) Acc@1 92.708 (92.547) Acc@5 98.611 (98.543) Mem 68106MB [2022-12-20 17:40:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4300 (0.4949) Acc@1 93.056 (92.597) Acc@5 98.264 (98.581) Mem 68106MB [2022-12-20 17:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:84] * Acc@1 92.559 Acc@5 98.584 [2022-12-20 17:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 17:40:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 17:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 17:41:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.56% [2022-12-20 17:41:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][0/1519] eta 0:34:46 lr 0.000002 time 1.3738 (1.3738) model_time 0.9987 (0.9987) loss 0.6773 (0.6773) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 17:41:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][10/1519] eta 0:26:07 lr 0.000002 time 0.9314 (1.0385) model_time 0.9313 (1.0041) loss 1.0178 (0.7786) grad_norm 6.9235 (7.7627/1.4590) mem 68106MB [2022-12-20 17:41:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][20/1519] eta 0:25:29 lr 0.000002 time 0.9230 (1.0200) model_time 0.9229 (1.0019) loss 0.6847 (0.7961) grad_norm 7.4641 (8.3578/1.4603) mem 68106MB [2022-12-20 17:41:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][30/1519] eta 0:25:15 lr 0.000002 time 0.9276 (1.0177) model_time 0.9275 (1.0053) loss 0.6856 (0.7879) grad_norm 7.7666 (8.3610/1.3987) mem 68106MB [2022-12-20 17:41:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][40/1519] eta 0:24:58 lr 0.000002 time 0.9258 (1.0130) model_time 0.9257 (1.0035) loss 0.8400 (0.7793) grad_norm 8.4622 (8.2434/1.3730) mem 68106MB [2022-12-20 17:42:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][50/1519] eta 0:24:44 lr 0.000002 time 0.9223 (1.0104) model_time 0.9222 (1.0028) loss 1.0271 (0.7854) grad_norm 8.9735 (8.4332/1.3211) mem 68106MB [2022-12-20 17:42:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][60/1519] eta 0:24:31 lr 0.000002 time 0.9243 (1.0085) model_time 0.9242 (1.0020) loss 0.8067 (0.7937) grad_norm 8.8551 (8.4686/1.3380) mem 68106MB [2022-12-20 17:42:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][70/1519] eta 0:24:19 lr 0.000002 time 0.9300 (1.0074) model_time 0.9298 (1.0018) loss 0.8391 (0.8017) grad_norm 7.1060 (8.2868/1.3616) mem 68106MB [2022-12-20 17:42:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][80/1519] eta 0:24:08 lr 0.000002 time 0.9232 (1.0064) model_time 0.9230 (1.0015) loss 0.8355 (0.8062) grad_norm 9.2822 (8.2038/1.3913) mem 68106MB [2022-12-20 17:42:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][90/1519] eta 0:23:57 lr 0.000002 time 0.9394 (1.0063) model_time 0.9392 (1.0018) loss 0.8572 (0.8087) grad_norm 8.4158 (8.3536/1.5558) mem 68106MB [2022-12-20 17:42:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][100/1519] eta 0:23:46 lr 0.000002 time 0.9224 (1.0055) model_time 0.9223 (1.0015) loss 0.8402 (0.8070) grad_norm 5.9926 (8.3102/1.5260) mem 68106MB [2022-12-20 17:43:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][110/1519] eta 0:23:36 lr 0.000002 time 0.9297 (1.0051) model_time 0.9296 (1.0014) loss 0.8506 (0.8058) grad_norm 7.5947 (8.3493/1.5781) mem 68106MB [2022-12-20 17:43:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][120/1519] eta 0:23:26 lr 0.000002 time 0.9301 (1.0050) model_time 0.9300 (1.0016) loss 0.6708 (0.8068) grad_norm 7.5018 (8.2662/1.5483) mem 68106MB [2022-12-20 17:43:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][130/1519] eta 0:23:15 lr 0.000002 time 0.9231 (1.0045) model_time 0.9229 (1.0013) loss 0.7612 (0.8046) grad_norm 8.0284 (8.2438/1.5178) mem 68106MB [2022-12-20 17:43:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][140/1519] eta 0:23:04 lr 0.000002 time 0.9347 (1.0040) model_time 0.9344 (1.0010) loss 0.6816 (0.8033) grad_norm 6.4081 (8.3083/1.5625) mem 68106MB [2022-12-20 17:43:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][150/1519] eta 0:22:54 lr 0.000002 time 0.9261 (1.0042) model_time 0.9259 (1.0014) loss 0.7288 (0.8008) grad_norm 7.3093 (8.2898/1.5460) mem 68106MB [2022-12-20 17:43:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][160/1519] eta 0:22:44 lr 0.000002 time 0.9276 (1.0042) model_time 0.9274 (1.0016) loss 0.6920 (0.7976) grad_norm 7.9174 (8.3293/1.5483) mem 68106MB [2022-12-20 17:44:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][170/1519] eta 0:22:35 lr 0.000002 time 0.9253 (1.0047) model_time 0.9251 (1.0021) loss 0.9103 (0.8028) grad_norm 9.8716 (8.3213/1.5419) mem 68106MB [2022-12-20 17:44:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][180/1519] eta 0:22:24 lr 0.000002 time 0.9357 (1.0044) model_time 0.9355 (1.0019) loss 0.7771 (0.8052) grad_norm 7.8308 (8.3570/1.5211) mem 68106MB [2022-12-20 17:44:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][190/1519] eta 0:22:14 lr 0.000002 time 0.9332 (1.0045) model_time 0.9330 (1.0021) loss 0.8116 (0.8043) grad_norm 7.8244 (8.3237/1.4945) mem 68106MB [2022-12-20 17:44:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][200/1519] eta 0:22:04 lr 0.000002 time 0.9341 (1.0042) model_time 0.9338 (1.0020) loss 0.8526 (0.8081) grad_norm 8.7323 (8.3662/1.5036) mem 68106MB [2022-12-20 17:44:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][210/1519] eta 0:21:55 lr 0.000002 time 0.9270 (1.0050) model_time 0.9268 (1.0029) loss 1.1564 (0.8092) grad_norm 8.0580 (8.3063/1.4964) mem 68106MB [2022-12-20 17:44:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][220/1519] eta 0:21:45 lr 0.000002 time 0.9274 (1.0049) model_time 0.9272 (1.0028) loss 0.6869 (0.8075) grad_norm 8.8053 (8.3262/1.4996) mem 68106MB [2022-12-20 17:45:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][230/1519] eta 0:21:35 lr 0.000002 time 0.9305 (1.0049) model_time 0.9304 (1.0029) loss 0.7528 (0.8037) grad_norm 7.1416 (8.3195/1.4876) mem 68106MB [2022-12-20 17:45:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][240/1519] eta 0:21:25 lr 0.000002 time 0.9225 (1.0051) model_time 0.9221 (1.0032) loss 0.6684 (0.8024) grad_norm 7.4580 (8.3170/1.5393) mem 68106MB [2022-12-20 17:45:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][250/1519] eta 0:21:16 lr 0.000002 time 0.9328 (1.0060) model_time 0.9325 (1.0041) loss 0.8493 (0.8011) grad_norm 11.1499 (8.3945/1.6208) mem 68106MB [2022-12-20 17:45:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][260/1519] eta 0:21:06 lr 0.000002 time 0.9306 (1.0062) model_time 0.9304 (1.0045) loss 0.9320 (0.8032) grad_norm 6.5542 (8.3915/1.6098) mem 68106MB [2022-12-20 17:45:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][270/1519] eta 0:20:56 lr 0.000002 time 0.9347 (1.0061) model_time 0.9346 (1.0044) loss 0.6844 (0.8032) grad_norm 9.0610 (8.4203/1.6015) mem 68106MB [2022-12-20 17:45:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][280/1519] eta 0:20:46 lr 0.000002 time 0.9342 (1.0058) model_time 0.9339 (1.0041) loss 0.8876 (0.8025) grad_norm 8.6637 (8.3984/1.5893) mem 68106MB [2022-12-20 17:46:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][290/1519] eta 0:20:36 lr 0.000002 time 0.9266 (1.0057) model_time 0.9264 (1.0041) loss 0.7054 (0.8044) grad_norm 7.9141 (8.3875/1.5740) mem 68106MB [2022-12-20 17:46:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][300/1519] eta 0:20:26 lr 0.000002 time 0.9782 (1.0065) model_time 0.9780 (1.0049) loss 1.1101 (0.8060) grad_norm 8.4588 (8.3793/1.5702) mem 68106MB [2022-12-20 17:46:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][310/1519] eta 0:20:16 lr 0.000002 time 0.9335 (1.0063) model_time 0.9333 (1.0047) loss 0.6937 (0.8067) grad_norm 6.8914 (8.3521/1.5598) mem 68106MB [2022-12-20 17:46:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][320/1519] eta 0:20:06 lr 0.000002 time 0.9309 (1.0060) model_time 0.9308 (1.0045) loss 0.8393 (0.8069) grad_norm 8.9550 (8.3715/1.5400) mem 68106MB [2022-12-20 17:46:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][330/1519] eta 0:19:55 lr 0.000002 time 0.9331 (1.0058) model_time 0.9330 (1.0043) loss 0.6698 (0.8070) grad_norm 13.7821 (8.3893/1.5819) mem 68106MB [2022-12-20 17:46:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][340/1519] eta 0:19:45 lr 0.000002 time 0.9343 (1.0058) model_time 0.9340 (1.0044) loss 0.7024 (0.8052) grad_norm 6.6886 (8.3645/1.5786) mem 68106MB [2022-12-20 17:47:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][350/1519] eta 0:19:35 lr 0.000002 time 0.9500 (1.0056) model_time 0.9499 (1.0042) loss 0.8962 (0.8052) grad_norm 8.7683 (8.3875/1.5823) mem 68106MB [2022-12-20 17:47:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][360/1519] eta 0:19:25 lr 0.000002 time 0.9349 (1.0055) model_time 0.9346 (1.0041) loss 0.8800 (0.8058) grad_norm 6.3185 (8.3857/1.6051) mem 68106MB [2022-12-20 17:47:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][370/1519] eta 0:19:15 lr 0.000002 time 0.9300 (1.0053) model_time 0.9298 (1.0039) loss 0.7278 (0.8043) grad_norm 6.2282 (8.3865/1.6068) mem 68106MB [2022-12-20 17:47:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][380/1519] eta 0:19:04 lr 0.000002 time 0.9289 (1.0052) model_time 0.9286 (1.0038) loss 1.0147 (0.8049) grad_norm 6.5724 (8.3765/1.6095) mem 68106MB [2022-12-20 17:47:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][390/1519] eta 0:18:54 lr 0.000002 time 0.9357 (1.0051) model_time 0.9354 (1.0038) loss 1.0024 (0.8066) grad_norm 8.0223 (8.3829/1.6100) mem 68106MB [2022-12-20 17:47:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][400/1519] eta 0:18:44 lr 0.000002 time 0.9307 (1.0050) model_time 0.9305 (1.0037) loss 0.6629 (0.8057) grad_norm 8.8330 (8.3767/1.6158) mem 68106MB [2022-12-20 17:48:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][410/1519] eta 0:18:34 lr 0.000002 time 0.9371 (1.0050) model_time 0.9370 (1.0037) loss 1.0960 (0.8063) grad_norm 8.5310 (8.4214/1.6276) mem 68106MB [2022-12-20 17:48:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][420/1519] eta 0:18:24 lr 0.000002 time 0.9399 (1.0049) model_time 0.9398 (1.0036) loss 0.6862 (0.8062) grad_norm 10.3206 (8.4367/1.6229) mem 68106MB [2022-12-20 17:48:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][430/1519] eta 0:18:14 lr 0.000002 time 0.9361 (1.0049) model_time 0.9359 (1.0036) loss 0.9392 (0.8091) grad_norm 7.0907 (8.4105/1.6220) mem 68106MB [2022-12-20 17:48:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][440/1519] eta 0:18:04 lr 0.000002 time 0.9329 (1.0049) model_time 0.9327 (1.0037) loss 0.7864 (0.8090) grad_norm 7.9032 (8.4095/1.6493) mem 68106MB [2022-12-20 17:48:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][450/1519] eta 0:17:54 lr 0.000002 time 0.9550 (1.0048) model_time 0.9549 (1.0036) loss 0.8149 (0.8097) grad_norm 6.1637 (8.3964/1.6553) mem 68106MB [2022-12-20 17:48:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][460/1519] eta 0:17:43 lr 0.000002 time 0.9320 (1.0046) model_time 0.9318 (1.0034) loss 0.8401 (0.8114) grad_norm 10.3837 (8.4517/1.6994) mem 68106MB [2022-12-20 17:49:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][470/1519] eta 0:17:34 lr 0.000002 time 0.9352 (1.0049) model_time 0.9351 (1.0038) loss 1.0073 (0.8115) grad_norm 8.5646 (8.4668/1.7066) mem 68106MB [2022-12-20 17:49:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][480/1519] eta 0:17:24 lr 0.000002 time 0.9984 (1.0053) model_time 0.9982 (1.0041) loss 0.6635 (0.8107) grad_norm 7.8599 (8.4520/1.6933) mem 68106MB [2022-12-20 17:49:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][490/1519] eta 0:17:14 lr 0.000002 time 0.9256 (1.0051) model_time 0.9255 (1.0040) loss 0.7520 (0.8109) grad_norm 16.5495 (8.4921/1.7735) mem 68106MB [2022-12-20 17:49:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][500/1519] eta 0:17:04 lr 0.000002 time 0.9225 (1.0051) model_time 0.9223 (1.0039) loss 0.7047 (0.8092) grad_norm 8.4334 (8.4934/1.7981) mem 68106MB [2022-12-20 17:49:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][510/1519] eta 0:16:53 lr 0.000002 time 0.9321 (1.0049) model_time 0.9320 (1.0038) loss 0.7138 (0.8088) grad_norm 7.5027 (8.4911/1.8044) mem 68106MB [2022-12-20 17:49:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][520/1519] eta 0:16:44 lr 0.000002 time 0.9346 (1.0050) model_time 0.9345 (1.0039) loss 0.6741 (0.8087) grad_norm 8.8834 (8.4904/1.7910) mem 68106MB [2022-12-20 17:50:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][530/1519] eta 0:16:33 lr 0.000002 time 0.9378 (1.0050) model_time 0.9375 (1.0040) loss 1.0376 (0.8090) grad_norm 7.5439 (8.4703/1.7838) mem 68106MB [2022-12-20 17:50:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][540/1519] eta 0:16:23 lr 0.000002 time 0.9325 (1.0050) model_time 0.9323 (1.0039) loss 0.8102 (0.8099) grad_norm 9.3487 (8.4827/1.7910) mem 68106MB [2022-12-20 17:50:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][550/1519] eta 0:16:13 lr 0.000002 time 0.9293 (1.0051) model_time 0.9291 (1.0041) loss 0.8529 (0.8098) grad_norm 13.1198 (8.4968/1.7985) mem 68106MB [2022-12-20 17:50:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][560/1519] eta 0:16:03 lr 0.000002 time 0.9701 (1.0052) model_time 0.9700 (1.0042) loss 0.6513 (0.8089) grad_norm 8.5261 (8.4889/1.7915) mem 68106MB [2022-12-20 17:50:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][570/1519] eta 0:15:53 lr 0.000002 time 0.9337 (1.0051) model_time 0.9335 (1.0040) loss 0.9430 (0.8093) grad_norm 8.0391 (8.4885/1.7842) mem 68106MB [2022-12-20 17:50:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][580/1519] eta 0:15:43 lr 0.000002 time 0.9292 (1.0050) model_time 0.9291 (1.0040) loss 1.0030 (0.8093) grad_norm 8.2461 (8.4760/1.7870) mem 68106MB [2022-12-20 17:51:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][590/1519] eta 0:15:33 lr 0.000002 time 0.9313 (1.0047) model_time 0.9310 (1.0037) loss 1.0053 (0.8092) grad_norm inf (8.4880/1.8019) mem 68106MB [2022-12-20 17:51:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][600/1519] eta 0:15:23 lr 0.000002 time 0.9259 (1.0046) model_time 0.9257 (1.0036) loss 0.7009 (0.8076) grad_norm 7.1015 (8.4678/1.7939) mem 68106MB [2022-12-20 17:51:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][610/1519] eta 0:15:13 lr 0.000002 time 0.9289 (1.0046) model_time 0.9288 (1.0036) loss 0.6943 (0.8075) grad_norm 7.2763 (8.4673/1.7984) mem 68106MB [2022-12-20 17:51:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][620/1519] eta 0:15:03 lr 0.000002 time 0.9196 (1.0046) model_time 0.9195 (1.0036) loss 0.6790 (0.8075) grad_norm 9.5706 (8.4449/1.8006) mem 68106MB [2022-12-20 17:51:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][630/1519] eta 0:14:52 lr 0.000002 time 0.9384 (1.0045) model_time 0.9382 (1.0035) loss 0.9048 (0.8076) grad_norm 8.8535 (8.4498/1.8074) mem 68106MB [2022-12-20 17:51:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][640/1519] eta 0:14:42 lr 0.000002 time 0.9297 (1.0044) model_time 0.9295 (1.0034) loss 0.8201 (0.8078) grad_norm 8.7983 (8.4824/1.8876) mem 68106MB [2022-12-20 17:52:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][650/1519] eta 0:14:32 lr 0.000002 time 0.9310 (1.0044) model_time 0.9308 (1.0035) loss 0.7185 (0.8078) grad_norm 10.4081 (8.4949/1.8956) mem 68106MB [2022-12-20 17:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][660/1519] eta 0:14:22 lr 0.000002 time 0.9791 (1.0045) model_time 0.9790 (1.0035) loss 0.8909 (0.8076) grad_norm 9.7504 (8.4797/1.8959) mem 68106MB [2022-12-20 17:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][670/1519] eta 0:14:12 lr 0.000002 time 0.9323 (1.0044) model_time 0.9322 (1.0034) loss 0.8214 (0.8083) grad_norm 7.9689 (8.4905/1.8893) mem 68106MB [2022-12-20 17:52:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][680/1519] eta 0:14:02 lr 0.000002 time 0.9333 (1.0043) model_time 0.9331 (1.0034) loss 0.9158 (0.8082) grad_norm 9.0153 (8.5121/1.8766) mem 68106MB [2022-12-20 17:52:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][690/1519] eta 0:13:52 lr 0.000002 time 0.9267 (1.0042) model_time 0.9265 (1.0033) loss 0.6787 (0.8080) grad_norm 7.3192 (8.5193/1.9654) mem 68106MB [2022-12-20 17:52:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][700/1519] eta 0:13:42 lr 0.000002 time 0.9429 (1.0042) model_time 0.9427 (1.0033) loss 0.9408 (0.8077) grad_norm 10.8307 (8.5217/1.9741) mem 68106MB [2022-12-20 17:53:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][710/1519] eta 0:13:32 lr 0.000002 time 0.9293 (1.0042) model_time 0.9292 (1.0033) loss 0.6592 (0.8071) grad_norm 10.0540 (8.5179/1.9598) mem 68106MB [2022-12-20 17:53:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][720/1519] eta 0:13:22 lr 0.000002 time 0.9154 (1.0043) model_time 0.9152 (1.0034) loss 0.7918 (0.8070) grad_norm 6.2427 (8.5212/1.9627) mem 68106MB [2022-12-20 17:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][730/1519] eta 0:13:12 lr 0.000002 time 0.9313 (1.0044) model_time 0.9312 (1.0035) loss 0.8671 (0.8074) grad_norm 6.2088 (8.5869/2.0749) mem 68106MB [2022-12-20 17:53:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][740/1519] eta 0:13:02 lr 0.000002 time 0.9972 (1.0045) model_time 0.9968 (1.0036) loss 1.0671 (0.8072) grad_norm 6.8495 (8.5712/2.0730) mem 68106MB [2022-12-20 17:53:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][750/1519] eta 0:12:52 lr 0.000002 time 0.9259 (1.0046) model_time 0.9257 (1.0037) loss 0.9470 (0.8071) grad_norm 7.3187 (8.5883/2.0703) mem 68106MB [2022-12-20 17:53:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][760/1519] eta 0:12:42 lr 0.000002 time 0.9337 (1.0045) model_time 0.9335 (1.0036) loss 0.8982 (0.8073) grad_norm 8.3854 (8.5974/2.0691) mem 68106MB [2022-12-20 17:54:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][770/1519] eta 0:12:32 lr 0.000002 time 0.9376 (1.0045) model_time 0.9373 (1.0036) loss 0.8131 (0.8068) grad_norm 7.6266 (8.5926/2.0739) mem 68106MB [2022-12-20 17:54:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][780/1519] eta 0:12:22 lr 0.000002 time 0.9589 (1.0046) model_time 0.9587 (1.0037) loss 0.9588 (0.8066) grad_norm 5.1199 (8.5566/2.0819) mem 68106MB [2022-12-20 17:54:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][790/1519] eta 0:12:12 lr 0.000002 time 0.9264 (1.0047) model_time 0.9262 (1.0038) loss 0.8462 (0.8074) grad_norm 4.8060 (8.5700/2.1070) mem 68106MB [2022-12-20 17:54:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][800/1519] eta 0:12:02 lr 0.000002 time 0.9318 (1.0048) model_time 0.9316 (1.0039) loss 0.6569 (0.8065) grad_norm 10.7923 (8.5817/2.1489) mem 68106MB [2022-12-20 17:54:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][810/1519] eta 0:11:52 lr 0.000002 time 0.9208 (1.0047) model_time 0.9201 (1.0039) loss 0.7770 (0.8067) grad_norm 15.7630 (8.6299/2.1835) mem 68106MB [2022-12-20 17:54:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][820/1519] eta 0:11:42 lr 0.000002 time 0.9284 (1.0046) model_time 0.9282 (1.0038) loss 0.8494 (0.8073) grad_norm 11.6644 (8.6291/2.1887) mem 68106MB [2022-12-20 17:55:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][830/1519] eta 0:11:32 lr 0.000002 time 0.9371 (1.0049) model_time 0.9369 (1.0041) loss 0.6762 (0.8068) grad_norm 6.8288 (8.6361/2.1915) mem 68106MB [2022-12-20 17:55:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][840/1519] eta 0:11:22 lr 0.000002 time 0.9244 (1.0049) model_time 0.9242 (1.0040) loss 0.6766 (0.8060) grad_norm 10.2522 (8.6739/2.1914) mem 68106MB [2022-12-20 17:55:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][850/1519] eta 0:11:12 lr 0.000002 time 0.9339 (1.0049) model_time 0.9337 (1.0040) loss 0.6858 (0.8061) grad_norm 5.4786 (8.6579/2.1787) mem 68106MB [2022-12-20 17:55:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][860/1519] eta 0:11:02 lr 0.000002 time 0.9386 (1.0048) model_time 0.9385 (1.0040) loss 1.0090 (0.8067) grad_norm 7.4662 (8.6338/2.1732) mem 68106MB [2022-12-20 17:55:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][870/1519] eta 0:10:52 lr 0.000002 time 0.9259 (1.0047) model_time 0.9258 (1.0039) loss 0.7514 (0.8068) grad_norm 9.5487 (8.6270/2.1703) mem 68106MB [2022-12-20 17:55:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][880/1519] eta 0:10:42 lr 0.000002 time 0.9575 (1.0047) model_time 0.9574 (1.0039) loss 0.6995 (0.8071) grad_norm 11.3327 (8.6435/2.1724) mem 68106MB [2022-12-20 17:56:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][890/1519] eta 0:10:31 lr 0.000002 time 0.9351 (1.0047) model_time 0.9349 (1.0039) loss 0.6963 (0.8071) grad_norm 6.7582 (8.6486/2.1731) mem 68106MB [2022-12-20 17:56:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][900/1519] eta 0:10:21 lr 0.000002 time 0.9301 (1.0046) model_time 0.9300 (1.0038) loss 0.7718 (0.8065) grad_norm 7.5947 (8.6478/2.1683) mem 68106MB [2022-12-20 17:56:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][910/1519] eta 0:10:11 lr 0.000002 time 0.9470 (1.0046) model_time 0.9468 (1.0038) loss 0.9159 (0.8069) grad_norm 13.9835 (8.6729/2.1904) mem 68106MB [2022-12-20 17:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][920/1519] eta 0:10:01 lr 0.000002 time 0.9288 (1.0045) model_time 0.9286 (1.0037) loss 0.7425 (0.8064) grad_norm 6.5835 (8.6514/2.2026) mem 68106MB [2022-12-20 17:56:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][930/1519] eta 0:09:51 lr 0.000002 time 0.9244 (1.0044) model_time 0.9243 (1.0037) loss 0.6664 (0.8069) grad_norm 8.4081 (8.6625/2.2083) mem 68106MB [2022-12-20 17:56:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][940/1519] eta 0:09:41 lr 0.000002 time 0.9338 (1.0044) model_time 0.9337 (1.0036) loss 0.8226 (0.8072) grad_norm 8.1928 (8.6415/2.1940) mem 68106MB [2022-12-20 17:57:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][950/1519] eta 0:09:31 lr 0.000002 time 0.9386 (1.0044) model_time 0.9384 (1.0036) loss 1.0663 (0.8070) grad_norm 7.3640 (8.6653/2.2109) mem 68106MB [2022-12-20 17:57:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][960/1519] eta 0:09:21 lr 0.000002 time 1.0125 (1.0044) model_time 1.0124 (1.0036) loss 0.9693 (0.8063) grad_norm 11.7384 (8.6647/2.2085) mem 68106MB [2022-12-20 17:57:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][970/1519] eta 0:09:11 lr 0.000002 time 0.9337 (1.0044) model_time 0.9336 (1.0036) loss 0.7060 (0.8064) grad_norm 8.4446 (8.6509/2.2060) mem 68106MB [2022-12-20 17:57:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][980/1519] eta 0:09:01 lr 0.000002 time 0.9250 (1.0044) model_time 0.9248 (1.0037) loss 0.8053 (0.8063) grad_norm 5.2077 (8.6364/2.2112) mem 68106MB [2022-12-20 17:57:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][990/1519] eta 0:08:51 lr 0.000002 time 0.9316 (1.0044) model_time 0.9315 (1.0037) loss 0.6481 (0.8060) grad_norm 8.6534 (8.6365/2.2005) mem 68106MB [2022-12-20 17:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1000/1519] eta 0:08:41 lr 0.000002 time 0.9265 (1.0044) model_time 0.9264 (1.0036) loss 0.7265 (0.8064) grad_norm 7.0877 (8.6400/2.1906) mem 68106MB [2022-12-20 17:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1010/1519] eta 0:08:31 lr 0.000002 time 0.9316 (1.0045) model_time 0.9315 (1.0037) loss 0.7691 (0.8064) grad_norm 8.0546 (8.6153/2.1891) mem 68106MB [2022-12-20 17:58:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1020/1519] eta 0:08:21 lr 0.000002 time 0.9280 (1.0045) model_time 0.9278 (1.0037) loss 0.7868 (0.8073) grad_norm 8.2248 (8.6142/2.1858) mem 68106MB [2022-12-20 17:58:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1030/1519] eta 0:08:11 lr 0.000002 time 0.9275 (1.0044) model_time 0.9270 (1.0037) loss 0.7419 (0.8067) grad_norm 9.7381 (8.6441/2.1987) mem 68106MB [2022-12-20 17:58:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1040/1519] eta 0:08:01 lr 0.000002 time 0.9111 (1.0047) model_time 0.9110 (1.0039) loss 0.7438 (0.8065) grad_norm 12.5281 (8.6551/2.2013) mem 68106MB [2022-12-20 17:58:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1050/1519] eta 0:07:51 lr 0.000002 time 0.9108 (1.0047) model_time 0.9106 (1.0040) loss 0.8809 (0.8064) grad_norm 7.7980 (8.6651/2.1935) mem 68106MB [2022-12-20 17:58:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1060/1519] eta 0:07:41 lr 0.000002 time 0.9412 (1.0049) model_time 0.9411 (1.0041) loss 0.6657 (0.8071) grad_norm 8.7744 (8.6367/2.1642) mem 68106MB [2022-12-20 17:59:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1070/1519] eta 0:07:31 lr 0.000002 time 0.9314 (1.0048) model_time 0.9313 (1.0041) loss 0.7211 (0.8069) grad_norm 8.8873 (8.6064/2.1563) mem 68106MB [2022-12-20 17:59:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1080/1519] eta 0:07:21 lr 0.000002 time 0.9215 (1.0048) model_time 0.9212 (1.0041) loss 0.9371 (0.8067) grad_norm 8.0684 (8.6063/2.1574) mem 68106MB [2022-12-20 17:59:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1090/1519] eta 0:07:11 lr 0.000002 time 0.9256 (1.0048) model_time 0.9255 (1.0041) loss 0.6889 (0.8065) grad_norm 8.2785 (8.5987/2.1446) mem 68106MB [2022-12-20 17:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1100/1519] eta 0:07:01 lr 0.000002 time 0.9606 (1.0048) model_time 0.9605 (1.0041) loss 0.7224 (0.8062) grad_norm 6.3045 (8.5516/2.0725) mem 68106MB [2022-12-20 17:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1110/1519] eta 0:06:51 lr 0.000002 time 0.9315 (1.0049) model_time 0.9314 (1.0042) loss 0.7959 (0.8064) grad_norm 12.3332 (8.5789/2.0740) mem 68106MB [2022-12-20 17:59:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1120/1519] eta 0:06:40 lr 0.000002 time 0.9280 (1.0049) model_time 0.9279 (1.0042) loss 0.9332 (0.8070) grad_norm 8.8301 (8.5714/2.0747) mem 68106MB [2022-12-20 18:00:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1130/1519] eta 0:06:30 lr 0.000002 time 0.9264 (1.0048) model_time 0.9262 (1.0041) loss 0.7759 (0.8070) grad_norm 7.5351 (8.6004/2.1063) mem 68106MB [2022-12-20 18:00:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1140/1519] eta 0:06:20 lr 0.000002 time 1.0073 (1.0048) model_time 1.0071 (1.0041) loss 0.6591 (0.8068) grad_norm 9.6807 (8.6269/2.1025) mem 68106MB [2022-12-20 18:00:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1150/1519] eta 0:06:10 lr 0.000002 time 0.9231 (1.0048) model_time 0.9229 (1.0041) loss 0.6536 (0.8067) grad_norm 7.2323 (8.6324/2.1063) mem 68106MB [2022-12-20 18:00:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1160/1519] eta 0:06:00 lr 0.000002 time 0.9337 (1.0048) model_time 0.9336 (1.0041) loss 1.1062 (0.8068) grad_norm 10.8038 (8.6296/2.0940) mem 68106MB [2022-12-20 18:00:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1170/1519] eta 0:05:50 lr 0.000002 time 0.9175 (1.0047) model_time 0.9174 (1.0040) loss 0.7741 (0.8067) grad_norm 5.2463 (8.6164/2.1023) mem 68106MB [2022-12-20 18:00:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1180/1519] eta 0:05:40 lr 0.000002 time 0.9245 (1.0047) model_time 0.9242 (1.0040) loss 0.8663 (0.8064) grad_norm 7.4171 (8.6232/2.0934) mem 68106MB [2022-12-20 18:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1190/1519] eta 0:05:30 lr 0.000002 time 0.9369 (1.0047) model_time 0.9367 (1.0040) loss 0.9503 (0.8069) grad_norm 8.9234 (8.6243/2.0777) mem 68106MB [2022-12-20 18:01:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1200/1519] eta 0:05:20 lr 0.000002 time 0.9420 (1.0047) model_time 0.9418 (1.0040) loss 1.1132 (0.8070) grad_norm 8.3989 (8.6615/2.1041) mem 68106MB [2022-12-20 18:01:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1210/1519] eta 0:05:10 lr 0.000002 time 0.9359 (1.0046) model_time 0.9357 (1.0040) loss 0.6843 (0.8065) grad_norm 11.6693 (8.7018/2.1458) mem 68106MB [2022-12-20 18:01:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1220/1519] eta 0:05:00 lr 0.000002 time 0.9302 (1.0047) model_time 0.9300 (1.0040) loss 0.6762 (0.8064) grad_norm 10.1766 (8.7282/2.1426) mem 68106MB [2022-12-20 18:01:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1230/1519] eta 0:04:50 lr 0.000002 time 0.9321 (1.0046) model_time 0.9319 (1.0039) loss 0.8278 (0.8062) grad_norm 10.3157 (8.7290/2.1392) mem 68106MB [2022-12-20 18:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1240/1519] eta 0:04:40 lr 0.000002 time 0.9298 (1.0046) model_time 0.9296 (1.0039) loss 0.7728 (0.8057) grad_norm 5.9351 (8.7113/2.0735) mem 68106MB [2022-12-20 18:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1250/1519] eta 0:04:30 lr 0.000002 time 0.9319 (1.0046) model_time 0.9318 (1.0039) loss 0.8655 (0.8061) grad_norm 9.4089 (8.7078/2.0733) mem 68106MB [2022-12-20 18:02:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1260/1519] eta 0:04:20 lr 0.000002 time 0.9313 (1.0046) model_time 0.9311 (1.0039) loss 0.6701 (0.8054) grad_norm 7.3234 (8.7065/2.0722) mem 68106MB [2022-12-20 18:02:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1270/1519] eta 0:04:10 lr 0.000002 time 0.9401 (1.0047) model_time 0.9399 (1.0040) loss 0.6837 (0.8056) grad_norm 7.0953 (8.7162/2.0923) mem 68106MB [2022-12-20 18:02:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1280/1519] eta 0:04:00 lr 0.000002 time 0.9123 (1.0047) model_time 0.9120 (1.0040) loss 0.9157 (0.8056) grad_norm 8.2237 (8.7114/2.0929) mem 68106MB [2022-12-20 18:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1290/1519] eta 0:03:50 lr 0.000002 time 0.9300 (1.0047) model_time 0.9299 (1.0041) loss 0.8585 (0.8051) grad_norm 5.9729 (8.6807/2.0112) mem 68106MB [2022-12-20 18:02:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1300/1519] eta 0:03:40 lr 0.000002 time 0.9289 (1.0047) model_time 0.9287 (1.0041) loss 0.8914 (0.8057) grad_norm 6.2415 (8.6500/2.0192) mem 68106MB [2022-12-20 18:03:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1310/1519] eta 0:03:29 lr 0.000002 time 0.9355 (1.0048) model_time 0.9353 (1.0041) loss 0.9380 (0.8064) grad_norm 10.2300 (8.6492/2.0190) mem 68106MB [2022-12-20 18:03:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1320/1519] eta 0:03:19 lr 0.000002 time 0.9198 (1.0049) model_time 0.9194 (1.0042) loss 0.6847 (0.8065) grad_norm 6.3722 (8.6460/2.0198) mem 68106MB [2022-12-20 18:03:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1330/1519] eta 0:03:09 lr 0.000002 time 0.9630 (1.0049) model_time 0.9628 (1.0043) loss 0.8168 (0.8063) grad_norm 9.1855 (8.5768/1.9089) mem 68106MB [2022-12-20 18:03:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1340/1519] eta 0:02:59 lr 0.000002 time 0.9402 (1.0049) model_time 0.9400 (1.0042) loss 0.8244 (0.8061) grad_norm 6.7840 (8.5511/1.9135) mem 68106MB [2022-12-20 18:03:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1350/1519] eta 0:02:49 lr 0.000002 time 0.9246 (1.0051) model_time 0.9244 (1.0044) loss 0.8510 (0.8061) grad_norm 7.8313 (8.5549/1.9117) mem 68106MB [2022-12-20 18:04:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1360/1519] eta 0:02:39 lr 0.000002 time 0.9779 (1.0052) model_time 0.9777 (1.0045) loss 0.6989 (0.8062) grad_norm 8.5125 (8.5335/1.9151) mem 68106MB [2022-12-20 18:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1370/1519] eta 0:02:29 lr 0.000002 time 0.9784 (1.0052) model_time 0.9782 (1.0046) loss 0.7661 (0.8060) grad_norm 9.9260 (8.5611/1.9121) mem 68106MB [2022-12-20 18:04:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1380/1519] eta 0:02:19 lr 0.000002 time 0.9347 (1.0052) model_time 0.9345 (1.0045) loss 0.7624 (0.8058) grad_norm 5.7333 (8.5672/1.9088) mem 68106MB [2022-12-20 18:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1390/1519] eta 0:02:09 lr 0.000002 time 0.9392 (1.0052) model_time 0.9390 (1.0045) loss 0.8941 (0.8056) grad_norm 15.6801 (8.5969/1.9222) mem 68106MB [2022-12-20 18:04:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1400/1519] eta 0:01:59 lr 0.000002 time 0.9754 (1.0052) model_time 0.9751 (1.0045) loss 0.9278 (0.8057) grad_norm 10.4891 (8.6111/1.9023) mem 68106MB [2022-12-20 18:04:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1410/1519] eta 0:01:49 lr 0.000002 time 0.9335 (1.0051) model_time 0.9333 (1.0044) loss 0.6742 (0.8059) grad_norm 8.8389 (8.5792/1.8531) mem 68106MB [2022-12-20 18:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1420/1519] eta 0:01:39 lr 0.000002 time 0.9320 (1.0053) model_time 0.9319 (1.0046) loss 0.6981 (0.8058) grad_norm 5.8015 (8.5835/1.8506) mem 68106MB [2022-12-20 18:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1430/1519] eta 0:01:29 lr 0.000002 time 0.9292 (1.0052) model_time 0.9291 (1.0046) loss 0.7643 (0.8055) grad_norm 7.0425 (8.5611/1.8476) mem 68106MB [2022-12-20 18:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1440/1519] eta 0:01:19 lr 0.000002 time 0.9340 (1.0052) model_time 0.9339 (1.0045) loss 0.6666 (0.8055) grad_norm 10.4061 (8.5293/1.8287) mem 68106MB [2022-12-20 18:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1450/1519] eta 0:01:09 lr 0.000002 time 0.9278 (1.0051) model_time 0.9277 (1.0045) loss 0.9216 (0.8053) grad_norm 9.3879 (8.5181/1.8237) mem 68106MB [2022-12-20 18:05:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1460/1519] eta 0:00:59 lr 0.000002 time 0.9265 (1.0051) model_time 0.9263 (1.0045) loss 0.8511 (0.8054) grad_norm 10.0717 (8.5151/1.8329) mem 68106MB [2022-12-20 18:05:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1470/1519] eta 0:00:49 lr 0.000002 time 0.9413 (1.0051) model_time 0.9411 (1.0044) loss 0.9347 (0.8056) grad_norm 8.8291 (8.5274/1.8286) mem 68106MB [2022-12-20 18:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1480/1519] eta 0:00:39 lr 0.000002 time 0.9300 (1.0050) model_time 0.9298 (1.0044) loss 0.6645 (0.8050) grad_norm 6.1785 (8.5029/1.8266) mem 68106MB [2022-12-20 18:06:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1490/1519] eta 0:00:29 lr 0.000002 time 0.9287 (1.0050) model_time 0.9286 (1.0044) loss 0.8821 (0.8057) grad_norm 7.2551 (8.5006/1.8359) mem 68106MB [2022-12-20 18:06:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1500/1519] eta 0:00:19 lr 0.000002 time 0.9677 (1.0050) model_time 0.9675 (1.0043) loss 0.7156 (0.8058) grad_norm 8.9707 (8.4999/1.8342) mem 68106MB [2022-12-20 18:06:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [85/100][1510/1519] eta 0:00:09 lr 0.000002 time 0.9204 (1.0049) model_time 0.9203 (1.0043) loss 0.6761 (0.8060) grad_norm 9.2229 (8.5116/1.8546) mem 68106MB [2022-12-20 18:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 85 training takes 0:25:26 [2022-12-20 18:06:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_85.pth saving...... [2022-12-20 18:07:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_85.pth saved !!! [2022-12-20 18:07:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.698 (0.698) Loss 0.5368 (0.5368) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 18:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.298 (0.335) Loss 0.5324 (0.5079) Acc@1 92.708 (92.835) Acc@5 98.264 (98.422) Mem 68106MB [2022-12-20 18:07:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.300 (0.317) Loss 0.4806 (0.5021) Acc@1 91.667 (92.824) Acc@5 98.958 (98.462) Mem 68106MB [2022-12-20 18:07:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.311) Loss 0.6332 (0.5092) Acc@1 90.625 (92.496) Acc@5 97.917 (98.421) Mem 68106MB [2022-12-20 18:07:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.308) Loss 0.4597 (0.5000) Acc@1 93.750 (92.598) Acc@5 99.306 (98.535) Mem 68106MB [2022-12-20 18:07:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.306) Loss 0.4875 (0.4973) Acc@1 92.014 (92.702) Acc@5 99.653 (98.597) Mem 68106MB [2022-12-20 18:07:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.306) Loss 0.5062 (0.4967) Acc@1 91.319 (92.629) Acc@5 98.264 (98.566) Mem 68106MB [2022-12-20 18:07:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.301 (0.305) Loss 0.5431 (0.4980) Acc@1 93.056 (92.562) Acc@5 97.917 (98.548) Mem 68106MB [2022-12-20 18:07:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.304) Loss 0.4300 (0.4966) Acc@1 93.403 (92.601) Acc@5 98.611 (98.590) Mem 68106MB [2022-12-20 18:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:85] * Acc@1 92.571 Acc@5 98.592 [2022-12-20 18:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 18:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 18:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 18:07:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 18:07:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][0/1519] eta 0:35:45 lr 0.000002 time 1.4126 (1.4126) model_time 0.9848 (0.9848) loss 0.8111 (0.8111) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 18:08:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][10/1519] eta 0:26:10 lr 0.000002 time 0.9248 (1.0410) model_time 0.9246 (1.0018) loss 0.7871 (0.8210) grad_norm 7.2013 (12.2341/6.0708) mem 68106MB [2022-12-20 18:08:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][20/1519] eta 0:25:36 lr 0.000002 time 0.9621 (1.0249) model_time 0.9620 (1.0042) loss 0.7818 (0.8523) grad_norm 7.2341 (10.8482/5.0745) mem 68106MB [2022-12-20 18:08:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][30/1519] eta 0:25:16 lr 0.000002 time 0.9467 (1.0186) model_time 0.9466 (1.0045) loss 0.6826 (0.8367) grad_norm 7.5120 (9.9970/4.3327) mem 68106MB [2022-12-20 18:08:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][40/1519] eta 0:24:59 lr 0.000002 time 0.9256 (1.0140) model_time 0.9255 (1.0032) loss 0.8660 (0.8451) grad_norm 9.2980 (9.9377/4.0817) mem 68106MB [2022-12-20 18:08:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][50/1519] eta 0:24:49 lr 0.000002 time 0.9291 (1.0137) model_time 0.9290 (1.0050) loss 0.6839 (0.8352) grad_norm 7.2832 (9.8055/3.6926) mem 68106MB [2022-12-20 18:08:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][60/1519] eta 0:24:37 lr 0.000002 time 0.9316 (1.0126) model_time 0.9315 (1.0053) loss 1.0757 (0.8343) grad_norm 10.1645 (9.5832/3.5085) mem 68106MB [2022-12-20 18:09:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][70/1519] eta 0:24:27 lr 0.000002 time 0.9390 (1.0128) model_time 0.9389 (1.0064) loss 0.7174 (0.8277) grad_norm 11.5033 (9.3510/3.3914) mem 68106MB [2022-12-20 18:09:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][80/1519] eta 0:24:16 lr 0.000002 time 0.9145 (1.0123) model_time 0.9144 (1.0066) loss 0.9076 (0.8205) grad_norm 7.8083 (9.1595/3.2719) mem 68106MB [2022-12-20 18:09:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][90/1519] eta 0:24:08 lr 0.000002 time 0.9527 (1.0134) model_time 0.9526 (1.0083) loss 0.7678 (0.8144) grad_norm 14.0202 (9.2543/3.1894) mem 68106MB [2022-12-20 18:09:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][100/1519] eta 0:23:58 lr 0.000002 time 0.9282 (1.0141) model_time 0.9280 (1.0095) loss 0.7466 (0.8152) grad_norm 8.9441 (9.1777/3.0418) mem 68106MB [2022-12-20 18:09:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][110/1519] eta 0:23:48 lr 0.000002 time 0.9588 (1.0135) model_time 0.9586 (1.0093) loss 0.9581 (0.8159) grad_norm 7.8705 (9.0818/2.9380) mem 68106MB [2022-12-20 18:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][120/1519] eta 0:23:40 lr 0.000002 time 0.9888 (1.0155) model_time 0.9886 (1.0116) loss 0.8552 (0.8146) grad_norm 9.5537 (8.9969/2.8765) mem 68106MB [2022-12-20 18:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][130/1519] eta 0:23:29 lr 0.000002 time 0.9204 (1.0149) model_time 0.9202 (1.0112) loss 0.6973 (0.8107) grad_norm 12.6032 (9.2366/3.4261) mem 68106MB [2022-12-20 18:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][140/1519] eta 0:23:18 lr 0.000002 time 0.9264 (1.0144) model_time 0.9262 (1.0110) loss 0.7520 (0.8036) grad_norm 11.2185 (9.2024/3.3353) mem 68106MB [2022-12-20 18:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][150/1519] eta 0:23:07 lr 0.000002 time 0.9713 (1.0135) model_time 0.9711 (1.0103) loss 0.8410 (0.8025) grad_norm 7.4571 (9.1211/3.2484) mem 68106MB [2022-12-20 18:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][160/1519] eta 0:22:56 lr 0.000002 time 0.9187 (1.0128) model_time 0.9186 (1.0097) loss 0.7565 (0.8000) grad_norm 10.5826 (9.0847/3.1901) mem 68106MB [2022-12-20 18:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][170/1519] eta 0:22:45 lr 0.000002 time 0.9348 (1.0121) model_time 0.9347 (1.0093) loss 0.7663 (0.7996) grad_norm 10.2374 (9.0720/3.1280) mem 68106MB [2022-12-20 18:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][180/1519] eta 0:22:34 lr 0.000002 time 0.9250 (1.0113) model_time 0.9249 (1.0086) loss 0.8223 (0.7994) grad_norm 7.9029 (8.9868/3.0658) mem 68106MB [2022-12-20 18:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][190/1519] eta 0:22:23 lr 0.000002 time 0.9280 (1.0106) model_time 0.9278 (1.0080) loss 0.6878 (0.7967) grad_norm 7.2198 (8.9599/2.9977) mem 68106MB [2022-12-20 18:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][200/1519] eta 0:22:12 lr 0.000002 time 0.9283 (1.0103) model_time 0.9282 (1.0078) loss 0.8165 (0.7954) grad_norm 6.4712 (8.9039/2.9392) mem 68106MB [2022-12-20 18:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][210/1519] eta 0:22:01 lr 0.000002 time 0.9269 (1.0099) model_time 0.9267 (1.0075) loss 0.6997 (0.7930) grad_norm 7.8682 (8.8830/2.9366) mem 68106MB [2022-12-20 18:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][220/1519] eta 0:21:51 lr 0.000002 time 0.9343 (1.0096) model_time 0.9341 (1.0073) loss 0.6810 (0.7927) grad_norm 7.4612 (8.8044/2.8958) mem 68106MB [2022-12-20 18:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][230/1519] eta 0:21:40 lr 0.000002 time 0.9323 (1.0093) model_time 0.9322 (1.0070) loss 0.6670 (0.7975) grad_norm 5.8398 (8.7434/2.8656) mem 68106MB [2022-12-20 18:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][240/1519] eta 0:21:30 lr 0.000002 time 0.9238 (1.0088) model_time 0.9236 (1.0067) loss 0.8866 (0.7963) grad_norm 7.4890 (8.7221/2.8282) mem 68106MB [2022-12-20 18:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][250/1519] eta 0:21:19 lr 0.000002 time 0.9328 (1.0085) model_time 0.9326 (1.0064) loss 0.7021 (0.7967) grad_norm 9.1671 (8.6824/2.7910) mem 68106MB [2022-12-20 18:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][260/1519] eta 0:21:09 lr 0.000002 time 0.9263 (1.0085) model_time 0.9262 (1.0065) loss 0.8601 (0.7959) grad_norm 7.7150 (8.7061/2.7578) mem 68106MB [2022-12-20 18:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][270/1519] eta 0:20:59 lr 0.000002 time 0.9296 (1.0082) model_time 0.9294 (1.0062) loss 0.6874 (0.7942) grad_norm 8.0502 (8.6745/2.7162) mem 68106MB [2022-12-20 18:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][280/1519] eta 0:20:48 lr 0.000002 time 0.9175 (1.0080) model_time 0.9174 (1.0061) loss 0.7638 (0.7937) grad_norm 6.5576 (8.6642/2.6871) mem 68106MB [2022-12-20 18:12:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][290/1519] eta 0:20:38 lr 0.000002 time 0.9294 (1.0077) model_time 0.9291 (1.0059) loss 0.6889 (0.7957) grad_norm 10.6797 (8.6698/2.6491) mem 68106MB [2022-12-20 18:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][300/1519] eta 0:20:28 lr 0.000002 time 0.9211 (1.0077) model_time 0.9209 (1.0059) loss 1.2151 (0.7978) grad_norm 9.2287 (8.6448/2.6136) mem 68106MB [2022-12-20 18:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][310/1519] eta 0:20:18 lr 0.000002 time 0.9364 (1.0075) model_time 0.9362 (1.0058) loss 0.7428 (0.7960) grad_norm 7.8543 (8.6147/2.6008) mem 68106MB [2022-12-20 18:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][320/1519] eta 0:20:07 lr 0.000002 time 0.9287 (1.0074) model_time 0.9285 (1.0057) loss 0.7805 (0.7963) grad_norm 9.7968 (8.6296/2.5875) mem 68106MB [2022-12-20 18:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][330/1519] eta 0:19:57 lr 0.000002 time 0.9297 (1.0071) model_time 0.9295 (1.0054) loss 1.0009 (0.7954) grad_norm 13.8600 (8.6434/2.5894) mem 68106MB [2022-12-20 18:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][340/1519] eta 0:19:47 lr 0.000002 time 0.9319 (1.0071) model_time 0.9317 (1.0054) loss 0.6838 (0.7971) grad_norm 7.7056 (8.6149/2.5575) mem 68106MB [2022-12-20 18:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][350/1519] eta 0:19:36 lr 0.000002 time 0.9375 (1.0068) model_time 0.9372 (1.0052) loss 0.9215 (0.7967) grad_norm 10.2402 (8.6455/2.5504) mem 68106MB [2022-12-20 18:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][360/1519] eta 0:19:26 lr 0.000002 time 0.9123 (1.0067) model_time 0.9121 (1.0051) loss 0.7020 (0.7967) grad_norm 8.3496 (8.7119/2.6349) mem 68106MB [2022-12-20 18:14:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][370/1519] eta 0:19:16 lr 0.000002 time 0.9318 (1.0066) model_time 0.9317 (1.0051) loss 0.6705 (0.7956) grad_norm 7.3586 (8.7296/2.6232) mem 68106MB [2022-12-20 18:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][380/1519] eta 0:19:06 lr 0.000002 time 0.9318 (1.0065) model_time 0.9317 (1.0050) loss 0.6858 (0.7943) grad_norm 8.2741 (8.7107/2.6010) mem 68106MB [2022-12-20 18:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][390/1519] eta 0:18:56 lr 0.000002 time 0.9168 (1.0067) model_time 0.9166 (1.0052) loss 0.6836 (0.7933) grad_norm 8.6274 (8.7083/2.5731) mem 68106MB [2022-12-20 18:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][400/1519] eta 0:18:46 lr 0.000002 time 0.9314 (1.0069) model_time 0.9312 (1.0054) loss 0.7054 (0.7938) grad_norm 8.3690 (8.7099/2.5636) mem 68106MB [2022-12-20 18:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][410/1519] eta 0:18:37 lr 0.000002 time 0.8826 (1.0076) model_time 0.8824 (1.0062) loss 0.6740 (0.7944) grad_norm 8.2964 (8.7252/2.5513) mem 68106MB [2022-12-20 18:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][420/1519] eta 0:18:27 lr 0.000002 time 0.9262 (1.0074) model_time 0.9261 (1.0060) loss 0.6848 (0.7949) grad_norm 12.6223 (8.7379/2.5403) mem 68106MB [2022-12-20 18:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][430/1519] eta 0:18:17 lr 0.000002 time 0.9129 (1.0075) model_time 0.9127 (1.0061) loss 0.7098 (0.7957) grad_norm 10.7773 (8.7376/2.5208) mem 68106MB [2022-12-20 18:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][440/1519] eta 0:18:07 lr 0.000002 time 0.9585 (1.0077) model_time 0.9582 (1.0063) loss 0.7041 (0.7946) grad_norm 6.4530 (8.7224/2.4991) mem 68106MB [2022-12-20 18:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][450/1519] eta 0:17:57 lr 0.000002 time 0.9303 (1.0077) model_time 0.9301 (1.0064) loss 0.7587 (0.7946) grad_norm 5.4572 (8.7007/2.4870) mem 68106MB [2022-12-20 18:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][460/1519] eta 0:17:47 lr 0.000002 time 0.9302 (1.0076) model_time 0.9300 (1.0063) loss 0.8925 (0.7950) grad_norm 7.9844 (8.6720/2.4677) mem 68106MB [2022-12-20 18:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][470/1519] eta 0:17:36 lr 0.000002 time 0.9304 (1.0074) model_time 0.9302 (1.0061) loss 0.9593 (0.7961) grad_norm 7.9356 (8.6976/2.4637) mem 68106MB [2022-12-20 18:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][480/1519] eta 0:17:26 lr 0.000002 time 0.9264 (1.0072) model_time 0.9263 (1.0059) loss 1.2159 (0.7968) grad_norm 8.4942 (8.6771/2.4471) mem 68106MB [2022-12-20 18:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][490/1519] eta 0:17:16 lr 0.000002 time 0.9380 (1.0071) model_time 0.9379 (1.0058) loss 0.7477 (0.7955) grad_norm 10.5782 (8.7103/2.4487) mem 68106MB [2022-12-20 18:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][500/1519] eta 0:17:06 lr 0.000002 time 0.9240 (1.0070) model_time 0.9238 (1.0057) loss 0.9353 (0.7949) grad_norm 8.3234 (8.6988/2.4347) mem 68106MB [2022-12-20 18:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][510/1519] eta 0:16:55 lr 0.000002 time 0.9298 (1.0068) model_time 0.9296 (1.0056) loss 0.8454 (0.7960) grad_norm 7.0457 (8.6858/2.4153) mem 68106MB [2022-12-20 18:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][520/1519] eta 0:16:45 lr 0.000002 time 0.9279 (1.0068) model_time 0.9277 (1.0056) loss 0.7528 (0.7948) grad_norm 9.8858 (8.6721/2.4022) mem 68106MB [2022-12-20 18:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][530/1519] eta 0:16:35 lr 0.000002 time 0.9418 (1.0067) model_time 0.9417 (1.0055) loss 0.6948 (0.7943) grad_norm 7.1492 (8.6633/2.3828) mem 68106MB [2022-12-20 18:16:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][540/1519] eta 0:16:25 lr 0.000002 time 0.9402 (1.0066) model_time 0.9401 (1.0054) loss 0.8190 (0.7948) grad_norm 9.1113 (8.6574/2.3656) mem 68106MB [2022-12-20 18:17:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][550/1519] eta 0:16:15 lr 0.000002 time 0.9275 (1.0065) model_time 0.9274 (1.0053) loss 0.7560 (0.7942) grad_norm 8.3089 (8.6587/2.3681) mem 68106MB [2022-12-20 18:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][560/1519] eta 0:16:05 lr 0.000002 time 0.9292 (1.0064) model_time 0.9290 (1.0053) loss 0.6878 (0.7933) grad_norm 7.5100 (8.6380/2.3537) mem 68106MB [2022-12-20 18:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][570/1519] eta 0:15:55 lr 0.000002 time 0.9299 (1.0064) model_time 0.9297 (1.0053) loss 0.6933 (0.7942) grad_norm 7.8825 (8.6393/2.3387) mem 68106MB [2022-12-20 18:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][580/1519] eta 0:15:44 lr 0.000002 time 0.9252 (1.0063) model_time 0.9248 (1.0052) loss 0.7954 (0.7937) grad_norm 8.3461 (8.6387/2.3206) mem 68106MB [2022-12-20 18:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][590/1519] eta 0:15:35 lr 0.000002 time 1.0073 (1.0065) model_time 1.0072 (1.0053) loss 0.8336 (0.7933) grad_norm 16.0871 (8.6408/2.3502) mem 68106MB [2022-12-20 18:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][600/1519] eta 0:15:24 lr 0.000002 time 0.9290 (1.0063) model_time 0.9288 (1.0052) loss 0.7492 (0.7932) grad_norm 10.7246 (8.6564/2.3425) mem 68106MB [2022-12-20 18:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][610/1519] eta 0:15:14 lr 0.000002 time 0.9291 (1.0064) model_time 0.9290 (1.0053) loss 0.9982 (0.7940) grad_norm 11.3437 (8.6190/2.1818) mem 68106MB [2022-12-20 18:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][620/1519] eta 0:15:04 lr 0.000002 time 0.9321 (1.0063) model_time 0.9319 (1.0052) loss 1.0110 (0.7939) grad_norm 8.0858 (8.6137/2.1477) mem 68106MB [2022-12-20 18:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][630/1519] eta 0:14:54 lr 0.000002 time 1.1955 (1.0066) model_time 1.1954 (1.0056) loss 1.0590 (0.7955) grad_norm 7.5515 (8.6366/2.1559) mem 68106MB [2022-12-20 18:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][640/1519] eta 0:14:44 lr 0.000002 time 0.9285 (1.0065) model_time 0.9283 (1.0055) loss 0.8697 (0.7970) grad_norm 8.1292 (8.6495/2.1474) mem 68106MB [2022-12-20 18:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][650/1519] eta 0:14:34 lr 0.000002 time 0.9303 (1.0064) model_time 0.9302 (1.0054) loss 0.8285 (0.7967) grad_norm 5.0140 (8.6090/2.1644) mem 68106MB [2022-12-20 18:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][660/1519] eta 0:14:24 lr 0.000002 time 0.9294 (1.0063) model_time 0.9292 (1.0053) loss 0.7581 (0.7966) grad_norm 11.1566 (8.6114/2.1615) mem 68106MB [2022-12-20 18:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][670/1519] eta 0:14:14 lr 0.000002 time 0.9280 (1.0062) model_time 0.9277 (1.0052) loss 0.7760 (0.7962) grad_norm 14.5797 (8.7084/2.2708) mem 68106MB [2022-12-20 18:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][680/1519] eta 0:14:04 lr 0.000002 time 1.0147 (1.0064) model_time 1.0146 (1.0053) loss 0.6845 (0.7965) grad_norm 9.3958 (8.7306/2.2806) mem 68106MB [2022-12-20 18:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][690/1519] eta 0:13:54 lr 0.000002 time 0.9336 (1.0062) model_time 0.9335 (1.0052) loss 0.7008 (0.7975) grad_norm 9.5507 (8.7071/2.2708) mem 68106MB [2022-12-20 18:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][700/1519] eta 0:13:44 lr 0.000002 time 0.9339 (1.0062) model_time 0.9337 (1.0052) loss 0.7998 (0.7971) grad_norm 8.2079 (8.6919/2.2820) mem 68106MB [2022-12-20 18:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][710/1519] eta 0:13:34 lr 0.000002 time 1.0386 (1.0063) model_time 1.0385 (1.0053) loss 0.7773 (0.7969) grad_norm 5.8967 (8.6826/2.2836) mem 68106MB [2022-12-20 18:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][720/1519] eta 0:13:24 lr 0.000002 time 0.9383 (1.0064) model_time 0.9382 (1.0054) loss 0.9745 (0.7969) grad_norm 10.7999 (8.6748/2.2845) mem 68106MB [2022-12-20 18:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][730/1519] eta 0:13:14 lr 0.000002 time 0.9386 (1.0064) model_time 0.9385 (1.0054) loss 0.6852 (0.7969) grad_norm 8.1510 (8.6262/2.0775) mem 68106MB [2022-12-20 18:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][740/1519] eta 0:13:03 lr 0.000002 time 0.9374 (1.0063) model_time 0.9373 (1.0053) loss 0.8498 (0.7976) grad_norm 7.1978 (8.6089/2.0697) mem 68106MB [2022-12-20 18:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][750/1519] eta 0:12:53 lr 0.000002 time 0.9294 (1.0064) model_time 0.9293 (1.0054) loss 0.7533 (0.7979) grad_norm 11.0496 (8.6103/2.0766) mem 68106MB [2022-12-20 18:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][760/1519] eta 0:12:43 lr 0.000002 time 0.9307 (1.0065) model_time 0.9305 (1.0056) loss 0.9895 (0.7984) grad_norm 5.9263 (8.6006/2.0698) mem 68106MB [2022-12-20 18:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][770/1519] eta 0:12:33 lr 0.000002 time 0.9345 (1.0065) model_time 0.9344 (1.0055) loss 0.7107 (0.7985) grad_norm 8.4677 (8.5987/2.0569) mem 68106MB [2022-12-20 18:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][780/1519] eta 0:12:23 lr 0.000002 time 0.9369 (1.0064) model_time 0.9368 (1.0055) loss 0.8018 (0.7979) grad_norm 8.1257 (8.6297/2.0654) mem 68106MB [2022-12-20 18:21:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][790/1519] eta 0:12:13 lr 0.000002 time 0.9325 (1.0063) model_time 0.9323 (1.0054) loss 0.7102 (0.7972) grad_norm 9.6348 (8.6357/2.0632) mem 68106MB [2022-12-20 18:21:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][800/1519] eta 0:12:03 lr 0.000002 time 0.9334 (1.0062) model_time 0.9333 (1.0053) loss 0.7841 (0.7977) grad_norm 10.5803 (8.6626/2.0752) mem 68106MB [2022-12-20 18:21:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][810/1519] eta 0:11:53 lr 0.000002 time 0.9821 (1.0062) model_time 0.9820 (1.0053) loss 0.9396 (0.7982) grad_norm 5.7514 (8.6774/2.1308) mem 68106MB [2022-12-20 18:21:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][820/1519] eta 0:11:43 lr 0.000002 time 0.9609 (1.0063) model_time 0.9608 (1.0053) loss 1.2424 (0.7988) grad_norm 6.6976 (8.6946/2.1296) mem 68106MB [2022-12-20 18:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][830/1519] eta 0:11:33 lr 0.000002 time 0.9293 (1.0062) model_time 0.9292 (1.0053) loss 0.6951 (0.7985) grad_norm 6.5344 (8.7186/2.1279) mem 68106MB [2022-12-20 18:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][840/1519] eta 0:11:23 lr 0.000002 time 0.9164 (1.0061) model_time 0.9162 (1.0052) loss 0.6805 (0.7987) grad_norm 6.7869 (8.7062/2.1229) mem 68106MB [2022-12-20 18:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][850/1519] eta 0:11:13 lr 0.000002 time 0.9188 (1.0060) model_time 0.9186 (1.0051) loss 0.9539 (0.7995) grad_norm 8.4983 (8.7114/2.1270) mem 68106MB [2022-12-20 18:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][860/1519] eta 0:11:02 lr 0.000002 time 0.9651 (1.0060) model_time 0.9648 (1.0051) loss 0.8018 (0.7997) grad_norm 8.9608 (8.7194/2.1280) mem 68106MB [2022-12-20 18:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][870/1519] eta 0:10:52 lr 0.000002 time 0.9278 (1.0059) model_time 0.9277 (1.0050) loss 0.8107 (0.8009) grad_norm 7.9973 (8.7321/2.1350) mem 68106MB [2022-12-20 18:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][880/1519] eta 0:10:42 lr 0.000002 time 0.9372 (1.0059) model_time 0.9370 (1.0050) loss 0.7043 (0.8006) grad_norm 8.4700 (8.7474/2.1561) mem 68106MB [2022-12-20 18:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][890/1519] eta 0:10:32 lr 0.000002 time 0.9298 (1.0058) model_time 0.9297 (1.0049) loss 0.7295 (0.8008) grad_norm 8.3651 (8.7232/2.1606) mem 68106MB [2022-12-20 18:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][900/1519] eta 0:10:22 lr 0.000002 time 0.9639 (1.0058) model_time 0.9637 (1.0049) loss 0.6976 (0.8003) grad_norm 8.5888 (8.7315/2.1568) mem 68106MB [2022-12-20 18:23:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][910/1519] eta 0:10:12 lr 0.000002 time 0.9348 (1.0059) model_time 0.9346 (1.0050) loss 0.6790 (0.8005) grad_norm 9.6764 (8.7601/2.1461) mem 68106MB [2022-12-20 18:23:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][920/1519] eta 0:10:02 lr 0.000002 time 0.9199 (1.0059) model_time 0.9197 (1.0051) loss 0.9353 (0.8006) grad_norm 9.6132 (8.7370/2.1442) mem 68106MB [2022-12-20 18:23:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][930/1519] eta 0:09:52 lr 0.000002 time 0.9313 (1.0059) model_time 0.9311 (1.0050) loss 0.7653 (0.7999) grad_norm 8.2781 (8.7276/2.1226) mem 68106MB [2022-12-20 18:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][940/1519] eta 0:09:42 lr 0.000002 time 0.9335 (1.0062) model_time 0.9334 (1.0054) loss 0.8101 (0.7995) grad_norm 8.2484 (8.7341/2.1260) mem 68106MB [2022-12-20 18:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][950/1519] eta 0:09:32 lr 0.000002 time 0.9325 (1.0061) model_time 0.9323 (1.0053) loss 0.8410 (0.8004) grad_norm 8.9649 (8.7175/2.1108) mem 68106MB [2022-12-20 18:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][960/1519] eta 0:09:22 lr 0.000002 time 0.9291 (1.0061) model_time 0.9290 (1.0052) loss 0.7825 (0.8002) grad_norm 8.3789 (8.6588/2.0324) mem 68106MB [2022-12-20 18:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][970/1519] eta 0:09:12 lr 0.000002 time 0.9274 (1.0060) model_time 0.9273 (1.0052) loss 0.7073 (0.8005) grad_norm 7.3807 (8.6309/2.0177) mem 68106MB [2022-12-20 18:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][980/1519] eta 0:09:02 lr 0.000002 time 0.9332 (1.0059) model_time 0.9331 (1.0051) loss 0.7351 (0.8006) grad_norm 8.5664 (8.6327/2.0176) mem 68106MB [2022-12-20 18:24:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][990/1519] eta 0:08:52 lr 0.000002 time 0.9240 (1.0062) model_time 0.9238 (1.0053) loss 0.6827 (0.7998) grad_norm 7.4414 (8.6319/2.0203) mem 68106MB [2022-12-20 18:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1000/1519] eta 0:08:42 lr 0.000002 time 0.9329 (1.0061) model_time 0.9327 (1.0053) loss 0.6697 (0.7996) grad_norm 9.7725 (8.6658/2.1010) mem 68106MB [2022-12-20 18:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1010/1519] eta 0:08:32 lr 0.000002 time 0.9321 (1.0061) model_time 0.9320 (1.0052) loss 0.6731 (0.7993) grad_norm 9.8291 (8.6661/2.0970) mem 68106MB [2022-12-20 18:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1020/1519] eta 0:08:22 lr 0.000002 time 0.9337 (1.0060) model_time 0.9335 (1.0052) loss 0.6893 (0.7996) grad_norm 13.9582 (8.6578/2.1094) mem 68106MB [2022-12-20 18:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1030/1519] eta 0:08:12 lr 0.000002 time 0.9923 (1.0062) model_time 0.9922 (1.0054) loss 0.8564 (0.7996) grad_norm 7.6506 (8.6362/2.1073) mem 68106MB [2022-12-20 18:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1040/1519] eta 0:08:01 lr 0.000002 time 0.9331 (1.0061) model_time 0.9330 (1.0053) loss 0.6871 (0.7990) grad_norm 10.1007 (8.6629/2.1069) mem 68106MB [2022-12-20 18:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1050/1519] eta 0:07:51 lr 0.000002 time 0.9328 (1.0060) model_time 0.9327 (1.0052) loss 0.9206 (0.7992) grad_norm 9.1114 (8.6700/2.1046) mem 68106MB [2022-12-20 18:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1060/1519] eta 0:07:41 lr 0.000002 time 0.9227 (1.0061) model_time 0.9225 (1.0053) loss 0.8517 (0.7989) grad_norm 5.8687 (8.6859/2.1091) mem 68106MB [2022-12-20 18:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1070/1519] eta 0:07:31 lr 0.000002 time 0.9243 (1.0061) model_time 0.9242 (1.0053) loss 0.7696 (0.7989) grad_norm 8.7532 (8.6721/2.0977) mem 68106MB [2022-12-20 18:26:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1080/1519] eta 0:07:21 lr 0.000002 time 0.9473 (1.0060) model_time 0.9471 (1.0052) loss 1.1232 (0.7990) grad_norm 9.2258 (8.6935/2.0936) mem 68106MB [2022-12-20 18:26:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1090/1519] eta 0:07:11 lr 0.000002 time 0.9290 (1.0060) model_time 0.9289 (1.0052) loss 0.6702 (0.7987) grad_norm 8.2239 (8.6543/2.0748) mem 68106MB [2022-12-20 18:26:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1100/1519] eta 0:07:01 lr 0.000002 time 0.9325 (1.0059) model_time 0.9324 (1.0051) loss 1.1270 (0.7990) grad_norm 8.8473 (8.6585/2.0716) mem 68106MB [2022-12-20 18:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1110/1519] eta 0:06:51 lr 0.000002 time 0.9242 (1.0059) model_time 0.9240 (1.0051) loss 0.7228 (0.7986) grad_norm 9.2007 (8.6562/2.0743) mem 68106MB [2022-12-20 18:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1120/1519] eta 0:06:41 lr 0.000002 time 0.9335 (1.0058) model_time 0.9333 (1.0050) loss 0.6733 (0.7984) grad_norm 5.7074 (8.6409/2.0823) mem 68106MB [2022-12-20 18:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1130/1519] eta 0:06:31 lr 0.000002 time 0.9329 (1.0058) model_time 0.9326 (1.0050) loss 0.8610 (0.7986) grad_norm 7.1510 (8.6916/2.1858) mem 68106MB [2022-12-20 18:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1140/1519] eta 0:06:21 lr 0.000002 time 0.9202 (1.0057) model_time 0.9200 (1.0050) loss 0.8604 (0.7991) grad_norm 7.3588 (8.6956/2.1905) mem 68106MB [2022-12-20 18:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1150/1519] eta 0:06:11 lr 0.000002 time 0.9296 (1.0057) model_time 0.9294 (1.0049) loss 1.0326 (0.7993) grad_norm 8.6147 (8.7044/2.1813) mem 68106MB [2022-12-20 18:27:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1160/1519] eta 0:06:01 lr 0.000002 time 0.9304 (1.0057) model_time 0.9303 (1.0049) loss 1.0214 (0.7996) grad_norm 7.7957 (8.7173/2.1887) mem 68106MB [2022-12-20 18:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1170/1519] eta 0:05:50 lr 0.000002 time 0.9466 (1.0057) model_time 0.9463 (1.0049) loss 0.7116 (0.7992) grad_norm 7.8168 (8.7066/2.1865) mem 68106MB [2022-12-20 18:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1180/1519] eta 0:05:40 lr 0.000002 time 0.9398 (1.0057) model_time 0.9395 (1.0049) loss 0.8503 (0.7989) grad_norm 7.8069 (8.6977/2.1880) mem 68106MB [2022-12-20 18:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1190/1519] eta 0:05:30 lr 0.000002 time 0.9915 (1.0059) model_time 0.9913 (1.0052) loss 0.6679 (0.7989) grad_norm 7.3430 (8.6609/2.1551) mem 68106MB [2022-12-20 18:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1200/1519] eta 0:05:20 lr 0.000002 time 0.9254 (1.0059) model_time 0.9250 (1.0052) loss 0.7254 (0.7991) grad_norm 5.7762 (8.6185/2.1639) mem 68106MB [2022-12-20 18:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1210/1519] eta 0:05:10 lr 0.000002 time 1.0045 (1.0061) model_time 1.0043 (1.0054) loss 0.9299 (0.7988) grad_norm 8.9883 (8.5906/2.1445) mem 68106MB [2022-12-20 18:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1220/1519] eta 0:05:00 lr 0.000002 time 0.9256 (1.0064) model_time 0.9255 (1.0056) loss 0.7142 (0.7986) grad_norm 8.9092 (8.5905/2.1403) mem 68106MB [2022-12-20 18:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1230/1519] eta 0:04:50 lr 0.000002 time 1.0073 (1.0064) model_time 1.0072 (1.0057) loss 0.8161 (0.7985) grad_norm 10.5517 (8.5953/2.1462) mem 68106MB [2022-12-20 18:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1240/1519] eta 0:04:40 lr 0.000002 time 0.9414 (1.0066) model_time 0.9413 (1.0059) loss 0.6731 (0.7983) grad_norm 7.3897 (8.5412/2.1161) mem 68106MB [2022-12-20 18:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1250/1519] eta 0:04:30 lr 0.000002 time 0.9191 (1.0067) model_time 0.9181 (1.0060) loss 0.7800 (0.7981) grad_norm 7.0870 (8.5616/2.1004) mem 68106MB [2022-12-20 18:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1260/1519] eta 0:04:20 lr 0.000002 time 0.9267 (1.0066) model_time 0.9265 (1.0059) loss 0.8039 (0.7981) grad_norm 9.0849 (8.5629/2.0906) mem 68106MB [2022-12-20 18:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1270/1519] eta 0:04:10 lr 0.000002 time 0.9241 (1.0066) model_time 0.9240 (1.0058) loss 0.7073 (0.7981) grad_norm 16.8602 (8.5343/2.0358) mem 68106MB [2022-12-20 18:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1280/1519] eta 0:04:00 lr 0.000002 time 0.9243 (1.0065) model_time 0.9242 (1.0058) loss 1.1866 (0.7992) grad_norm 9.3261 (8.5304/2.0097) mem 68106MB [2022-12-20 18:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1290/1519] eta 0:03:50 lr 0.000002 time 0.9215 (1.0064) model_time 0.9214 (1.0057) loss 0.6900 (0.7989) grad_norm 9.7661 (8.5283/2.0004) mem 68106MB [2022-12-20 18:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1300/1519] eta 0:03:40 lr 0.000002 time 0.9277 (1.0064) model_time 0.9276 (1.0057) loss 0.6636 (0.7988) grad_norm 10.4293 (8.5764/2.0728) mem 68106MB [2022-12-20 18:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1310/1519] eta 0:03:30 lr 0.000002 time 0.9223 (1.0064) model_time 0.9222 (1.0056) loss 0.6561 (0.7989) grad_norm 9.4311 (8.5876/2.0689) mem 68106MB [2022-12-20 18:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1320/1519] eta 0:03:20 lr 0.000002 time 0.9243 (1.0063) model_time 0.9242 (1.0056) loss 0.7753 (0.7989) grad_norm 5.9707 (8.6006/2.0702) mem 68106MB [2022-12-20 18:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1330/1519] eta 0:03:10 lr 0.000002 time 0.9228 (1.0063) model_time 0.9227 (1.0055) loss 0.8722 (0.7992) grad_norm 10.3567 (8.6049/2.0817) mem 68106MB [2022-12-20 18:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1340/1519] eta 0:03:00 lr 0.000002 time 0.9229 (1.0062) model_time 0.9227 (1.0055) loss 0.8954 (0.7991) grad_norm 7.8373 (8.6198/2.0840) mem 68106MB [2022-12-20 18:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1350/1519] eta 0:02:50 lr 0.000002 time 0.9212 (1.0062) model_time 0.9209 (1.0055) loss 0.8222 (0.7994) grad_norm 5.7019 (8.6137/2.0808) mem 68106MB [2022-12-20 18:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1360/1519] eta 0:02:39 lr 0.000002 time 0.9209 (1.0061) model_time 0.9207 (1.0054) loss 0.8190 (0.8000) grad_norm 10.6713 (8.6251/2.0788) mem 68106MB [2022-12-20 18:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1370/1519] eta 0:02:29 lr 0.000002 time 1.0035 (1.0061) model_time 1.0033 (1.0054) loss 0.7645 (0.8000) grad_norm 8.9575 (8.6464/2.0947) mem 68106MB [2022-12-20 18:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1380/1519] eta 0:02:19 lr 0.000002 time 0.9253 (1.0060) model_time 0.9252 (1.0053) loss 0.8140 (0.8003) grad_norm 8.9148 (8.6201/2.0835) mem 68106MB [2022-12-20 18:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1390/1519] eta 0:02:09 lr 0.000002 time 0.9268 (1.0060) model_time 0.9266 (1.0053) loss 0.7758 (0.8004) grad_norm 23.1190 (8.6603/2.2494) mem 68106MB [2022-12-20 18:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1400/1519] eta 0:01:59 lr 0.000002 time 0.9256 (1.0060) model_time 0.9254 (1.0053) loss 0.9254 (0.8003) grad_norm 6.2218 (8.6497/2.2478) mem 68106MB [2022-12-20 18:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1410/1519] eta 0:01:49 lr 0.000002 time 1.0130 (1.0060) model_time 1.0128 (1.0053) loss 0.6802 (0.8000) grad_norm 7.4469 (8.6244/2.1667) mem 68106MB [2022-12-20 18:31:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1420/1519] eta 0:01:39 lr 0.000002 time 0.9321 (1.0059) model_time 0.9319 (1.0052) loss 0.7343 (0.7998) grad_norm 7.0377 (8.6184/2.1646) mem 68106MB [2022-12-20 18:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1430/1519] eta 0:01:29 lr 0.000002 time 0.9249 (1.0059) model_time 0.9247 (1.0052) loss 0.8850 (0.8004) grad_norm 11.1679 (8.6335/2.1713) mem 68106MB [2022-12-20 18:32:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1440/1519] eta 0:01:19 lr 0.000002 time 0.9282 (1.0059) model_time 0.9280 (1.0052) loss 0.8547 (0.8008) grad_norm 8.8763 (8.6501/2.1728) mem 68106MB [2022-12-20 18:32:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1450/1519] eta 0:01:09 lr 0.000002 time 0.9716 (1.0058) model_time 0.9714 (1.0051) loss 0.6700 (0.8011) grad_norm 9.9208 (8.6810/2.1803) mem 68106MB [2022-12-20 18:32:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1460/1519] eta 0:00:59 lr 0.000002 time 0.9320 (1.0058) model_time 0.9318 (1.0051) loss 0.6883 (0.8005) grad_norm 6.1634 (8.6531/2.1791) mem 68106MB [2022-12-20 18:32:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1470/1519] eta 0:00:49 lr 0.000002 time 0.9271 (1.0058) model_time 0.9269 (1.0051) loss 0.6764 (0.8002) grad_norm 10.7844 (8.6759/2.1822) mem 68106MB [2022-12-20 18:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1480/1519] eta 0:00:39 lr 0.000002 time 0.9065 (1.0060) model_time 0.9064 (1.0053) loss 0.7832 (0.8000) grad_norm 6.0970 (8.6503/2.1564) mem 68106MB [2022-12-20 18:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1490/1519] eta 0:00:29 lr 0.000002 time 0.9331 (1.0060) model_time 0.9328 (1.0053) loss 0.7520 (0.8002) grad_norm 16.6626 (8.6921/2.2062) mem 68106MB [2022-12-20 18:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1500/1519] eta 0:00:19 lr 0.000002 time 0.9183 (1.0060) model_time 0.9181 (1.0053) loss 0.6742 (0.8001) grad_norm 7.9412 (8.6782/2.2108) mem 68106MB [2022-12-20 18:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [86/100][1510/1519] eta 0:00:09 lr 0.000002 time 0.9099 (1.0059) model_time 0.9098 (1.0053) loss 0.6735 (0.8004) grad_norm 9.0787 (8.6708/2.2057) mem 68106MB [2022-12-20 18:33:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 86 training takes 0:25:28 [2022-12-20 18:33:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_86.pth saving...... [2022-12-20 18:33:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_86.pth saved !!! [2022-12-20 18:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.661 (0.661) Loss 0.5396 (0.5396) Acc@1 92.361 (92.361) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 18:33:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.332) Loss 0.5322 (0.5096) Acc@1 92.708 (92.677) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 18:33:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.316) Loss 0.4848 (0.5039) Acc@1 91.319 (92.692) Acc@5 98.958 (98.446) Mem 68106MB [2022-12-20 18:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.314) Loss 0.6382 (0.5115) Acc@1 90.625 (92.484) Acc@5 98.264 (98.432) Mem 68106MB [2022-12-20 18:34:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.311) Loss 0.4596 (0.5024) Acc@1 93.750 (92.556) Acc@5 99.306 (98.543) Mem 68106MB [2022-12-20 18:34:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.308) Loss 0.4886 (0.4997) Acc@1 92.708 (92.627) Acc@5 99.653 (98.604) Mem 68106MB [2022-12-20 18:34:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.307) Loss 0.5117 (0.4994) Acc@1 90.972 (92.560) Acc@5 98.264 (98.577) Mem 68106MB [2022-12-20 18:34:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.306) Loss 0.5479 (0.5006) Acc@1 92.361 (92.513) Acc@5 97.917 (98.572) Mem 68106MB [2022-12-20 18:34:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.305) Loss 0.4288 (0.4991) Acc@1 93.403 (92.541) Acc@5 98.264 (98.598) Mem 68106MB [2022-12-20 18:34:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:86] * Acc@1 92.518 Acc@5 98.600 [2022-12-20 18:34:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 18:34:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 18:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][0/1519] eta 0:50:15 lr 0.000002 time 1.9855 (1.9855) model_time 1.3634 (1.3634) loss 1.1611 (1.1611) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 18:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][10/1519] eta 0:27:32 lr 0.000002 time 0.9248 (1.0954) model_time 0.9247 (1.0384) loss 0.6918 (0.8538) grad_norm 8.1710 (9.2774/0.8283) mem 68106MB [2022-12-20 18:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][20/1519] eta 0:26:23 lr 0.000002 time 0.9915 (1.0565) model_time 0.9914 (1.0265) loss 0.7074 (0.8181) grad_norm 7.5966 (8.0733/1.3797) mem 68106MB [2022-12-20 18:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][30/1519] eta 0:25:47 lr 0.000002 time 0.9165 (1.0396) model_time 0.9163 (1.0192) loss 0.9177 (0.8097) grad_norm 8.7571 (8.3230/1.5382) mem 68106MB [2022-12-20 18:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][40/1519] eta 0:25:24 lr 0.000002 time 0.9336 (1.0310) model_time 0.9334 (1.0154) loss 0.6680 (0.8016) grad_norm 6.6453 (8.0463/1.5470) mem 68106MB [2022-12-20 18:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][50/1519] eta 0:25:08 lr 0.000002 time 0.9311 (1.0269) model_time 0.9309 (1.0143) loss 0.9608 (0.8123) grad_norm 8.2410 (8.3860/1.7593) mem 68106MB [2022-12-20 18:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][60/1519] eta 0:24:54 lr 0.000002 time 0.9317 (1.0247) model_time 0.9315 (1.0140) loss 0.7824 (0.8027) grad_norm 10.7785 (8.4352/1.6886) mem 68106MB [2022-12-20 18:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][70/1519] eta 0:24:40 lr 0.000002 time 0.9348 (1.0219) model_time 0.9347 (1.0127) loss 0.6814 (0.8012) grad_norm 7.1539 (8.3298/1.6616) mem 68106MB [2022-12-20 18:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][80/1519] eta 0:24:26 lr 0.000002 time 0.9214 (1.0190) model_time 0.9212 (1.0109) loss 0.6857 (0.7958) grad_norm 6.5293 (8.3118/1.6054) mem 68106MB [2022-12-20 18:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][90/1519] eta 0:24:13 lr 0.000002 time 0.9291 (1.0169) model_time 0.9290 (1.0097) loss 0.6999 (0.7983) grad_norm 8.1350 (8.4936/1.7160) mem 68106MB [2022-12-20 18:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][100/1519] eta 0:24:00 lr 0.000002 time 0.9329 (1.0150) model_time 0.9328 (1.0085) loss 1.1100 (0.8092) grad_norm 8.0367 (8.4724/1.6458) mem 68106MB [2022-12-20 18:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][110/1519] eta 0:23:48 lr 0.000002 time 0.9431 (1.0137) model_time 0.9429 (1.0077) loss 0.9891 (0.8170) grad_norm 8.4900 (8.5293/1.6372) mem 68106MB [2022-12-20 18:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][120/1519] eta 0:23:36 lr 0.000002 time 0.9306 (1.0124) model_time 0.9304 (1.0069) loss 0.7133 (0.8125) grad_norm 12.5819 (8.7280/2.0292) mem 68106MB [2022-12-20 18:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][130/1519] eta 0:23:24 lr 0.000002 time 0.9304 (1.0114) model_time 0.9303 (1.0062) loss 0.7536 (0.8100) grad_norm 8.9992 (8.8057/2.0010) mem 68106MB [2022-12-20 18:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][140/1519] eta 0:23:14 lr 0.000002 time 0.9249 (1.0111) model_time 0.9247 (1.0063) loss 0.8251 (0.8058) grad_norm 11.1256 (8.7863/1.9754) mem 68106MB [2022-12-20 18:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][150/1519] eta 0:23:03 lr 0.000002 time 0.9382 (1.0104) model_time 0.9380 (1.0059) loss 0.7165 (0.8036) grad_norm 7.5341 (8.7317/1.9530) mem 68106MB [2022-12-20 18:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][160/1519] eta 0:22:52 lr 0.000002 time 0.9225 (1.0100) model_time 0.9224 (1.0057) loss 0.7582 (0.8016) grad_norm 15.2416 (8.8549/2.0951) mem 68106MB [2022-12-20 18:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][170/1519] eta 0:22:43 lr 0.000002 time 0.9230 (1.0107) model_time 0.9229 (1.0067) loss 0.6609 (0.7978) grad_norm 7.6238 (8.7740/2.0712) mem 68106MB [2022-12-20 18:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][180/1519] eta 0:22:32 lr 0.000002 time 0.9327 (1.0101) model_time 0.9325 (1.0063) loss 0.8904 (0.7974) grad_norm 9.2931 (8.7503/2.0453) mem 68106MB [2022-12-20 18:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][190/1519] eta 0:22:22 lr 0.000002 time 0.9289 (1.0101) model_time 0.9288 (1.0065) loss 0.6789 (0.7955) grad_norm 8.1003 (8.6864/2.0163) mem 68106MB [2022-12-20 18:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][200/1519] eta 0:22:12 lr 0.000002 time 0.9192 (1.0105) model_time 0.9191 (1.0070) loss 0.6648 (0.7996) grad_norm 10.5451 (8.6624/2.0070) mem 68106MB [2022-12-20 18:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][210/1519] eta 0:22:02 lr 0.000002 time 0.9250 (1.0105) model_time 0.9248 (1.0072) loss 0.7981 (0.7979) grad_norm 11.9363 (8.7313/2.0015) mem 68106MB [2022-12-20 18:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][220/1519] eta 0:21:52 lr 0.000002 time 0.9244 (1.0102) model_time 0.9241 (1.0070) loss 0.6654 (0.7959) grad_norm 10.9713 (8.7696/2.0257) mem 68106MB [2022-12-20 18:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][230/1519] eta 0:21:42 lr 0.000002 time 0.9312 (1.0104) model_time 0.9310 (1.0073) loss 0.7354 (0.7940) grad_norm 8.6092 (8.7201/2.0301) mem 68106MB [2022-12-20 18:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][240/1519] eta 0:21:31 lr 0.000002 time 0.9440 (1.0101) model_time 0.9439 (1.0071) loss 0.7862 (0.7942) grad_norm 6.1504 (8.6848/2.0333) mem 68106MB [2022-12-20 18:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][250/1519] eta 0:21:21 lr 0.000002 time 0.9343 (1.0097) model_time 0.9341 (1.0068) loss 0.6638 (0.7935) grad_norm 5.8275 (8.6040/2.0342) mem 68106MB [2022-12-20 18:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][260/1519] eta 0:21:10 lr 0.000002 time 0.9256 (1.0093) model_time 0.9252 (1.0065) loss 0.7769 (0.7976) grad_norm 7.0081 (8.6302/2.0316) mem 68106MB [2022-12-20 18:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][270/1519] eta 0:21:00 lr 0.000002 time 0.9270 (1.0089) model_time 0.9268 (1.0062) loss 0.8540 (0.7975) grad_norm 7.5599 (8.6441/2.0382) mem 68106MB [2022-12-20 18:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][280/1519] eta 0:20:50 lr 0.000002 time 0.9307 (1.0092) model_time 0.9306 (1.0066) loss 0.7123 (0.7982) grad_norm 6.6932 (8.6166/2.0163) mem 68106MB [2022-12-20 18:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][290/1519] eta 0:20:39 lr 0.000002 time 0.9210 (1.0088) model_time 0.9208 (1.0063) loss 0.9191 (0.7991) grad_norm 7.5438 (8.6141/1.9961) mem 68106MB [2022-12-20 18:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][300/1519] eta 0:20:29 lr 0.000002 time 0.9830 (1.0089) model_time 0.9829 (1.0064) loss 0.7241 (0.7983) grad_norm 10.3528 (8.6379/1.9792) mem 68106MB [2022-12-20 18:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][310/1519] eta 0:20:19 lr 0.000002 time 0.9299 (1.0090) model_time 0.9297 (1.0066) loss 0.6836 (0.7973) grad_norm 5.6600 (8.6384/1.9789) mem 68106MB [2022-12-20 18:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][320/1519] eta 0:20:09 lr 0.000002 time 0.9337 (1.0087) model_time 0.9335 (1.0064) loss 1.0498 (0.7983) grad_norm 10.8553 (8.6350/1.9665) mem 68106MB [2022-12-20 18:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][330/1519] eta 0:19:59 lr 0.000002 time 0.9259 (1.0086) model_time 0.9258 (1.0063) loss 0.7118 (0.7967) grad_norm 11.6019 (8.6635/1.9796) mem 68106MB [2022-12-20 18:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][340/1519] eta 0:19:49 lr 0.000002 time 0.9406 (1.0086) model_time 0.9404 (1.0064) loss 0.6885 (0.7961) grad_norm 7.9273 (8.6551/1.9634) mem 68106MB [2022-12-20 18:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][350/1519] eta 0:19:39 lr 0.000002 time 0.9386 (1.0090) model_time 0.9385 (1.0069) loss 0.9020 (0.7982) grad_norm 6.5041 (8.6331/1.9534) mem 68106MB [2022-12-20 18:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][360/1519] eta 0:19:29 lr 0.000002 time 0.9195 (1.0088) model_time 0.9194 (1.0067) loss 0.6728 (0.7992) grad_norm 9.5848 (8.7351/2.1232) mem 68106MB [2022-12-20 18:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][370/1519] eta 0:19:19 lr 0.000002 time 0.9370 (1.0090) model_time 0.9368 (1.0069) loss 0.7412 (0.7980) grad_norm 11.7765 (8.7402/2.1201) mem 68106MB [2022-12-20 18:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][380/1519] eta 0:19:09 lr 0.000002 time 0.9312 (1.0089) model_time 0.9310 (1.0069) loss 0.7063 (0.7981) grad_norm 10.4372 (8.7361/2.1161) mem 68106MB [2022-12-20 18:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][390/1519] eta 0:18:58 lr 0.000002 time 0.9244 (1.0087) model_time 0.9242 (1.0067) loss 0.6638 (0.7959) grad_norm 7.4808 (8.7333/2.0970) mem 68106MB [2022-12-20 18:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][400/1519] eta 0:18:48 lr 0.000002 time 0.9348 (1.0085) model_time 0.9346 (1.0066) loss 0.6689 (0.7950) grad_norm 5.9625 (8.7055/2.0934) mem 68106MB [2022-12-20 18:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][410/1519] eta 0:18:38 lr 0.000002 time 0.9573 (1.0084) model_time 0.9572 (1.0065) loss 0.6772 (0.7959) grad_norm 7.7481 (8.7094/2.0887) mem 68106MB [2022-12-20 18:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][420/1519] eta 0:18:27 lr 0.000002 time 0.9264 (1.0081) model_time 0.9263 (1.0062) loss 0.7087 (0.7957) grad_norm 9.4840 (8.7151/2.0765) mem 68106MB [2022-12-20 18:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][430/1519] eta 0:18:17 lr 0.000002 time 0.9222 (1.0079) model_time 0.9221 (1.0061) loss 0.6608 (0.7944) grad_norm 6.2899 (8.6833/2.0648) mem 68106MB [2022-12-20 18:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][440/1519] eta 0:18:07 lr 0.000002 time 0.9244 (1.0077) model_time 0.9243 (1.0059) loss 0.7552 (0.7940) grad_norm 10.9263 (8.6655/2.0575) mem 68106MB [2022-12-20 18:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][450/1519] eta 0:17:57 lr 0.000002 time 0.9301 (1.0077) model_time 0.9299 (1.0059) loss 0.6750 (0.7933) grad_norm 7.6459 (8.6549/2.0420) mem 68106MB [2022-12-20 18:41:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][460/1519] eta 0:17:47 lr 0.000002 time 0.9692 (1.0084) model_time 0.9691 (1.0067) loss 0.8753 (0.7934) grad_norm 8.7863 (8.6503/2.0219) mem 68106MB [2022-12-20 18:42:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][470/1519] eta 0:17:37 lr 0.000002 time 0.9483 (1.0082) model_time 0.9481 (1.0065) loss 0.7141 (0.7922) grad_norm 7.4830 (8.6529/2.0029) mem 68106MB [2022-12-20 18:42:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][480/1519] eta 0:17:27 lr 0.000002 time 0.9913 (1.0085) model_time 0.9911 (1.0069) loss 0.7395 (0.7931) grad_norm 8.4479 (8.6624/1.9857) mem 68106MB [2022-12-20 18:42:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][490/1519] eta 0:17:17 lr 0.000002 time 0.9284 (1.0085) model_time 0.9282 (1.0069) loss 0.8923 (0.7923) grad_norm 8.0181 (8.6450/1.9714) mem 68106MB [2022-12-20 18:42:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][500/1519] eta 0:17:07 lr 0.000002 time 0.9286 (1.0084) model_time 0.9284 (1.0068) loss 0.7856 (0.7916) grad_norm 11.5607 (8.6789/1.9789) mem 68106MB [2022-12-20 18:42:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][510/1519] eta 0:16:57 lr 0.000002 time 0.9330 (1.0085) model_time 0.9328 (1.0069) loss 0.7481 (0.7916) grad_norm 6.0106 (8.6658/1.9760) mem 68106MB [2022-12-20 18:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][520/1519] eta 0:16:47 lr 0.000002 time 0.9217 (1.0084) model_time 0.9215 (1.0068) loss 0.8763 (0.7926) grad_norm 8.5202 (8.6695/1.9682) mem 68106MB [2022-12-20 18:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][530/1519] eta 0:16:37 lr 0.000002 time 0.9278 (1.0086) model_time 0.9276 (1.0071) loss 0.9016 (0.7928) grad_norm 12.7769 (8.6693/1.9748) mem 68106MB [2022-12-20 18:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][540/1519] eta 0:16:27 lr 0.000002 time 0.9313 (1.0086) model_time 0.9311 (1.0071) loss 0.9461 (0.7945) grad_norm 12.3082 (8.6863/1.9745) mem 68106MB [2022-12-20 18:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][550/1519] eta 0:16:17 lr 0.000002 time 0.9315 (1.0085) model_time 0.9313 (1.0070) loss 0.6589 (0.7934) grad_norm 4.8144 (8.6608/1.9841) mem 68106MB [2022-12-20 18:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][560/1519] eta 0:16:06 lr 0.000002 time 0.9234 (1.0083) model_time 0.9232 (1.0068) loss 0.6868 (0.7926) grad_norm 7.4370 (8.6385/1.9761) mem 68106MB [2022-12-20 18:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][570/1519] eta 0:15:56 lr 0.000002 time 0.9278 (1.0081) model_time 0.9277 (1.0067) loss 0.6989 (0.7924) grad_norm 10.0345 (8.6442/1.9931) mem 68106MB [2022-12-20 18:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][580/1519] eta 0:15:46 lr 0.000002 time 0.9231 (1.0080) model_time 0.9229 (1.0066) loss 0.7664 (0.7941) grad_norm 6.2228 (8.6247/1.9836) mem 68106MB [2022-12-20 18:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][590/1519] eta 0:15:36 lr 0.000002 time 0.9337 (1.0079) model_time 0.9335 (1.0065) loss 0.9572 (0.7941) grad_norm 9.6635 (8.6483/1.9835) mem 68106MB [2022-12-20 18:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][600/1519] eta 0:15:26 lr 0.000002 time 0.9263 (1.0077) model_time 0.9261 (1.0063) loss 0.8309 (0.7930) grad_norm 6.6150 (8.6436/1.9872) mem 68106MB [2022-12-20 18:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][610/1519] eta 0:15:15 lr 0.000002 time 0.9329 (1.0076) model_time 0.9327 (1.0062) loss 0.7147 (0.7933) grad_norm 17.7677 (8.6723/2.0625) mem 68106MB [2022-12-20 18:44:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][620/1519] eta 0:15:05 lr 0.000002 time 0.9388 (1.0076) model_time 0.9386 (1.0062) loss 0.6691 (0.7932) grad_norm 7.0247 (8.6757/2.0602) mem 68106MB [2022-12-20 18:44:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][630/1519] eta 0:14:55 lr 0.000002 time 0.9337 (1.0075) model_time 0.9335 (1.0062) loss 0.7857 (0.7929) grad_norm 11.8868 (8.6966/2.0885) mem 68106MB [2022-12-20 18:44:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][640/1519] eta 0:14:45 lr 0.000002 time 0.9585 (1.0074) model_time 0.9584 (1.0061) loss 0.6939 (0.7938) grad_norm 6.8530 (8.7069/2.0805) mem 68106MB [2022-12-20 18:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][650/1519] eta 0:14:35 lr 0.000002 time 0.9340 (1.0074) model_time 0.9339 (1.0061) loss 0.8485 (0.7948) grad_norm 7.2104 (8.6777/2.0655) mem 68106MB [2022-12-20 18:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][660/1519] eta 0:14:25 lr 0.000002 time 0.8954 (1.0075) model_time 0.8953 (1.0062) loss 0.7765 (0.7951) grad_norm 8.8465 (8.6629/2.0657) mem 68106MB [2022-12-20 18:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][670/1519] eta 0:14:15 lr 0.000002 time 0.9330 (1.0074) model_time 0.9328 (1.0061) loss 1.0465 (0.7951) grad_norm 9.8947 (8.6896/2.0625) mem 68106MB [2022-12-20 18:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][680/1519] eta 0:14:05 lr 0.000002 time 1.0129 (1.0076) model_time 1.0128 (1.0063) loss 0.8665 (0.7951) grad_norm 9.8350 (8.6988/2.0745) mem 68106MB [2022-12-20 18:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][690/1519] eta 0:13:55 lr 0.000002 time 0.9312 (1.0075) model_time 0.9311 (1.0062) loss 0.6921 (0.7953) grad_norm 6.6706 (8.6708/2.0858) mem 68106MB [2022-12-20 18:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][700/1519] eta 0:13:45 lr 0.000002 time 0.9298 (1.0074) model_time 0.9296 (1.0062) loss 0.7604 (0.7955) grad_norm 10.3829 (8.6594/2.0970) mem 68106MB [2022-12-20 18:46:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][710/1519] eta 0:13:35 lr 0.000002 time 0.9357 (1.0074) model_time 0.9355 (1.0062) loss 0.7824 (0.7956) grad_norm 10.6005 (8.6472/2.1010) mem 68106MB [2022-12-20 18:46:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][720/1519] eta 0:13:24 lr 0.000002 time 0.9303 (1.0074) model_time 0.9301 (1.0062) loss 1.0669 (0.7959) grad_norm 12.0749 (8.6420/2.0614) mem 68106MB [2022-12-20 18:46:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][730/1519] eta 0:13:14 lr 0.000002 time 0.9314 (1.0073) model_time 0.9312 (1.0061) loss 0.6695 (0.7952) grad_norm 7.4127 (8.6262/2.0534) mem 68106MB [2022-12-20 18:46:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][740/1519] eta 0:13:04 lr 0.000002 time 0.9363 (1.0072) model_time 0.9362 (1.0060) loss 0.9810 (0.7961) grad_norm 7.5649 (8.6101/2.0542) mem 68106MB [2022-12-20 18:46:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][750/1519] eta 0:12:54 lr 0.000002 time 0.9285 (1.0072) model_time 0.9284 (1.0060) loss 0.8761 (0.7972) grad_norm 10.3629 (8.6238/2.0572) mem 68106MB [2022-12-20 18:47:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][760/1519] eta 0:12:44 lr 0.000002 time 0.9209 (1.0071) model_time 0.9207 (1.0059) loss 0.7523 (0.7973) grad_norm 6.6794 (8.5818/2.0048) mem 68106MB [2022-12-20 18:47:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][770/1519] eta 0:12:34 lr 0.000002 time 0.9058 (1.0071) model_time 0.9056 (1.0060) loss 0.6876 (0.7970) grad_norm 8.6372 (8.5986/2.0051) mem 68106MB [2022-12-20 18:47:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][780/1519] eta 0:12:24 lr 0.000002 time 0.9451 (1.0070) model_time 0.9450 (1.0059) loss 0.7482 (0.7970) grad_norm 9.0591 (8.6116/2.0040) mem 68106MB [2022-12-20 18:47:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][790/1519] eta 0:12:14 lr 0.000002 time 0.9220 (1.0070) model_time 0.9218 (1.0058) loss 0.9042 (0.7964) grad_norm 7.4767 (8.6034/2.0091) mem 68106MB [2022-12-20 18:47:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][800/1519] eta 0:12:04 lr 0.000002 time 0.9265 (1.0071) model_time 0.9264 (1.0059) loss 0.7672 (0.7965) grad_norm 7.4167 (8.5989/1.9991) mem 68106MB [2022-12-20 18:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][810/1519] eta 0:11:54 lr 0.000002 time 0.9550 (1.0072) model_time 0.9548 (1.0061) loss 0.8065 (0.7971) grad_norm 10.1193 (8.5934/2.0004) mem 68106MB [2022-12-20 18:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][820/1519] eta 0:11:44 lr 0.000002 time 0.9935 (1.0072) model_time 0.9934 (1.0061) loss 0.6827 (0.7972) grad_norm 6.3418 (8.5671/1.9814) mem 68106MB [2022-12-20 18:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][830/1519] eta 0:11:33 lr 0.000002 time 0.9184 (1.0071) model_time 0.9183 (1.0060) loss 0.7061 (0.7974) grad_norm 8.3884 (8.5781/1.9708) mem 68106MB [2022-12-20 18:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][840/1519] eta 0:11:23 lr 0.000002 time 0.9261 (1.0071) model_time 0.9259 (1.0060) loss 1.1597 (0.7982) grad_norm 7.3488 (8.5718/1.9609) mem 68106MB [2022-12-20 18:48:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][850/1519] eta 0:11:13 lr 0.000002 time 0.9205 (1.0071) model_time 0.9204 (1.0060) loss 0.6954 (0.7973) grad_norm 9.3520 (8.5890/1.9548) mem 68106MB [2022-12-20 18:48:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][860/1519] eta 0:11:03 lr 0.000002 time 1.0067 (1.0072) model_time 1.0065 (1.0061) loss 0.6947 (0.7967) grad_norm 6.8893 (8.5619/1.9441) mem 68106MB [2022-12-20 18:48:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][870/1519] eta 0:10:53 lr 0.000002 time 0.9262 (1.0071) model_time 0.9260 (1.0060) loss 0.6604 (0.7972) grad_norm 7.0268 (8.5505/1.9277) mem 68106MB [2022-12-20 18:49:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][880/1519] eta 0:10:43 lr 0.000002 time 0.9177 (1.0069) model_time 0.9176 (1.0059) loss 1.0426 (0.7970) grad_norm 7.6783 (8.5631/1.9251) mem 68106MB [2022-12-20 18:49:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][890/1519] eta 0:10:33 lr 0.000002 time 0.9232 (1.0068) model_time 0.9230 (1.0058) loss 0.6680 (0.7964) grad_norm 8.2770 (8.5646/1.9219) mem 68106MB [2022-12-20 18:49:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][900/1519] eta 0:10:23 lr 0.000002 time 0.9236 (1.0068) model_time 0.9234 (1.0058) loss 1.0236 (0.7967) grad_norm 6.5856 (8.5288/1.9224) mem 68106MB [2022-12-20 18:49:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][910/1519] eta 0:10:13 lr 0.000002 time 0.9316 (1.0068) model_time 0.9315 (1.0057) loss 0.6670 (0.7964) grad_norm 7.4540 (8.5511/1.9441) mem 68106MB [2022-12-20 18:49:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][920/1519] eta 0:10:03 lr 0.000002 time 0.9432 (1.0067) model_time 0.9430 (1.0057) loss 1.2125 (0.7974) grad_norm 8.2662 (8.5476/1.9350) mem 68106MB [2022-12-20 18:49:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][930/1519] eta 0:09:52 lr 0.000002 time 0.9248 (1.0067) model_time 0.9247 (1.0056) loss 0.7080 (0.7978) grad_norm 9.3979 (8.5069/1.9268) mem 68106MB [2022-12-20 18:50:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][940/1519] eta 0:09:42 lr 0.000002 time 0.9185 (1.0066) model_time 0.9183 (1.0056) loss 0.6951 (0.7974) grad_norm 7.0552 (8.5123/1.9557) mem 68106MB [2022-12-20 18:50:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][950/1519] eta 0:09:32 lr 0.000002 time 0.9222 (1.0065) model_time 0.9221 (1.0055) loss 0.7630 (0.7973) grad_norm 8.9673 (8.5361/1.9632) mem 68106MB [2022-12-20 18:50:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][960/1519] eta 0:09:22 lr 0.000002 time 0.9226 (1.0066) model_time 0.9224 (1.0056) loss 0.8643 (0.7981) grad_norm 5.2520 (8.4652/1.8462) mem 68106MB [2022-12-20 18:50:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][970/1519] eta 0:09:12 lr 0.000002 time 0.9177 (1.0066) model_time 0.9175 (1.0056) loss 0.7764 (0.7988) grad_norm 5.2386 (8.4461/1.8381) mem 68106MB [2022-12-20 18:50:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][980/1519] eta 0:09:02 lr 0.000002 time 0.9335 (1.0067) model_time 0.9333 (1.0057) loss 0.8298 (0.7985) grad_norm 12.3853 (8.4357/1.8488) mem 68106MB [2022-12-20 18:50:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][990/1519] eta 0:08:52 lr 0.000002 time 0.9338 (1.0066) model_time 0.9337 (1.0056) loss 0.7124 (0.7980) grad_norm 16.3113 (8.4541/1.9001) mem 68106MB [2022-12-20 18:51:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1000/1519] eta 0:08:42 lr 0.000002 time 0.9852 (1.0066) model_time 0.9850 (1.0056) loss 0.9149 (0.7978) grad_norm 7.4869 (8.4557/1.8939) mem 68106MB [2022-12-20 18:51:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1010/1519] eta 0:08:32 lr 0.000002 time 0.9075 (1.0066) model_time 0.9073 (1.0057) loss 0.7164 (0.7978) grad_norm 6.9199 (8.4327/1.8825) mem 68106MB [2022-12-20 18:51:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1020/1519] eta 0:08:22 lr 0.000002 time 0.9166 (1.0065) model_time 0.9165 (1.0056) loss 0.7278 (0.7981) grad_norm 6.9133 (8.4343/1.8998) mem 68106MB [2022-12-20 18:51:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1030/1519] eta 0:08:12 lr 0.000002 time 0.9310 (1.0065) model_time 0.9309 (1.0055) loss 0.8999 (0.7986) grad_norm 7.5258 (8.4334/1.9000) mem 68106MB [2022-12-20 18:51:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1040/1519] eta 0:08:02 lr 0.000002 time 0.9521 (1.0065) model_time 0.9519 (1.0056) loss 0.6947 (0.7989) grad_norm 10.4294 (8.4518/1.8975) mem 68106MB [2022-12-20 18:51:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1050/1519] eta 0:07:52 lr 0.000002 time 0.9296 (1.0065) model_time 0.9289 (1.0055) loss 0.7324 (0.7994) grad_norm 12.9586 (8.4831/1.9247) mem 68106MB [2022-12-20 18:52:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1060/1519] eta 0:07:41 lr 0.000002 time 0.9259 (1.0064) model_time 0.9255 (1.0055) loss 0.6889 (0.7991) grad_norm 6.7337 (8.4871/1.9310) mem 68106MB [2022-12-20 18:52:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1070/1519] eta 0:07:31 lr 0.000002 time 0.9780 (1.0064) model_time 0.9779 (1.0055) loss 1.0119 (0.7990) grad_norm 9.4071 (8.4784/1.9319) mem 68106MB [2022-12-20 18:52:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1080/1519] eta 0:07:21 lr 0.000002 time 0.9123 (1.0066) model_time 0.9122 (1.0057) loss 0.8777 (0.7992) grad_norm 7.2900 (8.4654/1.9350) mem 68106MB [2022-12-20 18:52:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1090/1519] eta 0:07:11 lr 0.000002 time 0.9218 (1.0066) model_time 0.9212 (1.0057) loss 1.0209 (0.7997) grad_norm 7.9271 (8.4745/1.9446) mem 68106MB [2022-12-20 18:52:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1100/1519] eta 0:07:01 lr 0.000002 time 0.9313 (1.0066) model_time 0.9311 (1.0057) loss 0.6607 (0.7996) grad_norm 7.5460 (8.4715/1.9966) mem 68106MB [2022-12-20 18:52:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1110/1519] eta 0:06:51 lr 0.000002 time 0.9278 (1.0066) model_time 0.9277 (1.0057) loss 0.8209 (0.7993) grad_norm 8.4274 (8.4725/1.9910) mem 68106MB [2022-12-20 18:53:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1120/1519] eta 0:06:41 lr 0.000002 time 0.9314 (1.0067) model_time 0.9313 (1.0058) loss 0.7025 (0.7994) grad_norm 8.0133 (8.4616/1.9814) mem 68106MB [2022-12-20 18:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1130/1519] eta 0:06:31 lr 0.000002 time 0.9260 (1.0067) model_time 0.9259 (1.0058) loss 0.6578 (0.7996) grad_norm 9.0882 (8.4455/1.9659) mem 68106MB [2022-12-20 18:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1140/1519] eta 0:06:21 lr 0.000002 time 0.9231 (1.0068) model_time 0.9230 (1.0059) loss 0.8255 (0.7999) grad_norm 7.8021 (8.4317/1.9522) mem 68106MB [2022-12-20 18:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1150/1519] eta 0:06:11 lr 0.000002 time 1.0767 (1.0069) model_time 1.0765 (1.0060) loss 0.8440 (0.7999) grad_norm 7.0456 (8.4272/1.9396) mem 68106MB [2022-12-20 18:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1160/1519] eta 0:06:01 lr 0.000002 time 1.2352 (1.0071) model_time 1.2350 (1.0062) loss 0.7184 (0.7995) grad_norm 6.5825 (8.4271/1.9393) mem 68106MB [2022-12-20 18:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1170/1519] eta 0:05:51 lr 0.000002 time 0.9199 (1.0071) model_time 0.9197 (1.0062) loss 0.7185 (0.7999) grad_norm 11.5645 (8.4398/1.9492) mem 68106MB [2022-12-20 18:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1180/1519] eta 0:05:41 lr 0.000002 time 0.9367 (1.0071) model_time 0.9366 (1.0062) loss 0.7432 (0.7999) grad_norm 8.1295 (8.4855/2.0196) mem 68106MB [2022-12-20 18:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1190/1519] eta 0:05:31 lr 0.000002 time 0.9213 (1.0070) model_time 0.9212 (1.0061) loss 0.6656 (0.7997) grad_norm 7.5595 (8.4597/2.0052) mem 68106MB [2022-12-20 18:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1200/1519] eta 0:05:21 lr 0.000002 time 0.9329 (1.0070) model_time 0.9327 (1.0061) loss 0.9551 (0.7999) grad_norm 6.3815 (8.4454/1.9984) mem 68106MB [2022-12-20 18:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1210/1519] eta 0:05:11 lr 0.000002 time 0.9262 (1.0069) model_time 0.9260 (1.0060) loss 1.1073 (0.8004) grad_norm 9.1868 (8.4055/1.9214) mem 68106MB [2022-12-20 18:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1220/1519] eta 0:05:01 lr 0.000002 time 0.9374 (1.0069) model_time 0.9373 (1.0060) loss 0.7805 (0.8003) grad_norm 6.6550 (8.4105/1.9240) mem 68106MB [2022-12-20 18:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1230/1519] eta 0:04:50 lr 0.000002 time 0.9230 (1.0068) model_time 0.9229 (1.0059) loss 0.9169 (0.8006) grad_norm 10.9271 (8.4117/1.9310) mem 68106MB [2022-12-20 18:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1240/1519] eta 0:04:40 lr 0.000002 time 0.9373 (1.0068) model_time 0.9372 (1.0059) loss 0.7833 (0.8002) grad_norm 9.2974 (8.4080/1.9339) mem 68106MB [2022-12-20 18:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1250/1519] eta 0:04:30 lr 0.000002 time 0.9983 (1.0068) model_time 0.9982 (1.0059) loss 0.6612 (0.8004) grad_norm 7.6532 (8.4103/1.9388) mem 68106MB [2022-12-20 18:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1260/1519] eta 0:04:20 lr 0.000002 time 0.9247 (1.0067) model_time 0.9246 (1.0059) loss 0.7146 (0.8004) grad_norm 6.3949 (8.4690/2.0208) mem 68106MB [2022-12-20 18:55:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1270/1519] eta 0:04:10 lr 0.000002 time 0.9209 (1.0067) model_time 0.9207 (1.0058) loss 0.7314 (0.8005) grad_norm 7.2194 (8.4649/2.0209) mem 68106MB [2022-12-20 18:55:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1280/1519] eta 0:04:00 lr 0.000002 time 0.9280 (1.0066) model_time 0.9279 (1.0058) loss 0.6927 (0.8004) grad_norm 12.0417 (8.4933/2.1051) mem 68106MB [2022-12-20 18:55:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1290/1519] eta 0:03:50 lr 0.000002 time 0.9343 (1.0066) model_time 0.9342 (1.0058) loss 0.7081 (0.8003) grad_norm 11.5534 (8.5117/2.0896) mem 68106MB [2022-12-20 18:56:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1300/1519] eta 0:03:40 lr 0.000002 time 0.9228 (1.0066) model_time 0.9227 (1.0057) loss 0.7427 (0.8003) grad_norm 8.2523 (8.5141/2.0803) mem 68106MB [2022-12-20 18:56:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1310/1519] eta 0:03:30 lr 0.000002 time 0.9206 (1.0067) model_time 0.9205 (1.0059) loss 0.7919 (0.8004) grad_norm 10.2484 (8.5385/2.0843) mem 68106MB [2022-12-20 18:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1320/1519] eta 0:03:20 lr 0.000002 time 0.9214 (1.0067) model_time 0.9212 (1.0059) loss 0.6800 (0.8003) grad_norm 12.0725 (8.5155/2.0561) mem 68106MB [2022-12-20 18:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1330/1519] eta 0:03:10 lr 0.000002 time 0.9213 (1.0067) model_time 0.9212 (1.0059) loss 0.9749 (0.8008) grad_norm 7.8633 (8.5161/2.0542) mem 68106MB [2022-12-20 18:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1340/1519] eta 0:03:00 lr 0.000002 time 1.0108 (1.0068) model_time 1.0106 (1.0059) loss 0.7349 (0.8004) grad_norm 8.1868 (8.5429/2.0581) mem 68106MB [2022-12-20 18:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1350/1519] eta 0:02:50 lr 0.000002 time 0.9218 (1.0067) model_time 0.9216 (1.0059) loss 1.2039 (0.8004) grad_norm 6.6502 (8.5253/2.0495) mem 68106MB [2022-12-20 18:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1360/1519] eta 0:02:40 lr 0.000002 time 0.9300 (1.0066) model_time 0.9298 (1.0058) loss 0.7742 (0.8001) grad_norm 8.8173 (8.5222/2.0528) mem 68106MB [2022-12-20 18:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1370/1519] eta 0:02:29 lr 0.000002 time 0.9236 (1.0066) model_time 0.9235 (1.0058) loss 0.9909 (0.8004) grad_norm 9.1084 (8.5342/2.0613) mem 68106MB [2022-12-20 18:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1380/1519] eta 0:02:19 lr 0.000002 time 0.9316 (1.0065) model_time 0.9315 (1.0057) loss 1.0416 (0.8005) grad_norm 9.0296 (8.5030/2.0634) mem 68106MB [2022-12-20 18:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1390/1519] eta 0:02:09 lr 0.000002 time 0.9799 (1.0065) model_time 0.9798 (1.0057) loss 0.6826 (0.8005) grad_norm 8.8235 (8.5374/2.0692) mem 68106MB [2022-12-20 18:57:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1400/1519] eta 0:01:59 lr 0.000002 time 0.9220 (1.0065) model_time 0.9219 (1.0057) loss 0.8366 (0.8007) grad_norm 7.2392 (8.5387/2.0714) mem 68106MB [2022-12-20 18:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1410/1519] eta 0:01:49 lr 0.000002 time 0.9221 (1.0064) model_time 0.9219 (1.0056) loss 0.7627 (0.8007) grad_norm 6.3101 (8.5043/2.0640) mem 68106MB [2022-12-20 18:58:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1420/1519] eta 0:01:39 lr 0.000002 time 0.9333 (1.0065) model_time 0.9331 (1.0057) loss 1.0053 (0.8004) grad_norm 6.1421 (8.5231/2.0728) mem 68106MB [2022-12-20 18:58:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1430/1519] eta 0:01:29 lr 0.000002 time 0.9240 (1.0066) model_time 0.9238 (1.0058) loss 0.7877 (0.8006) grad_norm 7.8796 (8.5214/2.0797) mem 68106MB [2022-12-20 18:58:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1440/1519] eta 0:01:19 lr 0.000002 time 0.9266 (1.0065) model_time 0.9264 (1.0058) loss 0.6728 (0.8003) grad_norm 7.5279 (8.5215/2.0798) mem 68106MB [2022-12-20 18:58:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1450/1519] eta 0:01:09 lr 0.000002 time 0.9303 (1.0066) model_time 0.9302 (1.0058) loss 0.8264 (0.8003) grad_norm 6.8294 (8.5273/2.0767) mem 68106MB [2022-12-20 18:58:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1460/1519] eta 0:00:59 lr 0.000002 time 0.9281 (1.0065) model_time 0.9280 (1.0058) loss 0.8133 (0.8002) grad_norm 6.5181 (8.5444/2.0809) mem 68106MB [2022-12-20 18:58:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1470/1519] eta 0:00:49 lr 0.000002 time 0.9211 (1.0066) model_time 0.9210 (1.0058) loss 0.9960 (0.8004) grad_norm 10.0325 (8.5432/2.0809) mem 68106MB [2022-12-20 18:59:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1480/1519] eta 0:00:39 lr 0.000002 time 0.9276 (1.0065) model_time 0.9274 (1.0058) loss 0.8088 (0.8002) grad_norm 7.1973 (8.5532/2.0906) mem 68106MB [2022-12-20 18:59:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1490/1519] eta 0:00:29 lr 0.000002 time 0.9335 (1.0066) model_time 0.9334 (1.0058) loss 0.6942 (0.8003) grad_norm 7.7761 (8.5380/2.0903) mem 68106MB [2022-12-20 18:59:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1500/1519] eta 0:00:19 lr 0.000002 time 0.9235 (1.0065) model_time 0.9233 (1.0057) loss 0.7017 (0.8000) grad_norm 11.1840 (8.5798/2.1193) mem 68106MB [2022-12-20 18:59:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [87/100][1510/1519] eta 0:00:09 lr 0.000002 time 0.9354 (1.0065) model_time 0.9353 (1.0057) loss 0.6607 (0.7996) grad_norm 10.3859 (8.5865/2.1040) mem 68106MB [2022-12-20 18:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 87 training takes 0:25:28 [2022-12-20 18:59:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_87.pth saving...... [2022-12-20 19:00:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_87.pth saved !!! [2022-12-20 19:00:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.624 (0.624) Loss 0.5391 (0.5391) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 19:00:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.327) Loss 0.5367 (0.5099) Acc@1 92.361 (92.614) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 19:00:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.312) Loss 0.4853 (0.5042) Acc@1 91.667 (92.692) Acc@5 99.306 (98.429) Mem 68106MB [2022-12-20 19:00:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.306 (0.307) Loss 0.6367 (0.5114) Acc@1 91.319 (92.484) Acc@5 97.917 (98.421) Mem 68106MB [2022-12-20 19:00:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.301 (0.306) Loss 0.4572 (0.5021) Acc@1 93.750 (92.573) Acc@5 99.306 (98.552) Mem 68106MB [2022-12-20 19:00:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.305) Loss 0.4914 (0.4996) Acc@1 92.014 (92.633) Acc@5 99.653 (98.604) Mem 68106MB [2022-12-20 19:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.295 (0.304) Loss 0.5104 (0.4992) Acc@1 90.972 (92.566) Acc@5 98.611 (98.588) Mem 68106MB [2022-12-20 19:00:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.5443 (0.5005) Acc@1 92.708 (92.518) Acc@5 98.264 (98.582) Mem 68106MB [2022-12-20 19:00:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4347 (0.4992) Acc@1 93.056 (92.541) Acc@5 98.264 (98.603) Mem 68106MB [2022-12-20 19:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:87] * Acc@1 92.514 Acc@5 98.604 [2022-12-20 19:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 19:00:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 19:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][0/1519] eta 0:47:04 lr 0.000002 time 1.8597 (1.8597) model_time 1.1173 (1.1173) loss 0.9295 (0.9295) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 19:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][10/1519] eta 0:27:05 lr 0.000002 time 0.9303 (1.0772) model_time 0.9300 (1.0093) loss 0.6892 (0.7790) grad_norm 10.3859 (8.8577/1.4466) mem 68106MB [2022-12-20 19:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][20/1519] eta 0:25:58 lr 0.000002 time 0.9260 (1.0400) model_time 0.9258 (1.0042) loss 0.7862 (0.7815) grad_norm 9.5747 (8.3902/1.5825) mem 68106MB [2022-12-20 19:01:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][30/1519] eta 0:25:28 lr 0.000002 time 0.9348 (1.0268) model_time 0.9347 (1.0025) loss 0.7351 (0.7613) grad_norm 7.5663 (8.0431/1.4928) mem 68106MB [2022-12-20 19:01:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][40/1519] eta 0:25:08 lr 0.000002 time 0.9350 (1.0202) model_time 0.9348 (1.0017) loss 0.9373 (0.7777) grad_norm 7.3079 (8.1606/1.5657) mem 68106MB [2022-12-20 19:01:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][50/1519] eta 0:24:57 lr 0.000002 time 0.9313 (1.0192) model_time 0.9311 (1.0042) loss 0.7521 (0.7786) grad_norm 8.6992 (8.4417/1.7547) mem 68106MB [2022-12-20 19:01:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][60/1519] eta 0:24:42 lr 0.000002 time 0.9305 (1.0160) model_time 0.9303 (1.0034) loss 0.7417 (0.7893) grad_norm 8.1726 (8.6383/1.7413) mem 68106MB [2022-12-20 19:01:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][70/1519] eta 0:24:30 lr 0.000002 time 0.9515 (1.0148) model_time 0.9514 (1.0039) loss 0.8264 (0.7855) grad_norm 6.6463 (8.6281/1.8412) mem 68106MB [2022-12-20 19:01:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][80/1519] eta 0:24:17 lr 0.000002 time 0.9299 (1.0131) model_time 0.9297 (1.0036) loss 1.1254 (0.7835) grad_norm 8.0044 (8.5542/1.7748) mem 68106MB [2022-12-20 19:02:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][90/1519] eta 0:24:06 lr 0.000001 time 0.9276 (1.0122) model_time 0.9275 (1.0037) loss 0.6768 (0.7844) grad_norm 13.5376 (8.7209/1.8724) mem 68106MB [2022-12-20 19:02:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][100/1519] eta 0:23:55 lr 0.000001 time 0.9328 (1.0119) model_time 0.9327 (1.0042) loss 0.8541 (0.7882) grad_norm 5.4034 (8.5887/1.8915) mem 68106MB [2022-12-20 19:02:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][110/1519] eta 0:23:45 lr 0.000001 time 0.9308 (1.0114) model_time 0.9306 (1.0043) loss 1.0178 (0.7898) grad_norm 7.5915 (8.4646/1.8514) mem 68106MB [2022-12-20 19:02:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][120/1519] eta 0:23:33 lr 0.000001 time 0.9641 (1.0105) model_time 0.9640 (1.0039) loss 0.7033 (0.7899) grad_norm 7.0363 (8.4268/1.8119) mem 68106MB [2022-12-20 19:02:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][130/1519] eta 0:23:22 lr 0.000001 time 0.9314 (1.0098) model_time 0.9313 (1.0037) loss 0.8241 (0.7910) grad_norm 9.7921 (8.4280/1.7812) mem 68106MB [2022-12-20 19:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][140/1519] eta 0:23:14 lr 0.000001 time 0.9219 (1.0115) model_time 0.9216 (1.0058) loss 1.1604 (0.7973) grad_norm 6.4268 (8.3838/1.7376) mem 68106MB [2022-12-20 19:03:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][150/1519] eta 0:23:03 lr 0.000001 time 0.9269 (1.0105) model_time 0.9267 (1.0052) loss 0.7653 (0.8018) grad_norm 9.2716 (8.4091/1.7642) mem 68106MB [2022-12-20 19:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][160/1519] eta 0:22:53 lr 0.000001 time 0.9312 (1.0109) model_time 0.9311 (1.0060) loss 0.7088 (0.7968) grad_norm 6.0272 (8.3843/1.7954) mem 68106MB [2022-12-20 19:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][170/1519] eta 0:22:42 lr 0.000001 time 0.9318 (1.0103) model_time 0.9316 (1.0055) loss 1.0731 (0.7980) grad_norm 7.6697 (8.3897/1.7756) mem 68106MB [2022-12-20 19:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][180/1519] eta 0:22:32 lr 0.000001 time 0.9186 (1.0098) model_time 0.9184 (1.0053) loss 0.7252 (0.8009) grad_norm 7.5355 (8.3718/1.7527) mem 68106MB [2022-12-20 19:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][190/1519] eta 0:22:21 lr 0.000001 time 0.9283 (1.0094) model_time 0.9281 (1.0051) loss 0.6553 (0.7999) grad_norm 11.6753 (8.3956/1.7531) mem 68106MB [2022-12-20 19:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][200/1519] eta 0:22:10 lr 0.000001 time 0.9292 (1.0088) model_time 0.9291 (1.0047) loss 0.7025 (0.7989) grad_norm 7.4245 (8.4189/1.7424) mem 68106MB [2022-12-20 19:04:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][210/1519] eta 0:22:00 lr 0.000001 time 0.9362 (1.0086) model_time 0.9361 (1.0047) loss 0.7912 (0.8020) grad_norm 13.2306 (8.4646/1.7806) mem 68106MB [2022-12-20 19:04:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][220/1519] eta 0:21:51 lr 0.000001 time 0.9183 (1.0093) model_time 0.9180 (1.0056) loss 0.8300 (0.8051) grad_norm 8.0110 (8.4348/1.7786) mem 68106MB [2022-12-20 19:04:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][230/1519] eta 0:21:40 lr 0.000001 time 0.9453 (1.0093) model_time 0.9452 (1.0057) loss 0.9666 (0.8043) grad_norm 11.8656 (8.4700/1.7838) mem 68106MB [2022-12-20 19:04:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][240/1519] eta 0:21:31 lr 0.000001 time 0.9229 (1.0099) model_time 0.9227 (1.0064) loss 0.6700 (0.8025) grad_norm 9.6132 (8.5048/1.7793) mem 68106MB [2022-12-20 19:04:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][250/1519] eta 0:21:21 lr 0.000001 time 0.9297 (1.0098) model_time 0.9296 (1.0064) loss 0.7540 (0.8015) grad_norm 10.7457 (8.5428/1.8128) mem 68106MB [2022-12-20 19:04:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][260/1519] eta 0:21:11 lr 0.000001 time 0.9343 (1.0096) model_time 0.9341 (1.0063) loss 0.7144 (0.8006) grad_norm 8.2858 (8.5365/1.7882) mem 68106MB [2022-12-20 19:05:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][270/1519] eta 0:21:00 lr 0.000001 time 0.9249 (1.0093) model_time 0.9248 (1.0062) loss 0.7784 (0.7997) grad_norm 12.6368 (8.5671/1.8211) mem 68106MB [2022-12-20 19:05:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][280/1519] eta 0:20:50 lr 0.000001 time 0.9385 (1.0092) model_time 0.9383 (1.0062) loss 0.7788 (0.8013) grad_norm 6.3263 (8.5264/1.8185) mem 68106MB [2022-12-20 19:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][290/1519] eta 0:20:40 lr 0.000001 time 0.9311 (1.0091) model_time 0.9309 (1.0061) loss 0.8724 (0.8027) grad_norm 9.1846 (8.5345/1.7931) mem 68106MB [2022-12-20 19:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][300/1519] eta 0:20:29 lr 0.000001 time 0.9212 (1.0087) model_time 0.9210 (1.0058) loss 0.7728 (0.8002) grad_norm 7.4212 (8.4731/1.7986) mem 68106MB [2022-12-20 19:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][310/1519] eta 0:20:19 lr 0.000001 time 0.9410 (1.0084) model_time 0.9409 (1.0056) loss 0.9687 (0.7996) grad_norm 9.5851 (8.5154/1.8199) mem 68106MB [2022-12-20 19:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][320/1519] eta 0:20:08 lr 0.000001 time 0.9193 (1.0081) model_time 0.9191 (1.0054) loss 0.6724 (0.7994) grad_norm 7.1096 (8.4941/1.8322) mem 68106MB [2022-12-20 19:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][330/1519] eta 0:19:58 lr 0.000001 time 0.9246 (1.0079) model_time 0.9244 (1.0053) loss 0.6582 (0.7997) grad_norm 8.4676 (8.4884/1.8069) mem 68106MB [2022-12-20 19:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][340/1519] eta 0:19:48 lr 0.000001 time 0.9308 (1.0077) model_time 0.9301 (1.0051) loss 0.7757 (0.8018) grad_norm 7.3009 (8.5011/1.7882) mem 68106MB [2022-12-20 19:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][350/1519] eta 0:19:37 lr 0.000001 time 0.9302 (1.0075) model_time 0.9301 (1.0050) loss 0.7756 (0.8004) grad_norm 13.1061 (8.5055/1.8065) mem 68106MB [2022-12-20 19:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][360/1519] eta 0:19:28 lr 0.000001 time 0.9322 (1.0078) model_time 0.9320 (1.0053) loss 0.6825 (0.7987) grad_norm 9.5537 (8.5107/1.8184) mem 68106MB [2022-12-20 19:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][370/1519] eta 0:19:17 lr 0.000001 time 0.9302 (1.0077) model_time 0.9300 (1.0053) loss 0.9665 (0.7995) grad_norm 8.0917 (8.5320/1.8244) mem 68106MB [2022-12-20 19:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][380/1519] eta 0:19:07 lr 0.000001 time 0.8891 (1.0077) model_time 0.8889 (1.0054) loss 0.7958 (0.8000) grad_norm 8.0148 (8.5179/1.8055) mem 68106MB [2022-12-20 19:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][390/1519] eta 0:18:57 lr 0.000001 time 0.9302 (1.0077) model_time 0.9300 (1.0054) loss 1.1604 (0.8025) grad_norm 14.2034 (8.5323/1.8346) mem 68106MB [2022-12-20 19:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][400/1519] eta 0:18:47 lr 0.000001 time 0.9594 (1.0076) model_time 0.9593 (1.0053) loss 0.6748 (0.8022) grad_norm 11.2602 (8.5385/1.8297) mem 68106MB [2022-12-20 19:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][410/1519] eta 0:18:37 lr 0.000001 time 0.9589 (1.0080) model_time 0.9587 (1.0058) loss 0.7123 (0.8007) grad_norm 7.4757 (8.5435/1.8161) mem 68106MB [2022-12-20 19:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][420/1519] eta 0:18:27 lr 0.000001 time 0.9161 (1.0079) model_time 0.9159 (1.0057) loss 0.7346 (0.8000) grad_norm 12.1307 (8.5391/1.8224) mem 68106MB [2022-12-20 19:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][430/1519] eta 0:18:17 lr 0.000001 time 0.9285 (1.0079) model_time 0.9282 (1.0057) loss 0.9483 (0.8017) grad_norm 5.6469 (8.5506/1.8295) mem 68106MB [2022-12-20 19:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][440/1519] eta 0:18:07 lr 0.000001 time 0.9157 (1.0079) model_time 0.9155 (1.0059) loss 0.6955 (0.8000) grad_norm 8.7889 (8.5720/1.8402) mem 68106MB [2022-12-20 19:08:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][450/1519] eta 0:17:57 lr 0.000001 time 0.9312 (1.0080) model_time 0.9310 (1.0060) loss 0.7135 (0.7999) grad_norm 10.1014 (8.5703/1.8253) mem 68106MB [2022-12-20 19:08:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][460/1519] eta 0:17:47 lr 0.000001 time 0.9254 (1.0077) model_time 0.9253 (1.0057) loss 0.7117 (0.7997) grad_norm 6.9893 (8.5476/1.8171) mem 68106MB [2022-12-20 19:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][470/1519] eta 0:17:37 lr 0.000001 time 0.9326 (1.0078) model_time 0.9324 (1.0058) loss 0.7229 (0.7994) grad_norm 7.7731 (8.5566/1.8092) mem 68106MB [2022-12-20 19:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][480/1519] eta 0:17:27 lr 0.000001 time 0.9229 (1.0078) model_time 0.9227 (1.0059) loss 0.6826 (0.8010) grad_norm 7.9673 (8.5601/1.8205) mem 68106MB [2022-12-20 19:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][490/1519] eta 0:17:16 lr 0.000001 time 0.9240 (1.0077) model_time 0.9238 (1.0058) loss 0.6785 (0.7993) grad_norm 8.2990 (8.5520/1.8046) mem 68106MB [2022-12-20 19:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][500/1519] eta 0:17:06 lr 0.000001 time 0.9352 (1.0075) model_time 0.9350 (1.0057) loss 0.6823 (0.7984) grad_norm 10.2996 (8.5514/1.8038) mem 68106MB [2022-12-20 19:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][510/1519] eta 0:16:56 lr 0.000001 time 0.9335 (1.0074) model_time 0.9333 (1.0056) loss 0.8417 (0.7984) grad_norm 8.5359 (8.5269/1.7980) mem 68106MB [2022-12-20 19:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][520/1519] eta 0:16:46 lr 0.000001 time 0.8927 (1.0073) model_time 0.8922 (1.0055) loss 0.6816 (0.7993) grad_norm 9.3405 (8.5204/1.7836) mem 68106MB [2022-12-20 19:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][530/1519] eta 0:16:36 lr 0.000001 time 0.9283 (1.0071) model_time 0.9281 (1.0053) loss 0.7717 (0.7996) grad_norm 12.4608 (8.5428/1.7952) mem 68106MB [2022-12-20 19:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][540/1519] eta 0:16:26 lr 0.000001 time 0.9222 (1.0072) model_time 0.9221 (1.0054) loss 1.0331 (0.7985) grad_norm 7.8903 (8.5717/1.8875) mem 68106MB [2022-12-20 19:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][550/1519] eta 0:16:15 lr 0.000001 time 0.9327 (1.0070) model_time 0.9326 (1.0053) loss 0.8047 (0.7981) grad_norm 11.7761 (8.5632/1.8914) mem 68106MB [2022-12-20 19:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][560/1519] eta 0:16:05 lr 0.000001 time 0.9321 (1.0069) model_time 0.9319 (1.0052) loss 0.6869 (0.7984) grad_norm 10.6706 (8.5550/1.8843) mem 68106MB [2022-12-20 19:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][570/1519] eta 0:15:55 lr 0.000001 time 0.9262 (1.0072) model_time 0.9261 (1.0055) loss 0.8471 (0.7980) grad_norm 6.6043 (8.5453/1.8724) mem 68106MB [2022-12-20 19:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][580/1519] eta 0:15:45 lr 0.000001 time 0.9319 (1.0071) model_time 0.9317 (1.0054) loss 0.6912 (0.7992) grad_norm 9.9957 (8.5526/1.8702) mem 68106MB [2022-12-20 19:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][590/1519] eta 0:15:35 lr 0.000001 time 0.9796 (1.0071) model_time 0.9794 (1.0055) loss 0.9523 (0.7988) grad_norm 8.2687 (8.5525/1.8545) mem 68106MB [2022-12-20 19:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][600/1519] eta 0:15:25 lr 0.000001 time 0.9084 (1.0072) model_time 0.9082 (1.0056) loss 0.7180 (0.7983) grad_norm 8.8873 (8.5420/1.8482) mem 68106MB [2022-12-20 19:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][610/1519] eta 0:15:15 lr 0.000001 time 0.9322 (1.0072) model_time 0.9320 (1.0056) loss 0.6989 (0.7988) grad_norm 7.5550 (8.5490/1.9116) mem 68106MB [2022-12-20 19:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9305 (1.0071) model_time 0.9303 (1.0055) loss 0.8231 (0.7989) grad_norm 9.2405 (8.5439/1.9081) mem 68106MB [2022-12-20 19:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9251 (1.0072) model_time 0.9250 (1.0056) loss 0.8424 (0.7981) grad_norm 6.7549 (8.5758/1.9208) mem 68106MB [2022-12-20 19:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][640/1519] eta 0:14:45 lr 0.000001 time 0.9193 (1.0071) model_time 0.9191 (1.0055) loss 0.7844 (0.7978) grad_norm 7.2541 (8.5946/1.9235) mem 68106MB [2022-12-20 19:11:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][650/1519] eta 0:14:35 lr 0.000001 time 0.9288 (1.0069) model_time 0.9286 (1.0054) loss 0.8460 (0.7983) grad_norm 8.8368 (8.5963/1.9096) mem 68106MB [2022-12-20 19:11:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][660/1519] eta 0:14:24 lr 0.000001 time 0.9292 (1.0068) model_time 0.9289 (1.0053) loss 1.0238 (0.7986) grad_norm 8.0463 (8.5894/1.9121) mem 68106MB [2022-12-20 19:11:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][670/1519] eta 0:14:14 lr 0.000001 time 0.9686 (1.0067) model_time 0.9680 (1.0052) loss 0.9011 (0.7996) grad_norm 8.4584 (8.5764/1.8951) mem 68106MB [2022-12-20 19:11:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][680/1519] eta 0:14:04 lr 0.000001 time 0.9282 (1.0067) model_time 0.9279 (1.0052) loss 0.7412 (0.7998) grad_norm 11.4587 (8.6084/1.9031) mem 68106MB [2022-12-20 19:12:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][690/1519] eta 0:13:54 lr 0.000001 time 0.9305 (1.0065) model_time 0.9303 (1.0051) loss 1.0449 (0.8003) grad_norm 9.6156 (8.5870/1.8793) mem 68106MB [2022-12-20 19:12:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][700/1519] eta 0:13:44 lr 0.000001 time 0.9254 (1.0065) model_time 0.9252 (1.0051) loss 0.6724 (0.8002) grad_norm 12.4284 (8.6170/1.8849) mem 68106MB [2022-12-20 19:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][710/1519] eta 0:13:34 lr 0.000001 time 0.9279 (1.0064) model_time 0.9278 (1.0049) loss 0.6870 (0.7996) grad_norm 9.5903 (8.6488/1.8837) mem 68106MB [2022-12-20 19:12:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][720/1519] eta 0:13:24 lr 0.000001 time 0.9259 (1.0064) model_time 0.9258 (1.0049) loss 0.7307 (0.7997) grad_norm 9.2992 (8.6562/1.8827) mem 68106MB [2022-12-20 19:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][730/1519] eta 0:13:13 lr 0.000001 time 0.9302 (1.0063) model_time 0.9300 (1.0049) loss 0.9545 (0.8009) grad_norm 9.6745 (8.6650/1.8775) mem 68106MB [2022-12-20 19:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9256 (1.0066) model_time 0.9255 (1.0052) loss 0.6760 (0.8003) grad_norm 6.1558 (8.6556/1.8843) mem 68106MB [2022-12-20 19:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][750/1519] eta 0:12:54 lr 0.000001 time 0.9206 (1.0066) model_time 0.9204 (1.0052) loss 0.6714 (0.7993) grad_norm 8.9022 (8.6524/1.8690) mem 68106MB [2022-12-20 19:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9234 (1.0066) model_time 0.9233 (1.0052) loss 0.8764 (0.7999) grad_norm 9.0309 (8.6884/1.9856) mem 68106MB [2022-12-20 19:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][770/1519] eta 0:12:33 lr 0.000001 time 0.9815 (1.0066) model_time 0.9813 (1.0052) loss 1.0644 (0.8001) grad_norm 7.4662 (8.6831/1.9805) mem 68106MB [2022-12-20 19:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9305 (1.0066) model_time 0.9303 (1.0053) loss 0.7536 (0.7999) grad_norm 6.5923 (8.7091/1.9913) mem 68106MB [2022-12-20 19:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9348 (1.0066) model_time 0.9345 (1.0053) loss 0.6698 (0.7994) grad_norm 7.1341 (8.7054/1.9902) mem 68106MB [2022-12-20 19:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][800/1519] eta 0:12:03 lr 0.000001 time 0.9166 (1.0065) model_time 0.9159 (1.0052) loss 0.7715 (0.7999) grad_norm 5.9771 (8.6845/1.9987) mem 68106MB [2022-12-20 19:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][810/1519] eta 0:11:53 lr 0.000001 time 0.9320 (1.0064) model_time 0.9318 (1.0051) loss 0.8467 (0.7996) grad_norm 8.1923 (8.6594/1.9832) mem 68106MB [2022-12-20 19:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][820/1519] eta 0:11:43 lr 0.000001 time 0.9287 (1.0063) model_time 0.9285 (1.0050) loss 0.7527 (0.7992) grad_norm 9.6027 (8.6806/1.9809) mem 68106MB [2022-12-20 19:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9318 (1.0062) model_time 0.9317 (1.0049) loss 0.7801 (0.7995) grad_norm 10.6331 (8.6719/1.9800) mem 68106MB [2022-12-20 19:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9275 (1.0062) model_time 0.9273 (1.0049) loss 0.6705 (0.7994) grad_norm 10.0491 (8.6560/1.9738) mem 68106MB [2022-12-20 19:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][850/1519] eta 0:11:13 lr 0.000001 time 0.8968 (1.0062) model_time 0.8967 (1.0049) loss 0.7579 (0.7992) grad_norm 9.2600 (8.6554/1.9612) mem 68106MB [2022-12-20 19:14:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][860/1519] eta 0:11:03 lr 0.000001 time 0.9299 (1.0061) model_time 0.9298 (1.0049) loss 1.0035 (0.8000) grad_norm 9.5466 (8.6701/1.9606) mem 68106MB [2022-12-20 19:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][870/1519] eta 0:10:52 lr 0.000001 time 0.9212 (1.0061) model_time 0.9211 (1.0048) loss 0.7493 (0.7999) grad_norm 8.7924 (8.6774/1.9453) mem 68106MB [2022-12-20 19:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][880/1519] eta 0:10:42 lr 0.000001 time 0.9359 (1.0061) model_time 0.9356 (1.0048) loss 0.6630 (0.7988) grad_norm 6.7406 (8.7015/1.9514) mem 68106MB [2022-12-20 19:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][890/1519] eta 0:10:32 lr 0.000001 time 0.9297 (1.0060) model_time 0.9296 (1.0048) loss 0.7264 (0.7993) grad_norm 6.2262 (8.6817/1.9711) mem 68106MB [2022-12-20 19:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][900/1519] eta 0:10:22 lr 0.000001 time 0.9243 (1.0060) model_time 0.9240 (1.0048) loss 0.7010 (0.7994) grad_norm 6.6909 (8.7039/1.9619) mem 68106MB [2022-12-20 19:15:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][910/1519] eta 0:10:12 lr 0.000001 time 0.9421 (1.0064) model_time 0.9419 (1.0051) loss 0.6660 (0.7989) grad_norm 7.1518 (8.6454/1.9667) mem 68106MB [2022-12-20 19:16:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][920/1519] eta 0:10:02 lr 0.000001 time 0.9324 (1.0064) model_time 0.9322 (1.0052) loss 0.7609 (0.7991) grad_norm 9.9137 (8.6610/1.9612) mem 68106MB [2022-12-20 19:16:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][930/1519] eta 0:09:52 lr 0.000001 time 0.9298 (1.0064) model_time 0.9296 (1.0052) loss 0.6846 (0.7985) grad_norm 9.0451 (8.6487/1.9680) mem 68106MB [2022-12-20 19:16:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9141 (1.0065) model_time 0.9139 (1.0054) loss 0.7126 (0.7989) grad_norm 7.3297 (8.6691/2.0368) mem 68106MB [2022-12-20 19:16:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][950/1519] eta 0:09:32 lr 0.000001 time 0.9368 (1.0065) model_time 0.9366 (1.0053) loss 0.7768 (0.7990) grad_norm 10.1302 (8.6822/2.0182) mem 68106MB [2022-12-20 19:16:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9335 (1.0065) model_time 0.9334 (1.0054) loss 1.0725 (0.7989) grad_norm 14.9408 (8.6940/2.0347) mem 68106MB [2022-12-20 19:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][970/1519] eta 0:09:12 lr 0.000001 time 0.9264 (1.0065) model_time 0.9262 (1.0053) loss 0.7822 (0.7989) grad_norm 8.3223 (8.6718/2.0225) mem 68106MB [2022-12-20 19:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9317 (1.0064) model_time 0.9316 (1.0052) loss 0.7899 (0.7993) grad_norm 6.6723 (8.7084/2.0544) mem 68106MB [2022-12-20 19:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9173 (1.0064) model_time 0.9171 (1.0053) loss 0.7345 (0.7988) grad_norm 6.3978 (8.6738/2.0390) mem 68106MB [2022-12-20 19:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9376 (1.0064) model_time 0.9375 (1.0053) loss 0.9427 (0.7990) grad_norm 6.3684 (8.6640/2.0410) mem 68106MB [2022-12-20 19:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9324 (1.0064) model_time 0.9322 (1.0052) loss 0.7794 (0.7993) grad_norm 9.3684 (8.6647/2.0410) mem 68106MB [2022-12-20 19:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1020/1519] eta 0:08:22 lr 0.000001 time 0.9289 (1.0064) model_time 0.9288 (1.0052) loss 0.7574 (0.7994) grad_norm 7.4653 (8.6998/2.0993) mem 68106MB [2022-12-20 19:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1030/1519] eta 0:08:12 lr 0.000001 time 0.9267 (1.0063) model_time 0.9264 (1.0052) loss 0.9023 (0.7995) grad_norm 6.6708 (8.6848/2.0886) mem 68106MB [2022-12-20 19:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1040/1519] eta 0:08:01 lr 0.000001 time 0.9253 (1.0062) model_time 0.9249 (1.0051) loss 0.9092 (0.8000) grad_norm 8.4834 (8.6682/2.0904) mem 68106MB [2022-12-20 19:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1050/1519] eta 0:07:51 lr 0.000001 time 0.9591 (1.0063) model_time 0.9590 (1.0052) loss 0.7293 (0.7996) grad_norm 6.8396 (8.6695/2.0918) mem 68106MB [2022-12-20 19:18:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1060/1519] eta 0:07:41 lr 0.000001 time 0.9258 (1.0063) model_time 0.9254 (1.0052) loss 0.8657 (0.7998) grad_norm 6.5884 (8.6702/2.0918) mem 68106MB [2022-12-20 19:18:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1070/1519] eta 0:07:31 lr 0.000001 time 0.9423 (1.0063) model_time 0.9421 (1.0052) loss 0.8394 (0.7998) grad_norm 8.9692 (8.6598/2.1008) mem 68106MB [2022-12-20 19:18:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1080/1519] eta 0:07:21 lr 0.000001 time 0.9359 (1.0062) model_time 0.9358 (1.0052) loss 0.9565 (0.7995) grad_norm 6.4451 (8.6531/2.0914) mem 68106MB [2022-12-20 19:18:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9511 (1.0062) model_time 0.9509 (1.0052) loss 0.6632 (0.7998) grad_norm 8.9215 (8.6506/2.0989) mem 68106MB [2022-12-20 19:19:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9330 (1.0062) model_time 0.9329 (1.0052) loss 0.8978 (0.8001) grad_norm 9.0454 (8.6613/2.0989) mem 68106MB [2022-12-20 19:19:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9258 (1.0062) model_time 0.9257 (1.0051) loss 0.7463 (0.8006) grad_norm 9.7502 (8.6755/2.0949) mem 68106MB [2022-12-20 19:19:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9415 (1.0061) model_time 0.9413 (1.0051) loss 0.8598 (0.8004) grad_norm 5.7747 (8.6595/2.1166) mem 68106MB [2022-12-20 19:19:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9288 (1.0061) model_time 0.9286 (1.0050) loss 0.6925 (0.8002) grad_norm 10.2902 (8.6443/2.1003) mem 68106MB [2022-12-20 19:19:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9335 (1.0060) model_time 0.9334 (1.0050) loss 0.8197 (0.8004) grad_norm 7.2404 (8.6137/2.0175) mem 68106MB [2022-12-20 19:19:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9232 (1.0060) model_time 0.9230 (1.0049) loss 0.8750 (0.8009) grad_norm 6.0177 (8.6160/2.0326) mem 68106MB [2022-12-20 19:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9321 (1.0060) model_time 0.9319 (1.0049) loss 0.6866 (0.8005) grad_norm 10.6437 (8.6469/2.0413) mem 68106MB [2022-12-20 19:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9341 (1.0060) model_time 0.9339 (1.0050) loss 0.6616 (0.8012) grad_norm 8.0860 (8.6499/2.0383) mem 68106MB [2022-12-20 19:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9252 (1.0060) model_time 0.9251 (1.0050) loss 0.7028 (0.8010) grad_norm 9.3201 (8.6609/2.0451) mem 68106MB [2022-12-20 19:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1190/1519] eta 0:05:31 lr 0.000001 time 0.9257 (1.0062) model_time 0.9256 (1.0052) loss 0.7833 (0.8015) grad_norm 12.9901 (8.6786/2.0622) mem 68106MB [2022-12-20 19:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1200/1519] eta 0:05:20 lr 0.000001 time 0.9242 (1.0061) model_time 0.9241 (1.0051) loss 0.7041 (0.8018) grad_norm 7.9675 (8.6862/2.0576) mem 68106MB [2022-12-20 19:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1210/1519] eta 0:05:10 lr 0.000001 time 0.9288 (1.0062) model_time 0.9286 (1.0052) loss 1.0267 (0.8023) grad_norm 10.9604 (8.6893/1.9981) mem 68106MB [2022-12-20 19:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1220/1519] eta 0:05:00 lr 0.000001 time 0.9277 (1.0062) model_time 0.9275 (1.0052) loss 0.8109 (0.8019) grad_norm 9.8037 (8.7303/2.0161) mem 68106MB [2022-12-20 19:21:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1230/1519] eta 0:04:50 lr 0.000001 time 0.9262 (1.0062) model_time 0.9260 (1.0052) loss 0.7592 (0.8019) grad_norm 10.6179 (8.7205/1.9994) mem 68106MB [2022-12-20 19:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9310 (1.0062) model_time 0.9308 (1.0052) loss 1.0747 (0.8020) grad_norm 8.6538 (8.7167/2.0037) mem 68106MB [2022-12-20 19:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9087 (1.0062) model_time 0.9086 (1.0052) loss 1.0380 (0.8020) grad_norm 9.5064 (8.7116/2.0146) mem 68106MB [2022-12-20 19:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9365 (1.0063) model_time 0.9364 (1.0053) loss 0.7084 (0.8013) grad_norm 11.6369 (8.7084/2.0236) mem 68106MB [2022-12-20 19:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9235 (1.0063) model_time 0.9233 (1.0053) loss 0.8836 (0.8019) grad_norm 7.1150 (8.7187/2.0369) mem 68106MB [2022-12-20 19:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9359 (1.0064) model_time 0.9358 (1.0055) loss 0.8507 (0.8021) grad_norm 8.7904 (8.6814/2.0335) mem 68106MB [2022-12-20 19:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9327 (1.0064) model_time 0.9326 (1.0054) loss 0.7563 (0.8018) grad_norm 14.9791 (8.6982/2.0655) mem 68106MB [2022-12-20 19:22:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9734 (1.0064) model_time 0.9733 (1.0054) loss 0.6637 (0.8017) grad_norm 8.8825 (8.6931/2.0520) mem 68106MB [2022-12-20 19:22:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9164 (1.0063) model_time 0.9162 (1.0054) loss 0.9956 (0.8013) grad_norm 6.3914 (8.6489/2.0659) mem 68106MB [2022-12-20 19:22:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9271 (1.0063) model_time 0.9270 (1.0054) loss 0.7738 (0.8014) grad_norm 9.4961 (8.6389/2.0744) mem 68106MB [2022-12-20 19:22:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9226 (1.0063) model_time 0.9225 (1.0053) loss 0.7233 (0.8013) grad_norm 11.0057 (8.6510/2.0833) mem 68106MB [2022-12-20 19:23:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9260 (1.0063) model_time 0.9258 (1.0053) loss 0.7521 (0.8013) grad_norm 10.0292 (8.6662/2.0852) mem 68106MB [2022-12-20 19:23:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9243 (1.0063) model_time 0.9241 (1.0053) loss 0.8107 (0.8007) grad_norm 8.2579 (8.6727/2.0871) mem 68106MB [2022-12-20 19:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1360/1519] eta 0:02:39 lr 0.000001 time 0.9245 (1.0062) model_time 0.9243 (1.0053) loss 0.6762 (0.8006) grad_norm 8.0024 (8.6285/1.9670) mem 68106MB [2022-12-20 19:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1370/1519] eta 0:02:29 lr 0.000001 time 0.9184 (1.0062) model_time 0.9183 (1.0053) loss 1.2158 (0.8007) grad_norm 7.1012 (8.6483/1.9808) mem 68106MB [2022-12-20 19:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9314 (1.0062) model_time 0.9312 (1.0052) loss 0.7633 (0.8004) grad_norm 7.1531 (8.6320/1.9665) mem 68106MB [2022-12-20 19:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9251 (1.0061) model_time 0.9250 (1.0052) loss 0.7091 (0.8010) grad_norm 6.4778 (8.6251/1.9681) mem 68106MB [2022-12-20 19:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9229 (1.0061) model_time 0.9228 (1.0052) loss 0.6987 (0.8011) grad_norm 6.8347 (8.6186/1.9615) mem 68106MB [2022-12-20 19:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9290 (1.0061) model_time 0.9288 (1.0052) loss 0.6668 (0.8007) grad_norm 10.0256 (8.6238/1.9631) mem 68106MB [2022-12-20 19:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9305 (1.0060) model_time 0.9304 (1.0051) loss 0.7457 (0.8006) grad_norm 7.6905 (8.6355/1.9644) mem 68106MB [2022-12-20 19:24:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9225 (1.0060) model_time 0.9223 (1.0051) loss 0.7468 (0.8010) grad_norm 11.0331 (8.6515/1.9601) mem 68106MB [2022-12-20 19:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9383 (1.0060) model_time 0.9381 (1.0050) loss 1.1120 (0.8019) grad_norm 9.1926 (8.6628/1.9796) mem 68106MB [2022-12-20 19:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9307 (1.0059) model_time 0.9304 (1.0050) loss 0.7128 (0.8018) grad_norm 7.2664 (8.6261/1.9839) mem 68106MB [2022-12-20 19:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9313 (1.0059) model_time 0.9311 (1.0050) loss 0.9236 (0.8016) grad_norm 8.1438 (8.6217/1.9825) mem 68106MB [2022-12-20 19:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9312 (1.0058) model_time 0.9311 (1.0049) loss 0.6678 (0.8012) grad_norm 7.2804 (8.5838/1.9787) mem 68106MB [2022-12-20 19:25:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9771 (1.0058) model_time 0.9769 (1.0049) loss 0.7221 (0.8012) grad_norm inf (8.5849/1.9928) mem 68106MB [2022-12-20 19:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9279 (1.0058) model_time 0.9276 (1.0049) loss 0.7705 (0.8011) grad_norm 5.7016 (8.5841/1.9932) mem 68106MB [2022-12-20 19:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9685 (1.0058) model_time 0.9684 (1.0049) loss 0.7299 (0.8007) grad_norm 7.0388 (8.5694/1.9928) mem 68106MB [2022-12-20 19:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [88/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9215 (1.0058) model_time 0.9213 (1.0049) loss 0.7707 (0.8008) grad_norm 11.1264 (8.6251/1.9869) mem 68106MB [2022-12-20 19:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 88 training takes 0:25:27 [2022-12-20 19:26:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_88.pth saving...... [2022-12-20 19:26:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_88.pth saved !!! [2022-12-20 19:26:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.638 (0.638) Loss 0.5390 (0.5390) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 19:26:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.295 (0.327) Loss 0.5348 (0.5087) Acc@1 92.708 (92.771) Acc@5 98.611 (98.516) Mem 68106MB [2022-12-20 19:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.312) Loss 0.4849 (0.5035) Acc@1 91.667 (92.774) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-20 19:26:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.308) Loss 0.6354 (0.5105) Acc@1 90.972 (92.540) Acc@5 97.917 (98.477) Mem 68106MB [2022-12-20 19:26:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.303 (0.305) Loss 0.4564 (0.5012) Acc@1 93.750 (92.607) Acc@5 99.306 (98.552) Mem 68106MB [2022-12-20 19:26:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.297 (0.304) Loss 0.4895 (0.4987) Acc@1 92.014 (92.667) Acc@5 99.653 (98.598) Mem 68106MB [2022-12-20 19:26:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.303) Loss 0.5101 (0.4984) Acc@1 90.972 (92.589) Acc@5 98.264 (98.571) Mem 68106MB [2022-12-20 19:26:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.302) Loss 0.5467 (0.4997) Acc@1 92.014 (92.532) Acc@5 97.917 (98.552) Mem 68106MB [2022-12-20 19:26:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.301) Loss 0.4289 (0.4984) Acc@1 93.056 (92.580) Acc@5 98.611 (98.590) Mem 68106MB [2022-12-20 19:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:88] * Acc@1 92.538 Acc@5 98.592 [2022-12-20 19:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 19:26:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 19:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][0/1519] eta 0:47:31 lr 0.000001 time 1.8773 (1.8773) model_time 1.0991 (1.0991) loss 0.6783 (0.6783) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 19:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][10/1519] eta 0:27:56 lr 0.000001 time 0.9235 (1.1108) model_time 0.9233 (1.0397) loss 0.6774 (0.7492) grad_norm 7.2299 (8.5484/1.5611) mem 68106MB [2022-12-20 19:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][20/1519] eta 0:26:32 lr 0.000001 time 0.9282 (1.0624) model_time 0.9281 (1.0249) loss 0.9413 (0.7890) grad_norm 9.3730 (9.6014/2.2289) mem 68106MB [2022-12-20 19:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][30/1519] eta 0:26:04 lr 0.000001 time 0.9058 (1.0505) model_time 0.9056 (1.0250) loss 0.6884 (0.7937) grad_norm 7.2736 (9.4330/1.9525) mem 68106MB [2022-12-20 19:27:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][40/1519] eta 0:25:53 lr 0.000001 time 0.9212 (1.0501) model_time 0.9210 (1.0307) loss 0.7330 (0.7861) grad_norm 7.5876 (9.1154/1.8434) mem 68106MB [2022-12-20 19:27:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][50/1519] eta 0:25:30 lr 0.000001 time 0.9311 (1.0420) model_time 0.9309 (1.0264) loss 0.6827 (0.7928) grad_norm 10.2788 (8.7412/1.9705) mem 68106MB [2022-12-20 19:27:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][60/1519] eta 0:25:09 lr 0.000001 time 0.9211 (1.0344) model_time 0.9210 (1.0213) loss 0.6635 (0.7931) grad_norm 9.2790 (8.5383/1.8944) mem 68106MB [2022-12-20 19:28:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][70/1519] eta 0:24:53 lr 0.000001 time 1.0074 (1.0308) model_time 1.0073 (1.0195) loss 0.6706 (0.8012) grad_norm 6.3693 (8.4554/1.8051) mem 68106MB [2022-12-20 19:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][80/1519] eta 0:24:37 lr 0.000001 time 0.9321 (1.0266) model_time 0.9320 (1.0166) loss 0.7758 (0.7985) grad_norm 7.6174 (8.3859/1.7412) mem 68106MB [2022-12-20 19:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][90/1519] eta 0:24:22 lr 0.000001 time 0.9256 (1.0234) model_time 0.9254 (1.0145) loss 0.8641 (0.7967) grad_norm 8.1925 (8.3844/1.6739) mem 68106MB [2022-12-20 19:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][100/1519] eta 0:24:08 lr 0.000001 time 0.9224 (1.0210) model_time 0.9223 (1.0129) loss 0.8864 (0.7962) grad_norm 9.9190 (8.3754/1.6922) mem 68106MB [2022-12-20 19:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][110/1519] eta 0:23:56 lr 0.000001 time 0.9307 (1.0196) model_time 0.9305 (1.0122) loss 0.6704 (0.7942) grad_norm 8.7485 (8.3356/1.6787) mem 68106MB [2022-12-20 19:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][120/1519] eta 0:23:44 lr 0.000001 time 0.9345 (1.0181) model_time 0.9338 (1.0113) loss 0.8611 (0.7952) grad_norm 20.9793 (8.4815/2.3197) mem 68106MB [2022-12-20 19:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][130/1519] eta 0:23:35 lr 0.000001 time 0.8863 (1.0191) model_time 0.8861 (1.0128) loss 0.6894 (0.7922) grad_norm 7.3181 (8.5128/2.2535) mem 68106MB [2022-12-20 19:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][140/1519] eta 0:23:23 lr 0.000001 time 0.9368 (1.0177) model_time 0.9365 (1.0118) loss 0.6987 (0.7943) grad_norm 15.8013 (8.5489/2.3643) mem 68106MB [2022-12-20 19:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][150/1519] eta 0:23:11 lr 0.000001 time 0.9179 (1.0166) model_time 0.9177 (1.0111) loss 0.6656 (0.7923) grad_norm 10.6240 (8.4925/2.3523) mem 68106MB [2022-12-20 19:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][160/1519] eta 0:23:00 lr 0.000001 time 0.9317 (1.0159) model_time 0.9315 (1.0107) loss 0.8544 (0.8001) grad_norm 9.2459 (8.5591/2.2998) mem 68106MB [2022-12-20 19:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][170/1519] eta 0:22:49 lr 0.000001 time 0.9281 (1.0149) model_time 0.9279 (1.0100) loss 0.8196 (0.7994) grad_norm 5.5925 (8.5735/2.2694) mem 68106MB [2022-12-20 19:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][180/1519] eta 0:22:38 lr 0.000001 time 0.9207 (1.0145) model_time 0.9205 (1.0098) loss 0.7229 (0.7984) grad_norm 7.1999 (8.4932/2.2665) mem 68106MB [2022-12-20 19:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][190/1519] eta 0:22:27 lr 0.000001 time 0.9282 (1.0141) model_time 0.9280 (1.0097) loss 0.8406 (0.7974) grad_norm 5.8371 (8.5356/2.3369) mem 68106MB [2022-12-20 19:30:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][200/1519] eta 0:22:16 lr 0.000001 time 0.9360 (1.0136) model_time 0.9358 (1.0093) loss 0.7191 (0.7972) grad_norm 7.2522 (8.6551/2.6523) mem 68106MB [2022-12-20 19:30:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][210/1519] eta 0:22:05 lr 0.000001 time 0.9225 (1.0130) model_time 0.9223 (1.0089) loss 0.6901 (0.7955) grad_norm 11.6498 (8.6724/2.6693) mem 68106MB [2022-12-20 19:30:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][220/1519] eta 0:21:55 lr 0.000001 time 0.9436 (1.0123) model_time 0.9434 (1.0084) loss 0.6586 (0.7928) grad_norm 8.7811 (8.7160/2.7099) mem 68106MB [2022-12-20 19:30:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][230/1519] eta 0:21:44 lr 0.000001 time 0.9291 (1.0124) model_time 0.9290 (1.0086) loss 0.7300 (0.7915) grad_norm 9.6342 (8.7976/2.7056) mem 68106MB [2022-12-20 19:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][240/1519] eta 0:21:34 lr 0.000001 time 0.9305 (1.0122) model_time 0.9304 (1.0085) loss 0.7290 (0.7909) grad_norm 7.6030 (8.7804/2.6691) mem 68106MB [2022-12-20 19:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][250/1519] eta 0:21:24 lr 0.000001 time 0.9248 (1.0119) model_time 0.9247 (1.0084) loss 0.7693 (0.7882) grad_norm 6.8620 (8.7167/2.6422) mem 68106MB [2022-12-20 19:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][260/1519] eta 0:21:13 lr 0.000001 time 0.9269 (1.0114) model_time 0.9268 (1.0080) loss 0.7125 (0.7874) grad_norm 7.7551 (8.7325/2.6092) mem 68106MB [2022-12-20 19:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][270/1519] eta 0:21:02 lr 0.000001 time 0.9430 (1.0110) model_time 0.9428 (1.0077) loss 0.7892 (0.7882) grad_norm 9.1413 (8.7102/2.5740) mem 68106MB [2022-12-20 19:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][280/1519] eta 0:20:52 lr 0.000001 time 0.9308 (1.0106) model_time 0.9306 (1.0074) loss 0.6893 (0.7870) grad_norm 8.6368 (8.6778/2.5451) mem 68106MB [2022-12-20 19:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][290/1519] eta 0:20:42 lr 0.000001 time 0.9829 (1.0107) model_time 0.9828 (1.0076) loss 0.7324 (0.7854) grad_norm 10.2855 (8.7484/2.7068) mem 68106MB [2022-12-20 19:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][300/1519] eta 0:20:31 lr 0.000001 time 0.9241 (1.0102) model_time 0.9240 (1.0072) loss 0.7566 (0.7842) grad_norm 7.8664 (8.6949/2.6827) mem 68106MB [2022-12-20 19:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][310/1519] eta 0:20:21 lr 0.000001 time 0.9305 (1.0100) model_time 0.9303 (1.0071) loss 0.7955 (0.7845) grad_norm 5.6570 (8.6622/2.6545) mem 68106MB [2022-12-20 19:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][320/1519] eta 0:20:10 lr 0.000001 time 0.9604 (1.0099) model_time 0.9602 (1.0071) loss 0.6546 (0.7853) grad_norm 9.5810 (8.6513/2.6163) mem 68106MB [2022-12-20 19:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][330/1519] eta 0:20:00 lr 0.000001 time 0.9621 (1.0097) model_time 0.9619 (1.0069) loss 0.6717 (0.7870) grad_norm 7.8045 (8.6421/2.5948) mem 68106MB [2022-12-20 19:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][340/1519] eta 0:19:50 lr 0.000001 time 0.9248 (1.0096) model_time 0.9246 (1.0070) loss 0.7301 (0.7893) grad_norm 8.6537 (8.6204/2.5650) mem 68106MB [2022-12-20 19:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][350/1519] eta 0:19:40 lr 0.000001 time 0.9916 (1.0101) model_time 0.9915 (1.0075) loss 0.9334 (0.7905) grad_norm 8.7376 (8.6042/2.5354) mem 68106MB [2022-12-20 19:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][360/1519] eta 0:19:30 lr 0.000001 time 0.9284 (1.0098) model_time 0.9283 (1.0073) loss 0.6609 (0.7920) grad_norm 10.3256 (8.6614/2.5952) mem 68106MB [2022-12-20 19:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][370/1519] eta 0:19:20 lr 0.000001 time 0.9464 (1.0096) model_time 0.9462 (1.0071) loss 0.7879 (0.7913) grad_norm 7.3866 (8.6726/2.5746) mem 68106MB [2022-12-20 19:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][380/1519] eta 0:19:09 lr 0.000001 time 0.9253 (1.0093) model_time 0.9252 (1.0069) loss 0.7990 (0.7918) grad_norm 8.2208 (8.6735/2.5521) mem 68106MB [2022-12-20 19:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][390/1519] eta 0:18:59 lr 0.000001 time 0.9188 (1.0093) model_time 0.9187 (1.0070) loss 0.6812 (0.7931) grad_norm 7.9369 (8.6612/2.5303) mem 68106MB [2022-12-20 19:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][400/1519] eta 0:18:49 lr 0.000001 time 0.9437 (1.0093) model_time 0.9435 (1.0070) loss 0.6747 (0.7919) grad_norm 8.6297 (8.6368/2.5117) mem 68106MB [2022-12-20 19:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][410/1519] eta 0:18:39 lr 0.000001 time 0.9290 (1.0091) model_time 0.9288 (1.0068) loss 0.6583 (0.7917) grad_norm 10.0533 (8.6250/2.4879) mem 68106MB [2022-12-20 19:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][420/1519] eta 0:18:28 lr 0.000001 time 0.9321 (1.0090) model_time 0.9319 (1.0068) loss 0.6968 (0.7934) grad_norm 5.5151 (8.6202/2.4857) mem 68106MB [2022-12-20 19:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][430/1519] eta 0:18:18 lr 0.000001 time 0.9399 (1.0089) model_time 0.9397 (1.0067) loss 0.9213 (0.7928) grad_norm 7.0772 (8.6191/2.4774) mem 68106MB [2022-12-20 19:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][440/1519] eta 0:18:08 lr 0.000001 time 0.9271 (1.0087) model_time 0.9269 (1.0066) loss 0.8528 (0.7935) grad_norm 8.5043 (8.6168/2.4598) mem 68106MB [2022-12-20 19:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][450/1519] eta 0:17:58 lr 0.000001 time 0.9256 (1.0086) model_time 0.9254 (1.0065) loss 0.8683 (0.7954) grad_norm 7.5039 (8.5918/2.4457) mem 68106MB [2022-12-20 19:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][460/1519] eta 0:17:47 lr 0.000001 time 0.9326 (1.0084) model_time 0.9325 (1.0064) loss 0.7050 (0.7956) grad_norm 7.0694 (8.5914/2.4274) mem 68106MB [2022-12-20 19:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][470/1519] eta 0:17:37 lr 0.000001 time 1.0016 (1.0085) model_time 1.0015 (1.0065) loss 1.1574 (0.7957) grad_norm 18.7006 (8.6222/2.4993) mem 68106MB [2022-12-20 19:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][480/1519] eta 0:17:27 lr 0.000001 time 0.9296 (1.0083) model_time 0.9295 (1.0064) loss 0.6794 (0.7965) grad_norm 10.7523 (8.6383/2.4841) mem 68106MB [2022-12-20 19:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][490/1519] eta 0:17:17 lr 0.000001 time 0.9313 (1.0083) model_time 0.9312 (1.0064) loss 0.6637 (0.7983) grad_norm 8.8197 (8.6627/2.4768) mem 68106MB [2022-12-20 19:35:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][500/1519] eta 0:17:07 lr 0.000001 time 0.9225 (1.0082) model_time 0.9223 (1.0063) loss 0.8826 (0.7979) grad_norm 6.4403 (8.6497/2.4649) mem 68106MB [2022-12-20 19:35:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][510/1519] eta 0:16:57 lr 0.000001 time 0.9791 (1.0083) model_time 0.9788 (1.0065) loss 0.7536 (0.7974) grad_norm 10.7814 (8.6484/2.4519) mem 68106MB [2022-12-20 19:35:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][520/1519] eta 0:16:47 lr 0.000001 time 0.9239 (1.0082) model_time 0.9237 (1.0064) loss 0.9026 (0.7993) grad_norm 6.5908 (8.6688/2.4460) mem 68106MB [2022-12-20 19:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][530/1519] eta 0:16:37 lr 0.000001 time 1.0040 (1.0082) model_time 1.0039 (1.0064) loss 0.7780 (0.7988) grad_norm 7.7565 (8.6631/2.4296) mem 68106MB [2022-12-20 19:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][540/1519] eta 0:16:27 lr 0.000001 time 0.9165 (1.0083) model_time 0.9164 (1.0065) loss 0.6975 (0.7980) grad_norm 8.8001 (8.6733/2.4190) mem 68106MB [2022-12-20 19:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][550/1519] eta 0:16:17 lr 0.000001 time 0.9683 (1.0083) model_time 0.9681 (1.0065) loss 0.6691 (0.7987) grad_norm 9.6794 (8.6574/2.4042) mem 68106MB [2022-12-20 19:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][560/1519] eta 0:16:06 lr 0.000001 time 0.9301 (1.0082) model_time 0.9299 (1.0064) loss 0.6705 (0.7997) grad_norm 12.5676 (8.6680/2.3988) mem 68106MB [2022-12-20 19:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][570/1519] eta 0:15:56 lr 0.000001 time 0.9295 (1.0082) model_time 0.9293 (1.0065) loss 1.1078 (0.8010) grad_norm 9.8408 (8.6549/2.3836) mem 68106MB [2022-12-20 19:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][580/1519] eta 0:15:46 lr 0.000001 time 0.9196 (1.0081) model_time 0.9194 (1.0064) loss 0.9113 (0.8014) grad_norm 6.8138 (8.6652/2.3821) mem 68106MB [2022-12-20 19:36:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][590/1519] eta 0:15:36 lr 0.000001 time 0.9253 (1.0080) model_time 0.9251 (1.0063) loss 1.0270 (0.8027) grad_norm 9.2356 (8.6636/2.3723) mem 68106MB [2022-12-20 19:36:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][600/1519] eta 0:15:26 lr 0.000001 time 0.9250 (1.0079) model_time 0.9248 (1.0062) loss 0.7056 (0.8025) grad_norm 8.9640 (8.6991/2.3989) mem 68106MB [2022-12-20 19:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][610/1519] eta 0:15:15 lr 0.000001 time 0.9231 (1.0077) model_time 0.9230 (1.0061) loss 0.6924 (0.8032) grad_norm 7.6553 (8.7002/2.4077) mem 68106MB [2022-12-20 19:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9287 (1.0076) model_time 0.9286 (1.0060) loss 0.7891 (0.8032) grad_norm 8.1993 (8.6387/2.3880) mem 68106MB [2022-12-20 19:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9260 (1.0079) model_time 0.9258 (1.0063) loss 0.8978 (0.8035) grad_norm 9.0173 (8.6184/2.3868) mem 68106MB [2022-12-20 19:37:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][640/1519] eta 0:14:45 lr 0.000001 time 0.9235 (1.0078) model_time 0.9233 (1.0062) loss 0.8623 (0.8043) grad_norm 8.3991 (8.6093/2.3877) mem 68106MB [2022-12-20 19:37:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][650/1519] eta 0:14:35 lr 0.000001 time 0.9232 (1.0078) model_time 0.9231 (1.0062) loss 0.6777 (0.8037) grad_norm 9.8106 (8.6271/2.3817) mem 68106MB [2022-12-20 19:37:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][660/1519] eta 0:14:26 lr 0.000001 time 1.1923 (1.0082) model_time 1.1922 (1.0067) loss 0.8851 (0.8039) grad_norm 7.9439 (8.6330/2.3803) mem 68106MB [2022-12-20 19:38:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][670/1519] eta 0:14:16 lr 0.000001 time 0.9250 (1.0083) model_time 0.9248 (1.0068) loss 0.6963 (0.8030) grad_norm 7.5881 (8.6344/2.3802) mem 68106MB [2022-12-20 19:38:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][680/1519] eta 0:14:05 lr 0.000001 time 0.9193 (1.0082) model_time 0.9191 (1.0067) loss 0.6957 (0.8019) grad_norm 7.2466 (8.6334/2.3780) mem 68106MB [2022-12-20 19:38:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][690/1519] eta 0:13:55 lr 0.000001 time 0.9783 (1.0081) model_time 0.9781 (1.0067) loss 0.6632 (0.8019) grad_norm 5.2451 (8.6356/2.3987) mem 68106MB [2022-12-20 19:38:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][700/1519] eta 0:13:45 lr 0.000001 time 0.9215 (1.0080) model_time 0.9214 (1.0065) loss 1.0763 (0.8027) grad_norm 12.4067 (8.6731/2.4077) mem 68106MB [2022-12-20 19:38:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][710/1519] eta 0:13:35 lr 0.000001 time 0.9266 (1.0079) model_time 0.9264 (1.0065) loss 0.8157 (0.8019) grad_norm 9.5214 (8.6846/2.4006) mem 68106MB [2022-12-20 19:38:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][720/1519] eta 0:13:25 lr 0.000001 time 0.9235 (1.0078) model_time 0.9234 (1.0064) loss 1.1044 (0.8025) grad_norm 7.8457 (8.6645/2.3142) mem 68106MB [2022-12-20 19:39:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][730/1519] eta 0:13:15 lr 0.000001 time 0.9600 (1.0077) model_time 0.9598 (1.0063) loss 0.9555 (0.8022) grad_norm 8.7405 (8.6750/2.3191) mem 68106MB [2022-12-20 19:39:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9214 (1.0076) model_time 0.9212 (1.0062) loss 0.8913 (0.8031) grad_norm 7.7546 (8.6572/2.2797) mem 68106MB [2022-12-20 19:39:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][750/1519] eta 0:12:54 lr 0.000001 time 0.9247 (1.0075) model_time 0.9246 (1.0061) loss 0.8353 (0.8029) grad_norm 10.2969 (8.7035/2.2767) mem 68106MB [2022-12-20 19:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9234 (1.0073) model_time 0.9233 (1.0060) loss 0.8821 (0.8031) grad_norm 7.2588 (8.6961/2.2941) mem 68106MB [2022-12-20 19:39:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][770/1519] eta 0:12:34 lr 0.000001 time 0.9213 (1.0073) model_time 0.9212 (1.0060) loss 0.8838 (0.8034) grad_norm 7.5694 (8.6891/2.2847) mem 68106MB [2022-12-20 19:39:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][780/1519] eta 0:12:24 lr 0.000001 time 0.9625 (1.0073) model_time 0.9623 (1.0060) loss 0.6765 (0.8050) grad_norm 10.6234 (8.7975/2.4108) mem 68106MB [2022-12-20 19:40:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][790/1519] eta 0:12:14 lr 0.000001 time 0.9221 (1.0073) model_time 0.9219 (1.0060) loss 0.7371 (0.8051) grad_norm 9.4923 (8.7771/2.3777) mem 68106MB [2022-12-20 19:40:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][800/1519] eta 0:12:04 lr 0.000001 time 0.9232 (1.0072) model_time 0.9231 (1.0059) loss 0.8226 (0.8054) grad_norm 10.9213 (8.7256/2.2615) mem 68106MB [2022-12-20 19:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][810/1519] eta 0:11:54 lr 0.000001 time 0.9198 (1.0072) model_time 0.9196 (1.0059) loss 0.9538 (0.8058) grad_norm 8.1373 (8.7022/2.2363) mem 68106MB [2022-12-20 19:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][820/1519] eta 0:11:44 lr 0.000001 time 0.9284 (1.0074) model_time 0.9282 (1.0061) loss 0.7043 (0.8056) grad_norm 9.7230 (8.7090/2.2330) mem 68106MB [2022-12-20 19:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][830/1519] eta 0:11:34 lr 0.000001 time 0.9186 (1.0074) model_time 0.9184 (1.0061) loss 0.8548 (0.8058) grad_norm 7.8355 (8.6725/2.2081) mem 68106MB [2022-12-20 19:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][840/1519] eta 0:11:24 lr 0.000001 time 0.9975 (1.0075) model_time 0.9973 (1.0062) loss 0.6716 (0.8052) grad_norm 6.6008 (8.6610/2.2043) mem 68106MB [2022-12-20 19:41:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][850/1519] eta 0:11:14 lr 0.000001 time 1.0230 (1.0079) model_time 1.0229 (1.0067) loss 0.7579 (0.8047) grad_norm 9.2564 (8.6759/2.1948) mem 68106MB [2022-12-20 19:41:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][860/1519] eta 0:11:04 lr 0.000001 time 0.9222 (1.0078) model_time 0.9221 (1.0066) loss 0.6523 (0.8040) grad_norm 8.5472 (8.7123/2.2557) mem 68106MB [2022-12-20 19:41:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][870/1519] eta 0:10:54 lr 0.000001 time 0.9431 (1.0078) model_time 0.9429 (1.0066) loss 0.8975 (0.8040) grad_norm 11.9921 (8.7336/2.2640) mem 68106MB [2022-12-20 19:41:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][880/1519] eta 0:10:44 lr 0.000001 time 0.9243 (1.0078) model_time 0.9241 (1.0066) loss 0.7737 (0.8035) grad_norm 7.8426 (8.8424/2.7717) mem 68106MB [2022-12-20 19:41:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][890/1519] eta 0:10:33 lr 0.000001 time 0.9274 (1.0078) model_time 0.9273 (1.0066) loss 0.6524 (0.8043) grad_norm 13.4832 (8.8198/2.6971) mem 68106MB [2022-12-20 19:41:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][900/1519] eta 0:10:23 lr 0.000001 time 0.9229 (1.0077) model_time 0.9227 (1.0065) loss 0.6729 (0.8036) grad_norm 9.7647 (8.8540/2.6874) mem 68106MB [2022-12-20 19:42:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][910/1519] eta 0:10:13 lr 0.000001 time 0.9253 (1.0076) model_time 0.9252 (1.0064) loss 0.6668 (0.8029) grad_norm 6.2294 (8.8620/2.6948) mem 68106MB [2022-12-20 19:42:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][920/1519] eta 0:10:03 lr 0.000001 time 0.9274 (1.0075) model_time 0.9273 (1.0064) loss 0.7417 (0.8039) grad_norm 9.4320 (8.8636/2.6960) mem 68106MB [2022-12-20 19:42:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][930/1519] eta 0:09:53 lr 0.000001 time 0.9534 (1.0075) model_time 0.9533 (1.0063) loss 0.8049 (0.8040) grad_norm 11.2898 (8.8811/2.6909) mem 68106MB [2022-12-20 19:42:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][940/1519] eta 0:09:43 lr 0.000001 time 0.9240 (1.0075) model_time 0.9239 (1.0063) loss 0.6756 (0.8038) grad_norm 6.8167 (8.9049/2.6941) mem 68106MB [2022-12-20 19:42:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][950/1519] eta 0:09:33 lr 0.000001 time 0.9252 (1.0074) model_time 0.9250 (1.0062) loss 0.7162 (0.8039) grad_norm 8.9114 (8.9359/2.7030) mem 68106MB [2022-12-20 19:42:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][960/1519] eta 0:09:23 lr 0.000001 time 0.9259 (1.0073) model_time 0.9257 (1.0062) loss 0.8405 (0.8035) grad_norm 10.1715 (8.9398/2.6733) mem 68106MB [2022-12-20 19:43:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][970/1519] eta 0:09:13 lr 0.000001 time 0.9285 (1.0075) model_time 0.9283 (1.0064) loss 0.6764 (0.8031) grad_norm 10.4368 (8.9530/2.7003) mem 68106MB [2022-12-20 19:43:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][980/1519] eta 0:09:03 lr 0.000001 time 0.9214 (1.0076) model_time 0.9212 (1.0065) loss 0.8006 (0.8034) grad_norm 7.0920 (8.9573/2.6983) mem 68106MB [2022-12-20 19:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9253 (1.0075) model_time 0.9252 (1.0064) loss 0.6826 (0.8033) grad_norm 11.9972 (8.9792/2.7096) mem 68106MB [2022-12-20 19:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9398 (1.0075) model_time 0.9397 (1.0064) loss 1.2553 (0.8037) grad_norm 13.1130 (8.9991/2.7196) mem 68106MB [2022-12-20 19:43:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9301 (1.0075) model_time 0.9299 (1.0064) loss 0.6829 (0.8037) grad_norm 8.3327 (9.0615/2.8726) mem 68106MB [2022-12-20 19:44:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1020/1519] eta 0:08:22 lr 0.000001 time 0.9265 (1.0075) model_time 0.9263 (1.0064) loss 0.7812 (0.8034) grad_norm 8.3148 (9.0627/2.8611) mem 68106MB [2022-12-20 19:44:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1030/1519] eta 0:08:12 lr 0.000001 time 0.9227 (1.0074) model_time 0.9225 (1.0063) loss 0.6932 (0.8027) grad_norm 7.5508 (9.0660/2.8510) mem 68106MB [2022-12-20 19:44:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1040/1519] eta 0:08:02 lr 0.000001 time 0.9255 (1.0073) model_time 0.9254 (1.0063) loss 1.0101 (0.8032) grad_norm 8.2579 (9.0755/2.8462) mem 68106MB [2022-12-20 19:44:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1050/1519] eta 0:07:52 lr 0.000001 time 0.9271 (1.0075) model_time 0.9270 (1.0064) loss 0.7754 (0.8031) grad_norm 6.9004 (9.0816/2.8442) mem 68106MB [2022-12-20 19:44:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1060/1519] eta 0:07:42 lr 0.000001 time 0.9313 (1.0074) model_time 0.9312 (1.0064) loss 0.9797 (0.8034) grad_norm 7.7694 (9.0721/2.8421) mem 68106MB [2022-12-20 19:44:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1070/1519] eta 0:07:32 lr 0.000001 time 0.9322 (1.0074) model_time 0.9321 (1.0063) loss 0.7574 (0.8032) grad_norm 8.6770 (9.0399/2.7841) mem 68106MB [2022-12-20 19:45:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1080/1519] eta 0:07:22 lr 0.000001 time 0.9087 (1.0073) model_time 0.9085 (1.0063) loss 0.6618 (0.8027) grad_norm 13.3081 (9.0452/2.7915) mem 68106MB [2022-12-20 19:45:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1090/1519] eta 0:07:12 lr 0.000001 time 0.9228 (1.0073) model_time 0.9227 (1.0063) loss 0.6549 (0.8025) grad_norm 7.3447 (9.0159/2.7924) mem 68106MB [2022-12-20 19:45:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1100/1519] eta 0:07:02 lr 0.000001 time 0.9233 (1.0073) model_time 0.9232 (1.0063) loss 1.0093 (0.8026) grad_norm 9.3993 (9.0319/2.7840) mem 68106MB [2022-12-20 19:45:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9196 (1.0073) model_time 0.9194 (1.0062) loss 0.8997 (0.8028) grad_norm 12.6402 (9.0513/2.7879) mem 68106MB [2022-12-20 19:45:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9198 (1.0074) model_time 0.9196 (1.0064) loss 0.9646 (0.8029) grad_norm 7.6514 (9.0278/2.7850) mem 68106MB [2022-12-20 19:45:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9251 (1.0074) model_time 0.9249 (1.0064) loss 1.0448 (0.8028) grad_norm 7.5904 (9.0296/2.7978) mem 68106MB [2022-12-20 19:46:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9238 (1.0074) model_time 0.9237 (1.0064) loss 0.6564 (0.8025) grad_norm 6.7203 (9.0082/2.8046) mem 68106MB [2022-12-20 19:46:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9309 (1.0075) model_time 0.9306 (1.0065) loss 0.8809 (0.8031) grad_norm 10.7230 (9.0135/2.8087) mem 68106MB [2022-12-20 19:46:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9829 (1.0078) model_time 0.9827 (1.0069) loss 0.7024 (0.8031) grad_norm 11.4630 (9.0175/2.8041) mem 68106MB [2022-12-20 19:46:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9256 (1.0079) model_time 0.9254 (1.0069) loss 0.7016 (0.8032) grad_norm 9.1409 (9.0276/2.7995) mem 68106MB [2022-12-20 19:46:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9186 (1.0078) model_time 0.9184 (1.0068) loss 0.8132 (0.8029) grad_norm 8.8413 (9.0070/2.7899) mem 68106MB [2022-12-20 19:46:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1190/1519] eta 0:05:31 lr 0.000001 time 0.9293 (1.0079) model_time 0.9291 (1.0069) loss 0.9017 (0.8026) grad_norm 6.4926 (8.9906/2.7921) mem 68106MB [2022-12-20 19:47:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1200/1519] eta 0:05:21 lr 0.000001 time 0.9266 (1.0078) model_time 0.9265 (1.0068) loss 0.8591 (0.8028) grad_norm 9.5713 (8.9772/2.7681) mem 68106MB [2022-12-20 19:47:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1210/1519] eta 0:05:11 lr 0.000001 time 0.9285 (1.0077) model_time 0.9284 (1.0068) loss 0.8190 (0.8027) grad_norm 8.4325 (8.9766/2.7660) mem 68106MB [2022-12-20 19:47:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1220/1519] eta 0:05:01 lr 0.000001 time 0.8916 (1.0077) model_time 0.8914 (1.0067) loss 0.9689 (0.8026) grad_norm 7.3713 (9.0111/2.7648) mem 68106MB [2022-12-20 19:47:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1230/1519] eta 0:04:51 lr 0.000001 time 0.9243 (1.0077) model_time 0.9242 (1.0068) loss 0.9347 (0.8029) grad_norm 7.7984 (9.0375/2.7734) mem 68106MB [2022-12-20 19:47:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1240/1519] eta 0:04:41 lr 0.000001 time 0.9226 (1.0076) model_time 0.9225 (1.0067) loss 0.7169 (0.8029) grad_norm 9.1271 (9.0487/2.7743) mem 68106MB [2022-12-20 19:47:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1250/1519] eta 0:04:31 lr 0.000001 time 0.9216 (1.0076) model_time 0.9214 (1.0067) loss 1.1583 (0.8027) grad_norm 6.8309 (9.0512/2.7693) mem 68106MB [2022-12-20 19:48:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9368 (1.0076) model_time 0.9367 (1.0066) loss 0.6900 (0.8022) grad_norm 5.6096 (9.0643/2.8182) mem 68106MB [2022-12-20 19:48:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9231 (1.0075) model_time 0.9230 (1.0066) loss 0.6706 (0.8022) grad_norm 7.8495 (9.0754/2.8167) mem 68106MB [2022-12-20 19:48:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9252 (1.0075) model_time 0.9250 (1.0065) loss 0.6711 (0.8020) grad_norm 7.5094 (9.0918/2.8216) mem 68106MB [2022-12-20 19:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9361 (1.0074) model_time 0.9359 (1.0065) loss 1.2294 (0.8019) grad_norm 8.4846 (9.1217/2.8101) mem 68106MB [2022-12-20 19:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9247 (1.0074) model_time 0.9246 (1.0065) loss 0.7526 (0.8019) grad_norm 8.9379 (9.1186/2.8341) mem 68106MB [2022-12-20 19:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9245 (1.0074) model_time 0.9243 (1.0065) loss 0.7391 (0.8020) grad_norm 5.2077 (9.1170/2.8517) mem 68106MB [2022-12-20 19:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9299 (1.0073) model_time 0.9297 (1.0064) loss 0.8394 (0.8017) grad_norm 8.0679 (9.1057/2.8332) mem 68106MB [2022-12-20 19:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9280 (1.0073) model_time 0.9278 (1.0064) loss 0.7762 (0.8017) grad_norm 8.1064 (9.1114/2.8408) mem 68106MB [2022-12-20 19:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9942 (1.0073) model_time 0.9940 (1.0064) loss 0.9304 (0.8015) grad_norm 12.1944 (9.1188/2.8470) mem 68106MB [2022-12-20 19:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9182 (1.0072) model_time 0.9180 (1.0063) loss 0.6776 (0.8012) grad_norm 12.8058 (9.0953/2.8517) mem 68106MB [2022-12-20 19:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1360/1519] eta 0:02:40 lr 0.000001 time 0.9243 (1.0072) model_time 0.9241 (1.0063) loss 0.7431 (0.8014) grad_norm 7.9326 (9.1245/2.8801) mem 68106MB [2022-12-20 19:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1370/1519] eta 0:02:30 lr 0.000001 time 0.9161 (1.0072) model_time 0.9159 (1.0063) loss 0.7231 (0.8013) grad_norm 6.1369 (9.1179/2.8914) mem 68106MB [2022-12-20 19:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9240 (1.0071) model_time 0.9238 (1.0062) loss 0.7671 (0.8011) grad_norm 11.5036 (9.0452/2.7900) mem 68106MB [2022-12-20 19:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9300 (1.0071) model_time 0.9299 (1.0062) loss 0.8637 (0.8013) grad_norm 7.7197 (9.0370/2.7926) mem 68106MB [2022-12-20 19:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9198 (1.0071) model_time 0.9197 (1.0062) loss 0.8787 (0.8012) grad_norm 7.1796 (9.0643/2.8088) mem 68106MB [2022-12-20 19:50:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9311 (1.0070) model_time 0.9309 (1.0062) loss 1.1173 (0.8012) grad_norm 9.3552 (9.0813/2.8035) mem 68106MB [2022-12-20 19:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9362 (1.0070) model_time 0.9361 (1.0061) loss 0.7425 (0.8011) grad_norm 7.3717 (9.0704/2.7804) mem 68106MB [2022-12-20 19:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9291 (1.0070) model_time 0.9290 (1.0061) loss 0.8693 (0.8011) grad_norm 9.9638 (9.0709/2.7853) mem 68106MB [2022-12-20 19:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9272 (1.0070) model_time 0.9270 (1.0061) loss 0.8445 (0.8012) grad_norm 8.8842 (9.0797/2.7842) mem 68106MB [2022-12-20 19:51:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9237 (1.0069) model_time 0.9236 (1.0061) loss 0.6783 (0.8015) grad_norm 6.4690 (9.0688/2.7940) mem 68106MB [2022-12-20 19:51:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9449 (1.0070) model_time 0.9447 (1.0061) loss 0.6794 (0.8013) grad_norm 7.3052 (9.0356/2.7491) mem 68106MB [2022-12-20 19:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9279 (1.0072) model_time 0.9278 (1.0064) loss 0.6882 (0.8016) grad_norm 8.3137 (9.0222/2.7390) mem 68106MB [2022-12-20 19:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9211 (1.0072) model_time 0.9209 (1.0063) loss 0.7663 (0.8015) grad_norm 7.4017 (8.9204/2.2243) mem 68106MB [2022-12-20 19:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9308 (1.0071) model_time 0.9305 (1.0063) loss 0.7046 (0.8016) grad_norm 7.5274 (8.9186/2.2190) mem 68106MB [2022-12-20 19:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1500/1519] eta 0:00:19 lr 0.000001 time 1.0060 (1.0071) model_time 1.0059 (1.0063) loss 0.7141 (0.8019) grad_norm 6.8197 (8.9061/2.2220) mem 68106MB [2022-12-20 19:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [89/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9251 (1.0071) model_time 0.9250 (1.0062) loss 0.7437 (0.8018) grad_norm 6.2258 (8.9067/2.2099) mem 68106MB [2022-12-20 19:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 89 training takes 0:25:29 [2022-12-20 19:52:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_89.pth saving...... [2022-12-20 19:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_89.pth saved !!! [2022-12-20 19:52:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.683 (0.683) Loss 0.5405 (0.5405) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 19:52:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.294 (0.332) Loss 0.5368 (0.5096) Acc@1 92.361 (92.866) Acc@5 98.264 (98.453) Mem 68106MB [2022-12-20 19:52:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.315) Loss 0.4878 (0.5050) Acc@1 91.319 (92.808) Acc@5 98.958 (98.429) Mem 68106MB [2022-12-20 19:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.310) Loss 0.6393 (0.5122) Acc@1 90.625 (92.552) Acc@5 97.917 (98.410) Mem 68106MB [2022-12-20 19:52:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.307) Loss 0.4591 (0.5028) Acc@1 94.097 (92.607) Acc@5 99.306 (98.509) Mem 68106MB [2022-12-20 19:53:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.299 (0.306) Loss 0.4910 (0.5002) Acc@1 92.361 (92.661) Acc@5 99.653 (98.563) Mem 68106MB [2022-12-20 19:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.305) Loss 0.5117 (0.5000) Acc@1 90.972 (92.589) Acc@5 98.264 (98.549) Mem 68106MB [2022-12-20 19:53:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.304) Loss 0.5453 (0.5013) Acc@1 93.403 (92.542) Acc@5 97.917 (98.533) Mem 68106MB [2022-12-20 19:53:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.293 (0.303) Loss 0.4321 (0.4999) Acc@1 93.403 (92.575) Acc@5 98.264 (98.564) Mem 68106MB [2022-12-20 19:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:89] * Acc@1 92.543 Acc@5 98.563 [2022-12-20 19:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.5% [2022-12-20 19:53:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 19:53:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][0/1519] eta 0:47:44 lr 0.000001 time 1.8859 (1.8859) model_time 1.1835 (1.1835) loss 0.7528 (0.7528) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 19:53:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][10/1519] eta 0:27:11 lr 0.000001 time 0.9269 (1.0810) model_time 0.9268 (1.0168) loss 0.7810 (0.7075) grad_norm 9.6060 (8.0893/1.1244) mem 68106MB [2022-12-20 19:53:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][20/1519] eta 0:26:07 lr 0.000001 time 0.9333 (1.0457) model_time 0.9331 (1.0119) loss 0.6640 (0.7428) grad_norm 6.1640 (8.0304/1.5658) mem 68106MB [2022-12-20 19:53:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][30/1519] eta 0:25:35 lr 0.000001 time 0.9280 (1.0310) model_time 0.9279 (1.0080) loss 1.0664 (0.7424) grad_norm 6.7799 (8.5501/2.3702) mem 68106MB [2022-12-20 19:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][40/1519] eta 0:25:15 lr 0.000001 time 0.9334 (1.0249) model_time 0.9333 (1.0075) loss 0.6834 (0.7596) grad_norm 8.1041 (8.5949/2.4221) mem 68106MB [2022-12-20 19:54:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][50/1519] eta 0:24:59 lr 0.000001 time 0.9320 (1.0205) model_time 0.9319 (1.0065) loss 0.8084 (0.7844) grad_norm 8.1934 (8.5868/2.2958) mem 68106MB [2022-12-20 19:54:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][60/1519] eta 0:24:44 lr 0.000001 time 0.9262 (1.0173) model_time 0.9260 (1.0055) loss 0.7837 (0.7795) grad_norm 7.3228 (8.7851/2.2786) mem 68106MB [2022-12-20 19:54:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][70/1519] eta 0:24:31 lr 0.000001 time 0.9313 (1.0159) model_time 0.9311 (1.0057) loss 1.1247 (0.7868) grad_norm 8.1080 (8.7288/2.2280) mem 68106MB [2022-12-20 19:54:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][80/1519] eta 0:24:19 lr 0.000001 time 0.9271 (1.0140) model_time 0.9269 (1.0050) loss 0.9198 (0.7934) grad_norm 6.2315 (8.7937/2.3331) mem 68106MB [2022-12-20 19:54:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][90/1519] eta 0:24:08 lr 0.000001 time 0.9843 (1.0138) model_time 0.9842 (1.0057) loss 0.7553 (0.7926) grad_norm 6.8899 (8.6428/2.2715) mem 68106MB [2022-12-20 19:54:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][100/1519] eta 0:24:00 lr 0.000001 time 1.1493 (1.0149) model_time 1.1491 (1.0076) loss 1.0942 (0.7928) grad_norm 7.4802 (8.6954/2.3380) mem 68106MB [2022-12-20 19:55:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][110/1519] eta 0:23:48 lr 0.000001 time 0.9162 (1.0139) model_time 0.9160 (1.0073) loss 0.6914 (0.7991) grad_norm 8.5338 (8.6944/2.2366) mem 68106MB [2022-12-20 19:55:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][120/1519] eta 0:23:38 lr 0.000001 time 0.9296 (1.0137) model_time 0.9295 (1.0076) loss 0.7909 (0.7976) grad_norm 7.4596 (8.5470/2.2012) mem 68106MB [2022-12-20 19:55:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][130/1519] eta 0:23:27 lr 0.000001 time 0.9304 (1.0132) model_time 0.9303 (1.0075) loss 0.8993 (0.7922) grad_norm 8.2731 (8.5263/2.1221) mem 68106MB [2022-12-20 19:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][140/1519] eta 0:23:19 lr 0.000001 time 0.9232 (1.0146) model_time 0.9230 (1.0093) loss 0.8214 (0.7906) grad_norm 6.1977 (8.4564/2.0718) mem 68106MB [2022-12-20 19:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][150/1519] eta 0:23:09 lr 0.000001 time 1.0342 (1.0150) model_time 1.0341 (1.0100) loss 0.6626 (0.7900) grad_norm 10.6147 (8.5626/2.1915) mem 68106MB [2022-12-20 19:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][160/1519] eta 0:22:59 lr 0.000001 time 0.9439 (1.0148) model_time 0.9437 (1.0101) loss 0.6739 (0.7938) grad_norm 6.4489 (8.4449/2.1744) mem 68106MB [2022-12-20 19:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][170/1519] eta 0:22:47 lr 0.000001 time 0.9222 (1.0140) model_time 0.9220 (1.0096) loss 0.8417 (0.7929) grad_norm 8.5493 (8.4156/2.1230) mem 68106MB [2022-12-20 19:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][180/1519] eta 0:22:36 lr 0.000001 time 0.9206 (1.0133) model_time 0.9205 (1.0091) loss 0.8584 (0.7939) grad_norm 10.0040 (8.5254/2.2242) mem 68106MB [2022-12-20 19:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][190/1519] eta 0:22:26 lr 0.000001 time 0.9261 (1.0130) model_time 0.9259 (1.0090) loss 0.6690 (0.7929) grad_norm 12.3599 (8.5339/2.2147) mem 68106MB [2022-12-20 19:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][200/1519] eta 0:22:16 lr 0.000001 time 0.9224 (1.0130) model_time 0.9222 (1.0092) loss 0.7455 (0.7918) grad_norm 12.1335 (8.5348/2.2041) mem 68106MB [2022-12-20 19:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][210/1519] eta 0:22:05 lr 0.000001 time 0.9238 (1.0123) model_time 0.9237 (1.0087) loss 0.8910 (0.7930) grad_norm 8.7438 (8.5364/2.1639) mem 68106MB [2022-12-20 19:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][220/1519] eta 0:21:54 lr 0.000001 time 0.9258 (1.0119) model_time 0.9256 (1.0085) loss 1.0091 (0.7955) grad_norm 6.2202 (8.5461/2.1577) mem 68106MB [2022-12-20 19:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][230/1519] eta 0:21:43 lr 0.000001 time 0.9306 (1.0116) model_time 0.9305 (1.0083) loss 0.6670 (0.7938) grad_norm 10.2599 (8.5282/2.1308) mem 68106MB [2022-12-20 19:57:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][240/1519] eta 0:21:33 lr 0.000001 time 0.9264 (1.0114) model_time 0.9262 (1.0082) loss 0.6736 (0.7910) grad_norm 6.6705 (8.5021/2.0961) mem 68106MB [2022-12-20 19:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][250/1519] eta 0:21:23 lr 0.000001 time 0.9234 (1.0114) model_time 0.9232 (1.0083) loss 0.8061 (0.7920) grad_norm 8.7697 (8.4887/2.0837) mem 68106MB [2022-12-20 19:57:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][260/1519] eta 0:21:13 lr 0.000001 time 0.9076 (1.0113) model_time 0.9075 (1.0083) loss 0.8447 (0.7924) grad_norm 6.8570 (8.4783/2.0705) mem 68106MB [2022-12-20 19:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][270/1519] eta 0:21:03 lr 0.000001 time 1.0017 (1.0113) model_time 1.0015 (1.0085) loss 0.6980 (0.7927) grad_norm 9.8588 (8.5211/2.0520) mem 68106MB [2022-12-20 19:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][280/1519] eta 0:20:52 lr 0.000001 time 1.0183 (1.0112) model_time 1.0181 (1.0084) loss 0.6603 (0.7902) grad_norm 7.4447 (8.4892/2.0249) mem 68106MB [2022-12-20 19:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][290/1519] eta 0:20:42 lr 0.000001 time 0.9242 (1.0112) model_time 0.9241 (1.0085) loss 0.6757 (0.7916) grad_norm 7.4334 (8.5658/2.0853) mem 68106MB [2022-12-20 19:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][300/1519] eta 0:20:32 lr 0.000001 time 0.9243 (1.0108) model_time 0.9241 (1.0082) loss 0.9188 (0.7918) grad_norm 9.4270 (8.5787/2.0806) mem 68106MB [2022-12-20 19:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][310/1519] eta 0:20:21 lr 0.000001 time 0.9196 (1.0105) model_time 0.9195 (1.0079) loss 0.7416 (0.7902) grad_norm 10.4031 (8.6248/2.1073) mem 68106MB [2022-12-20 19:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][320/1519] eta 0:20:11 lr 0.000001 time 0.9265 (1.0104) model_time 0.9263 (1.0080) loss 1.0051 (0.7919) grad_norm 13.4241 (8.6232/2.1292) mem 68106MB [2022-12-20 19:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][330/1519] eta 0:20:00 lr 0.000001 time 0.9228 (1.0101) model_time 0.9226 (1.0077) loss 0.6626 (0.7897) grad_norm 6.8264 (8.6018/2.1063) mem 68106MB [2022-12-20 19:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][340/1519] eta 0:19:50 lr 0.000001 time 0.9262 (1.0099) model_time 0.9261 (1.0075) loss 0.7863 (0.7913) grad_norm 18.6074 (8.6634/2.2242) mem 68106MB [2022-12-20 19:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][350/1519] eta 0:19:40 lr 0.000001 time 0.9233 (1.0096) model_time 0.9231 (1.0073) loss 0.7822 (0.7912) grad_norm 5.5625 (8.7037/2.5172) mem 68106MB [2022-12-20 19:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][360/1519] eta 0:19:29 lr 0.000001 time 0.9233 (1.0092) model_time 0.9232 (1.0070) loss 0.6621 (0.7925) grad_norm 7.6147 (8.7053/2.4974) mem 68106MB [2022-12-20 19:59:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][370/1519] eta 0:19:19 lr 0.000001 time 0.9193 (1.0089) model_time 0.9191 (1.0067) loss 0.8455 (0.7935) grad_norm 7.6857 (8.7133/2.4753) mem 68106MB [2022-12-20 19:59:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][380/1519] eta 0:19:08 lr 0.000001 time 0.9353 (1.0087) model_time 0.9352 (1.0065) loss 0.7663 (0.7942) grad_norm 6.4508 (8.7027/2.4600) mem 68106MB [2022-12-20 19:59:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][390/1519] eta 0:18:58 lr 0.000001 time 0.9339 (1.0085) model_time 0.9338 (1.0064) loss 1.0232 (0.7937) grad_norm 8.1316 (8.7046/2.4317) mem 68106MB [2022-12-20 19:59:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][400/1519] eta 0:18:48 lr 0.000001 time 0.9192 (1.0087) model_time 0.9190 (1.0066) loss 0.8257 (0.7933) grad_norm 9.9137 (8.7286/2.4179) mem 68106MB [2022-12-20 20:00:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][410/1519] eta 0:18:38 lr 0.000001 time 0.9351 (1.0084) model_time 0.9349 (1.0064) loss 0.9745 (0.7945) grad_norm 12.0832 (8.7721/2.4118) mem 68106MB [2022-12-20 20:00:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][420/1519] eta 0:18:28 lr 0.000001 time 0.9234 (1.0082) model_time 0.9233 (1.0063) loss 1.1601 (0.7942) grad_norm 8.3258 (8.8070/2.4512) mem 68106MB [2022-12-20 20:00:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][430/1519] eta 0:18:18 lr 0.000001 time 1.2187 (1.0089) model_time 1.2186 (1.0069) loss 0.9228 (0.7938) grad_norm 9.4880 (8.8104/2.4313) mem 68106MB [2022-12-20 20:00:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][440/1519] eta 0:18:08 lr 0.000001 time 0.8909 (1.0087) model_time 0.8907 (1.0068) loss 0.7138 (0.7938) grad_norm 9.0355 (8.8329/2.4343) mem 68106MB [2022-12-20 20:00:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][450/1519] eta 0:17:58 lr 0.000001 time 0.9202 (1.0085) model_time 0.9201 (1.0067) loss 0.7062 (0.7926) grad_norm 9.8199 (8.8393/2.4620) mem 68106MB [2022-12-20 20:00:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][460/1519] eta 0:17:48 lr 0.000001 time 0.9273 (1.0085) model_time 0.9272 (1.0067) loss 0.6856 (0.7942) grad_norm 6.9843 (8.8379/2.4613) mem 68106MB [2022-12-20 20:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][470/1519] eta 0:17:38 lr 0.000001 time 0.9256 (1.0086) model_time 0.9254 (1.0069) loss 0.7520 (0.7955) grad_norm 9.3838 (8.8355/2.4445) mem 68106MB [2022-12-20 20:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][480/1519] eta 0:17:27 lr 0.000001 time 0.9231 (1.0084) model_time 0.9229 (1.0067) loss 0.8820 (0.7949) grad_norm 5.8172 (8.8097/2.4350) mem 68106MB [2022-12-20 20:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][490/1519] eta 0:17:17 lr 0.000001 time 0.9248 (1.0083) model_time 0.9247 (1.0065) loss 0.7791 (0.7953) grad_norm 7.4676 (8.7916/2.4156) mem 68106MB [2022-12-20 20:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][500/1519] eta 0:17:07 lr 0.000001 time 0.9378 (1.0081) model_time 0.9376 (1.0064) loss 0.6866 (0.7969) grad_norm 6.8612 (8.7790/2.4062) mem 68106MB [2022-12-20 20:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][510/1519] eta 0:16:57 lr 0.000001 time 0.9078 (1.0081) model_time 0.9076 (1.0064) loss 0.7548 (0.7974) grad_norm 7.5064 (8.7850/2.3870) mem 68106MB [2022-12-20 20:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][520/1519] eta 0:16:47 lr 0.000001 time 0.9348 (1.0080) model_time 0.9347 (1.0064) loss 0.7933 (0.7978) grad_norm 9.9236 (8.7961/2.3718) mem 68106MB [2022-12-20 20:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][530/1519] eta 0:16:36 lr 0.000001 time 0.9293 (1.0080) model_time 0.9291 (1.0063) loss 0.8112 (0.7979) grad_norm 12.8613 (8.8393/2.4312) mem 68106MB [2022-12-20 20:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][540/1519] eta 0:16:26 lr 0.000001 time 0.9316 (1.0080) model_time 0.9315 (1.0064) loss 0.9357 (0.7982) grad_norm 12.3698 (8.8467/2.4299) mem 68106MB [2022-12-20 20:02:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][550/1519] eta 0:16:16 lr 0.000001 time 0.9296 (1.0079) model_time 0.9294 (1.0063) loss 0.8008 (0.7983) grad_norm 10.7706 (8.8348/2.4201) mem 68106MB [2022-12-20 20:02:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][560/1519] eta 0:16:06 lr 0.000001 time 0.9148 (1.0079) model_time 0.9147 (1.0064) loss 0.6545 (0.7974) grad_norm 7.6880 (8.8649/2.4399) mem 68106MB [2022-12-20 20:02:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][570/1519] eta 0:15:56 lr 0.000001 time 0.9300 (1.0079) model_time 0.9299 (1.0063) loss 0.7350 (0.7977) grad_norm 8.8929 (8.8746/2.4247) mem 68106MB [2022-12-20 20:02:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][580/1519] eta 0:15:46 lr 0.000001 time 0.9225 (1.0078) model_time 0.9224 (1.0063) loss 0.7256 (0.7976) grad_norm 8.7934 (8.8677/2.4064) mem 68106MB [2022-12-20 20:03:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][590/1519] eta 0:15:36 lr 0.000001 time 0.9289 (1.0078) model_time 0.9288 (1.0063) loss 0.6901 (0.7974) grad_norm 8.0053 (8.8541/2.3904) mem 68106MB [2022-12-20 20:03:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][600/1519] eta 0:15:26 lr 0.000001 time 0.9218 (1.0079) model_time 0.9216 (1.0065) loss 0.7841 (0.7975) grad_norm 9.1905 (8.8331/2.3848) mem 68106MB [2022-12-20 20:03:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][610/1519] eta 0:15:16 lr 0.000001 time 0.9893 (1.0079) model_time 0.9892 (1.0064) loss 0.8521 (0.7995) grad_norm 8.8132 (8.8314/2.3845) mem 68106MB [2022-12-20 20:03:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9267 (1.0078) model_time 0.9266 (1.0063) loss 0.9733 (0.8004) grad_norm 7.8327 (8.8324/2.3716) mem 68106MB [2022-12-20 20:03:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9209 (1.0079) model_time 0.9207 (1.0065) loss 0.7364 (0.8000) grad_norm 7.2190 (8.8105/2.3373) mem 68106MB [2022-12-20 20:03:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][640/1519] eta 0:14:45 lr 0.000001 time 0.9401 (1.0078) model_time 0.9400 (1.0064) loss 0.8030 (0.7994) grad_norm 5.1335 (8.7777/2.3319) mem 68106MB [2022-12-20 20:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][650/1519] eta 0:14:35 lr 0.000001 time 0.9065 (1.0080) model_time 0.9063 (1.0066) loss 0.7037 (0.7991) grad_norm 7.7621 (8.7661/2.3245) mem 68106MB [2022-12-20 20:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][660/1519] eta 0:14:25 lr 0.000001 time 0.9169 (1.0078) model_time 0.9168 (1.0065) loss 0.9270 (0.7990) grad_norm 9.4154 (8.7481/2.3096) mem 68106MB [2022-12-20 20:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][670/1519] eta 0:14:15 lr 0.000001 time 0.9219 (1.0077) model_time 0.9217 (1.0063) loss 0.8956 (0.7994) grad_norm 10.4917 (8.7517/2.3014) mem 68106MB [2022-12-20 20:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][680/1519] eta 0:14:05 lr 0.000001 time 0.9250 (1.0076) model_time 0.9248 (1.0062) loss 0.7861 (0.7998) grad_norm 7.1632 (8.7314/2.2769) mem 68106MB [2022-12-20 20:04:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][690/1519] eta 0:13:55 lr 0.000001 time 0.9291 (1.0075) model_time 0.9290 (1.0062) loss 0.7073 (0.7993) grad_norm 11.1967 (8.7711/2.2824) mem 68106MB [2022-12-20 20:04:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][700/1519] eta 0:13:45 lr 0.000001 time 0.9373 (1.0074) model_time 0.9371 (1.0061) loss 0.6671 (0.7984) grad_norm 8.3055 (8.7384/2.2650) mem 68106MB [2022-12-20 20:05:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][710/1519] eta 0:13:35 lr 0.000001 time 0.9110 (1.0076) model_time 0.9108 (1.0063) loss 1.0684 (0.7999) grad_norm 10.5420 (8.7466/2.2744) mem 68106MB [2022-12-20 20:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][720/1519] eta 0:13:25 lr 0.000001 time 0.9374 (1.0077) model_time 0.9373 (1.0064) loss 0.7552 (0.8003) grad_norm 6.1732 (8.7557/2.2703) mem 68106MB [2022-12-20 20:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][730/1519] eta 0:13:14 lr 0.000001 time 0.9200 (1.0076) model_time 0.9199 (1.0063) loss 0.7830 (0.8003) grad_norm 7.4364 (8.7653/2.2967) mem 68106MB [2022-12-20 20:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9491 (1.0075) model_time 0.9490 (1.0063) loss 0.8086 (0.7999) grad_norm 8.1048 (8.7880/2.2965) mem 68106MB [2022-12-20 20:05:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][750/1519] eta 0:12:54 lr 0.000001 time 0.9305 (1.0074) model_time 0.9303 (1.0062) loss 0.6653 (0.8004) grad_norm 17.5220 (8.7857/2.3234) mem 68106MB [2022-12-20 20:05:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9205 (1.0074) model_time 0.9204 (1.0061) loss 0.6639 (0.7998) grad_norm 7.2884 (8.8011/2.3131) mem 68106MB [2022-12-20 20:06:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][770/1519] eta 0:12:34 lr 0.000001 time 0.9260 (1.0074) model_time 0.9259 (1.0061) loss 0.8338 (0.8002) grad_norm 7.4500 (8.8506/2.3422) mem 68106MB [2022-12-20 20:06:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][780/1519] eta 0:12:24 lr 0.000001 time 0.9230 (1.0074) model_time 0.9229 (1.0062) loss 0.8351 (0.8001) grad_norm 15.7158 (8.8508/2.3379) mem 68106MB [2022-12-20 20:06:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][790/1519] eta 0:12:14 lr 0.000001 time 0.9218 (1.0073) model_time 0.9217 (1.0061) loss 0.8783 (0.7997) grad_norm 6.4555 (8.8433/2.3276) mem 68106MB [2022-12-20 20:06:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][800/1519] eta 0:12:04 lr 0.000001 time 0.9214 (1.0072) model_time 0.9213 (1.0060) loss 0.9443 (0.7994) grad_norm 10.0732 (8.8429/2.3262) mem 68106MB [2022-12-20 20:06:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][810/1519] eta 0:11:54 lr 0.000001 time 0.9723 (1.0071) model_time 0.9720 (1.0060) loss 0.9563 (0.7993) grad_norm 8.5370 (8.8379/2.3256) mem 68106MB [2022-12-20 20:06:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][820/1519] eta 0:11:43 lr 0.000001 time 0.9240 (1.0070) model_time 0.9239 (1.0059) loss 0.7408 (0.7988) grad_norm 6.5242 (8.8429/2.3334) mem 68106MB [2022-12-20 20:07:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9246 (1.0070) model_time 0.9244 (1.0058) loss 0.6810 (0.7986) grad_norm 8.5250 (8.8809/2.3395) mem 68106MB [2022-12-20 20:07:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9309 (1.0069) model_time 0.9308 (1.0057) loss 0.6896 (0.7991) grad_norm 8.8504 (8.9100/2.3524) mem 68106MB [2022-12-20 20:07:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][850/1519] eta 0:11:13 lr 0.000001 time 0.9340 (1.0068) model_time 0.9338 (1.0056) loss 0.6950 (0.7983) grad_norm 6.6832 (8.9199/2.3622) mem 68106MB [2022-12-20 20:07:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][860/1519] eta 0:11:03 lr 0.000001 time 0.9269 (1.0068) model_time 0.9267 (1.0056) loss 0.6932 (0.7986) grad_norm 9.0813 (8.9369/2.3668) mem 68106MB [2022-12-20 20:07:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][870/1519] eta 0:10:53 lr 0.000001 time 0.9225 (1.0067) model_time 0.9223 (1.0056) loss 0.7867 (0.7985) grad_norm 7.2183 (8.9176/2.3717) mem 68106MB [2022-12-20 20:07:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][880/1519] eta 0:10:43 lr 0.000001 time 0.8934 (1.0067) model_time 0.8930 (1.0056) loss 0.8585 (0.7983) grad_norm 6.0137 (8.9163/2.3929) mem 68106MB [2022-12-20 20:08:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][890/1519] eta 0:10:33 lr 0.000001 time 0.9277 (1.0067) model_time 0.9276 (1.0056) loss 1.0401 (0.7992) grad_norm 9.9587 (8.8784/2.3615) mem 68106MB [2022-12-20 20:08:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][900/1519] eta 0:10:23 lr 0.000001 time 0.9331 (1.0067) model_time 0.9330 (1.0056) loss 0.9648 (0.8002) grad_norm 9.9084 (8.8756/2.3616) mem 68106MB [2022-12-20 20:08:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][910/1519] eta 0:10:13 lr 0.000001 time 0.9291 (1.0067) model_time 0.9290 (1.0056) loss 0.8073 (0.8001) grad_norm 8.3485 (8.8340/2.3478) mem 68106MB [2022-12-20 20:08:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][920/1519] eta 0:10:02 lr 0.000001 time 0.9333 (1.0067) model_time 0.9332 (1.0056) loss 0.6739 (0.8002) grad_norm 9.6286 (8.8429/2.3347) mem 68106MB [2022-12-20 20:08:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][930/1519] eta 0:09:52 lr 0.000001 time 0.9265 (1.0067) model_time 0.9263 (1.0056) loss 0.9900 (0.8005) grad_norm 6.5289 (8.8384/2.3358) mem 68106MB [2022-12-20 20:08:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9530 (1.0066) model_time 0.9529 (1.0056) loss 0.7062 (0.7997) grad_norm 10.4585 (8.8083/2.2668) mem 68106MB [2022-12-20 20:09:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][950/1519] eta 0:09:32 lr 0.000001 time 0.9197 (1.0066) model_time 0.9196 (1.0055) loss 1.1393 (0.8007) grad_norm 10.8039 (8.7866/2.0712) mem 68106MB [2022-12-20 20:09:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9187 (1.0066) model_time 0.9185 (1.0056) loss 0.7075 (0.8011) grad_norm 10.0283 (8.7972/2.0677) mem 68106MB [2022-12-20 20:09:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][970/1519] eta 0:09:12 lr 0.000001 time 0.9224 (1.0066) model_time 0.9223 (1.0055) loss 0.7918 (0.8010) grad_norm 7.3287 (8.7819/2.0748) mem 68106MB [2022-12-20 20:09:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9217 (1.0065) model_time 0.9216 (1.0055) loss 0.8638 (0.8012) grad_norm 8.1776 (8.7898/2.0682) mem 68106MB [2022-12-20 20:09:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9860 (1.0065) model_time 0.9858 (1.0055) loss 0.6970 (0.8014) grad_norm 7.2283 (8.7946/2.0850) mem 68106MB [2022-12-20 20:09:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9219 (1.0065) model_time 0.9218 (1.0054) loss 0.8423 (0.8011) grad_norm 7.0574 (8.7663/2.0809) mem 68106MB [2022-12-20 20:10:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9211 (1.0064) model_time 0.9210 (1.0054) loss 0.8349 (0.8007) grad_norm 8.7233 (8.7180/2.0692) mem 68106MB [2022-12-20 20:10:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1020/1519] eta 0:08:22 lr 0.000001 time 0.9624 (1.0063) model_time 0.9622 (1.0053) loss 1.0578 (0.8013) grad_norm 9.0900 (8.6876/2.0155) mem 68106MB [2022-12-20 20:10:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1030/1519] eta 0:08:12 lr 0.000001 time 0.9140 (1.0064) model_time 0.9137 (1.0054) loss 0.6851 (0.8010) grad_norm 7.4430 (8.6732/2.0148) mem 68106MB [2022-12-20 20:10:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1040/1519] eta 0:08:02 lr 0.000001 time 0.9253 (1.0064) model_time 0.9252 (1.0054) loss 0.8243 (0.8012) grad_norm 11.2206 (8.6478/2.0062) mem 68106MB [2022-12-20 20:10:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1050/1519] eta 0:07:51 lr 0.000001 time 0.9389 (1.0063) model_time 0.9388 (1.0054) loss 0.6752 (0.8008) grad_norm 11.5779 (8.6420/1.9659) mem 68106MB [2022-12-20 20:10:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1060/1519] eta 0:07:41 lr 0.000001 time 0.9268 (1.0064) model_time 0.9267 (1.0054) loss 0.7220 (0.8010) grad_norm 7.1046 (8.6192/1.9530) mem 68106MB [2022-12-20 20:11:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1070/1519] eta 0:07:31 lr 0.000001 time 0.9234 (1.0064) model_time 0.9233 (1.0055) loss 0.6722 (0.8008) grad_norm 17.1241 (8.6355/2.0139) mem 68106MB [2022-12-20 20:11:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1080/1519] eta 0:07:21 lr 0.000001 time 1.0118 (1.0064) model_time 1.0116 (1.0055) loss 0.7004 (0.8005) grad_norm 7.3307 (8.6359/2.0083) mem 68106MB [2022-12-20 20:11:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9378 (1.0065) model_time 0.9377 (1.0055) loss 0.9505 (0.8007) grad_norm 8.1999 (8.6338/2.0137) mem 68106MB [2022-12-20 20:11:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9333 (1.0064) model_time 0.9331 (1.0055) loss 0.7456 (0.8005) grad_norm 6.2600 (8.6331/2.0068) mem 68106MB [2022-12-20 20:11:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9277 (1.0064) model_time 0.9276 (1.0054) loss 0.7067 (0.8004) grad_norm 8.9914 (8.6403/2.0157) mem 68106MB [2022-12-20 20:11:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9313 (1.0064) model_time 0.9312 (1.0054) loss 0.6696 (0.8001) grad_norm 10.5891 (8.6332/2.0118) mem 68106MB [2022-12-20 20:12:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9292 (1.0063) model_time 0.9290 (1.0053) loss 0.7379 (0.7997) grad_norm 7.1200 (8.6081/1.9786) mem 68106MB [2022-12-20 20:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9301 (1.0063) model_time 0.9299 (1.0054) loss 0.8039 (0.7999) grad_norm 7.9488 (8.6037/1.9668) mem 68106MB [2022-12-20 20:12:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9333 (1.0063) model_time 0.9331 (1.0053) loss 1.0026 (0.8001) grad_norm 8.4303 (8.6081/1.9595) mem 68106MB [2022-12-20 20:12:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9229 (1.0062) model_time 0.9228 (1.0053) loss 1.0542 (0.8003) grad_norm 10.9263 (8.5850/1.9270) mem 68106MB [2022-12-20 20:12:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9251 (1.0061) model_time 0.9249 (1.0052) loss 0.7169 (0.8006) grad_norm 8.2454 (8.5742/1.9203) mem 68106MB [2022-12-20 20:12:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9205 (1.0061) model_time 0.9203 (1.0052) loss 0.7291 (0.8001) grad_norm 8.3981 (8.5792/1.9238) mem 68106MB [2022-12-20 20:13:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1190/1519] eta 0:05:30 lr 0.000001 time 0.9227 (1.0061) model_time 0.9225 (1.0052) loss 0.9984 (0.8008) grad_norm 9.9827 (8.5784/1.9324) mem 68106MB [2022-12-20 20:13:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1200/1519] eta 0:05:20 lr 0.000001 time 0.9654 (1.0060) model_time 0.9653 (1.0051) loss 0.6694 (0.8004) grad_norm 10.5977 (8.5947/1.9288) mem 68106MB [2022-12-20 20:13:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1210/1519] eta 0:05:10 lr 0.000001 time 0.9297 (1.0060) model_time 0.9296 (1.0051) loss 0.8895 (0.8003) grad_norm 6.4390 (8.5882/1.9353) mem 68106MB [2022-12-20 20:13:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1220/1519] eta 0:05:00 lr 0.000001 time 0.9269 (1.0061) model_time 0.9267 (1.0052) loss 0.6802 (0.7998) grad_norm 7.5189 (8.6140/1.9765) mem 68106MB [2022-12-20 20:13:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1230/1519] eta 0:04:50 lr 0.000001 time 0.9109 (1.0060) model_time 0.9107 (1.0051) loss 0.6614 (0.7997) grad_norm 7.7847 (8.6356/2.0351) mem 68106MB [2022-12-20 20:13:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9301 (1.0061) model_time 0.9298 (1.0052) loss 0.7170 (0.7997) grad_norm 9.7110 (8.7032/2.0647) mem 68106MB [2022-12-20 20:14:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9342 (1.0061) model_time 0.9341 (1.0052) loss 0.9908 (0.8008) grad_norm 7.0130 (8.7074/2.0847) mem 68106MB [2022-12-20 20:14:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1260/1519] eta 0:04:20 lr 0.000001 time 1.0192 (1.0062) model_time 1.0191 (1.0053) loss 1.0838 (0.8007) grad_norm 9.3948 (8.6955/2.0908) mem 68106MB [2022-12-20 20:14:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9241 (1.0063) model_time 0.9239 (1.0054) loss 0.7667 (0.8006) grad_norm 13.1402 (8.7030/2.1083) mem 68106MB [2022-12-20 20:14:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9329 (1.0062) model_time 0.9328 (1.0054) loss 0.8155 (0.8005) grad_norm 17.5521 (8.7357/2.1646) mem 68106MB [2022-12-20 20:14:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9249 (1.0062) model_time 0.9248 (1.0053) loss 0.9843 (0.8013) grad_norm 7.0257 (8.7866/2.3748) mem 68106MB [2022-12-20 20:15:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9272 (1.0062) model_time 0.9271 (1.0053) loss 0.6956 (0.8014) grad_norm 6.8664 (8.8201/2.3859) mem 68106MB [2022-12-20 20:15:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9245 (1.0062) model_time 0.9243 (1.0053) loss 0.9844 (0.8014) grad_norm 7.8327 (8.8007/2.3795) mem 68106MB [2022-12-20 20:15:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9372 (1.0061) model_time 0.9370 (1.0053) loss 0.8027 (0.8013) grad_norm 8.0623 (8.8264/2.3888) mem 68106MB [2022-12-20 20:15:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9233 (1.0061) model_time 0.9232 (1.0053) loss 0.6919 (0.8017) grad_norm 7.7999 (8.8235/2.3710) mem 68106MB [2022-12-20 20:15:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9187 (1.0062) model_time 0.9185 (1.0054) loss 0.7541 (0.8017) grad_norm 6.5867 (8.8101/2.3886) mem 68106MB [2022-12-20 20:15:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9123 (1.0062) model_time 0.9122 (1.0054) loss 1.1944 (0.8025) grad_norm 11.2662 (8.7888/2.3379) mem 68106MB [2022-12-20 20:16:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1360/1519] eta 0:02:39 lr 0.000001 time 0.9055 (1.0062) model_time 0.9054 (1.0054) loss 0.8954 (0.8026) grad_norm 8.6312 (8.7892/2.3425) mem 68106MB [2022-12-20 20:16:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1370/1519] eta 0:02:29 lr 0.000001 time 0.9197 (1.0062) model_time 0.9196 (1.0053) loss 0.6641 (0.8025) grad_norm 7.7000 (8.7392/2.3143) mem 68106MB [2022-12-20 20:16:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9070 (1.0063) model_time 0.9069 (1.0054) loss 0.8414 (0.8026) grad_norm 6.1431 (8.6936/2.2871) mem 68106MB [2022-12-20 20:16:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9234 (1.0063) model_time 0.9232 (1.0054) loss 1.0670 (0.8026) grad_norm 11.3249 (8.6914/2.2959) mem 68106MB [2022-12-20 20:16:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9146 (1.0064) model_time 0.9144 (1.0056) loss 0.7750 (0.8033) grad_norm 7.1505 (8.7160/2.3074) mem 68106MB [2022-12-20 20:16:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9274 (1.0063) model_time 0.9272 (1.0055) loss 0.7813 (0.8034) grad_norm 8.7791 (8.7399/2.3112) mem 68106MB [2022-12-20 20:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9329 (1.0063) model_time 0.9328 (1.0054) loss 0.9094 (0.8037) grad_norm 7.5138 (8.7326/2.3431) mem 68106MB [2022-12-20 20:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9320 (1.0062) model_time 0.9318 (1.0054) loss 0.8151 (0.8038) grad_norm 7.8582 (8.7112/2.3411) mem 68106MB [2022-12-20 20:17:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9405 (1.0063) model_time 0.9401 (1.0054) loss 0.9746 (0.8040) grad_norm 12.6392 (8.6994/2.3431) mem 68106MB [2022-12-20 20:17:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9225 (1.0062) model_time 0.9223 (1.0054) loss 0.7116 (0.8039) grad_norm 7.4475 (8.6912/2.3292) mem 68106MB [2022-12-20 20:17:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9394 (1.0062) model_time 0.9390 (1.0054) loss 0.6797 (0.8040) grad_norm 7.1891 (8.6977/2.3195) mem 68106MB [2022-12-20 20:17:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9506 (1.0062) model_time 0.9504 (1.0054) loss 0.7020 (0.8040) grad_norm 10.2839 (8.7128/2.3418) mem 68106MB [2022-12-20 20:18:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9326 (1.0062) model_time 0.9325 (1.0054) loss 0.6640 (0.8034) grad_norm 10.0382 (8.7293/2.3220) mem 68106MB [2022-12-20 20:18:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9321 (1.0061) model_time 0.9319 (1.0053) loss 0.7304 (0.8033) grad_norm 10.2784 (8.7379/2.3257) mem 68106MB [2022-12-20 20:18:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9399 (1.0061) model_time 0.9397 (1.0053) loss 0.7033 (0.8039) grad_norm 6.7184 (8.7218/2.3192) mem 68106MB [2022-12-20 20:18:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [90/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9237 (1.0060) model_time 0.9236 (1.0052) loss 0.6758 (0.8039) grad_norm 9.4504 (8.7496/2.3140) mem 68106MB [2022-12-20 20:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 90 training takes 0:25:28 [2022-12-20 20:18:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_90.pth saving...... [2022-12-20 20:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_90.pth saved !!! [2022-12-20 20:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.674 (0.674) Loss 0.5374 (0.5374) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 20:19:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.302 (0.332) Loss 0.5348 (0.5074) Acc@1 92.361 (92.898) Acc@5 98.611 (98.516) Mem 68106MB [2022-12-20 20:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.303 (0.316) Loss 0.4848 (0.5030) Acc@1 90.972 (92.808) Acc@5 99.306 (98.512) Mem 68106MB [2022-12-20 20:19:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.300 (0.311) Loss 0.6356 (0.5099) Acc@1 90.972 (92.529) Acc@5 98.264 (98.466) Mem 68106MB [2022-12-20 20:19:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.308) Loss 0.4585 (0.5006) Acc@1 93.750 (92.615) Acc@5 99.306 (98.560) Mem 68106MB [2022-12-20 20:19:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.306) Loss 0.4911 (0.4979) Acc@1 92.014 (92.674) Acc@5 99.653 (98.604) Mem 68106MB [2022-12-20 20:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.299 (0.305) Loss 0.5094 (0.4977) Acc@1 90.972 (92.594) Acc@5 98.264 (98.583) Mem 68106MB [2022-12-20 20:19:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.304) Loss 0.5435 (0.4990) Acc@1 92.708 (92.557) Acc@5 98.264 (98.572) Mem 68106MB [2022-12-20 20:19:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.295 (0.303) Loss 0.4270 (0.4977) Acc@1 93.056 (92.597) Acc@5 98.264 (98.598) Mem 68106MB [2022-12-20 20:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:90] * Acc@1 92.567 Acc@5 98.600 [2022-12-20 20:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 20:19:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 20:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][0/1519] eta 0:46:46 lr 0.000001 time 1.8476 (1.8476) model_time 1.1719 (1.1719) loss 0.7035 (0.7035) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 20:19:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][10/1519] eta 0:27:16 lr 0.000001 time 0.9246 (1.0843) model_time 0.9244 (1.0222) loss 0.9836 (0.8254) grad_norm 9.1178 (8.5406/1.6927) mem 68106MB [2022-12-20 20:19:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][20/1519] eta 0:26:04 lr 0.000001 time 0.9305 (1.0438) model_time 0.9304 (1.0111) loss 0.6607 (0.7717) grad_norm 7.9134 (8.7100/2.4606) mem 68106MB [2022-12-20 20:20:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][30/1519] eta 0:25:48 lr 0.000001 time 0.9194 (1.0397) model_time 0.9193 (1.0173) loss 0.7187 (0.7795) grad_norm 14.4214 (8.9476/2.4986) mem 68106MB [2022-12-20 20:20:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][40/1519] eta 0:25:26 lr 0.000001 time 0.9216 (1.0318) model_time 0.9213 (1.0148) loss 0.8221 (0.7706) grad_norm 7.0721 (8.5542/2.3369) mem 68106MB [2022-12-20 20:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][50/1519] eta 0:25:07 lr 0.000001 time 0.9264 (1.0259) model_time 0.9262 (1.0122) loss 0.7388 (0.7649) grad_norm 6.9506 (8.5608/2.1627) mem 68106MB [2022-12-20 20:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][60/1519] eta 0:24:52 lr 0.000001 time 0.9201 (1.0231) model_time 0.9200 (1.0115) loss 0.8892 (0.7741) grad_norm 9.8242 (8.7079/2.3239) mem 68106MB [2022-12-20 20:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][70/1519] eta 0:24:38 lr 0.000001 time 0.9229 (1.0202) model_time 0.9228 (1.0102) loss 0.8031 (0.7859) grad_norm 10.1070 (8.5701/2.2729) mem 68106MB [2022-12-20 20:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][80/1519] eta 0:24:24 lr 0.000001 time 0.9308 (1.0180) model_time 0.9305 (1.0092) loss 0.6680 (0.7859) grad_norm 8.6089 (8.6080/2.1814) mem 68106MB [2022-12-20 20:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][90/1519] eta 0:24:14 lr 0.000001 time 0.9344 (1.0178) model_time 0.9342 (1.0098) loss 0.6944 (0.7890) grad_norm 10.3825 (8.6984/2.0982) mem 68106MB [2022-12-20 20:21:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][100/1519] eta 0:24:04 lr 0.000001 time 0.9808 (1.0182) model_time 0.9805 (1.0110) loss 0.8767 (0.8045) grad_norm 11.0500 (8.6995/2.0596) mem 68106MB [2022-12-20 20:21:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][110/1519] eta 0:23:53 lr 0.000001 time 0.9311 (1.0174) model_time 0.9310 (1.0108) loss 0.6699 (0.8027) grad_norm 9.0407 (8.8330/2.0645) mem 68106MB [2022-12-20 20:21:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][120/1519] eta 0:23:41 lr 0.000001 time 0.9280 (1.0160) model_time 0.9279 (1.0100) loss 0.6922 (0.7968) grad_norm 6.7841 (8.7806/2.0030) mem 68106MB [2022-12-20 20:21:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][130/1519] eta 0:23:30 lr 0.000001 time 0.9341 (1.0153) model_time 0.9340 (1.0097) loss 0.7770 (0.7972) grad_norm 9.5359 (8.7831/1.9552) mem 68106MB [2022-12-20 20:21:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][140/1519] eta 0:23:18 lr 0.000001 time 0.9236 (1.0145) model_time 0.9234 (1.0092) loss 0.9142 (0.8016) grad_norm 9.0182 (8.7778/1.9016) mem 68106MB [2022-12-20 20:22:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][150/1519] eta 0:23:07 lr 0.000001 time 0.9453 (1.0136) model_time 0.9452 (1.0087) loss 0.7653 (0.7987) grad_norm 6.8051 (8.7652/1.8796) mem 68106MB [2022-12-20 20:22:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][160/1519] eta 0:22:56 lr 0.000001 time 0.9285 (1.0131) model_time 0.9283 (1.0085) loss 0.9066 (0.8019) grad_norm 6.4798 (8.6356/1.8901) mem 68106MB [2022-12-20 20:22:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][170/1519] eta 0:22:45 lr 0.000001 time 0.9211 (1.0123) model_time 0.9210 (1.0079) loss 0.7589 (0.8020) grad_norm 11.9848 (8.6471/1.9428) mem 68106MB [2022-12-20 20:22:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][180/1519] eta 0:22:34 lr 0.000001 time 0.9248 (1.0116) model_time 0.9246 (1.0075) loss 0.6912 (0.8036) grad_norm 7.6649 (8.6976/1.9396) mem 68106MB [2022-12-20 20:22:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][190/1519] eta 0:22:24 lr 0.000001 time 0.9201 (1.0118) model_time 0.9200 (1.0079) loss 0.8209 (0.8031) grad_norm 16.9808 (8.7153/2.1029) mem 68106MB [2022-12-20 20:22:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][200/1519] eta 0:22:14 lr 0.000001 time 0.9199 (1.0115) model_time 0.9198 (1.0078) loss 0.8390 (0.8012) grad_norm 9.3778 (8.7437/2.0674) mem 68106MB [2022-12-20 20:23:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][210/1519] eta 0:22:03 lr 0.000001 time 0.9381 (1.0112) model_time 0.9379 (1.0076) loss 0.7750 (0.7999) grad_norm 7.1052 (8.7748/2.1448) mem 68106MB [2022-12-20 20:23:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][220/1519] eta 0:21:53 lr 0.000001 time 0.9218 (1.0108) model_time 0.9216 (1.0073) loss 0.8972 (0.8016) grad_norm 6.7763 (8.8266/2.2021) mem 68106MB [2022-12-20 20:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][230/1519] eta 0:21:42 lr 0.000001 time 0.9264 (1.0108) model_time 0.9262 (1.0075) loss 0.6766 (0.8000) grad_norm 7.5620 (8.7878/2.1748) mem 68106MB [2022-12-20 20:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][240/1519] eta 0:21:32 lr 0.000001 time 0.9262 (1.0106) model_time 0.9260 (1.0074) loss 0.6665 (0.8016) grad_norm 10.9105 (8.8234/2.1420) mem 68106MB [2022-12-20 20:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][250/1519] eta 0:21:22 lr 0.000001 time 0.9260 (1.0104) model_time 0.9259 (1.0073) loss 0.8842 (0.8023) grad_norm 7.9066 (8.8540/2.1450) mem 68106MB [2022-12-20 20:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][260/1519] eta 0:21:11 lr 0.000001 time 0.9231 (1.0100) model_time 0.9230 (1.0070) loss 0.7520 (0.8023) grad_norm 8.1671 (8.8391/2.1170) mem 68106MB [2022-12-20 20:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][270/1519] eta 0:21:00 lr 0.000001 time 0.9341 (1.0096) model_time 0.9339 (1.0067) loss 0.7708 (0.8014) grad_norm 11.2259 (8.9541/2.2360) mem 68106MB [2022-12-20 20:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][280/1519] eta 0:20:50 lr 0.000001 time 0.9227 (1.0092) model_time 0.9226 (1.0064) loss 0.6662 (0.7993) grad_norm 7.6273 (9.0454/2.3585) mem 68106MB [2022-12-20 20:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][290/1519] eta 0:20:40 lr 0.000001 time 0.9242 (1.0090) model_time 0.9241 (1.0063) loss 0.6901 (0.8003) grad_norm 8.5005 (9.0251/2.3246) mem 68106MB [2022-12-20 20:24:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][300/1519] eta 0:20:29 lr 0.000001 time 0.9284 (1.0088) model_time 0.9283 (1.0061) loss 0.7905 (0.8007) grad_norm 9.3834 (8.9917/2.2975) mem 68106MB [2022-12-20 20:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][310/1519] eta 0:20:19 lr 0.000001 time 0.9313 (1.0087) model_time 0.9312 (1.0062) loss 0.7018 (0.8014) grad_norm 7.0589 (8.9494/2.3028) mem 68106MB [2022-12-20 20:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][320/1519] eta 0:20:09 lr 0.000001 time 0.8923 (1.0085) model_time 0.8921 (1.0061) loss 0.8344 (0.8010) grad_norm 9.9803 (8.9346/2.2811) mem 68106MB [2022-12-20 20:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][330/1519] eta 0:19:58 lr 0.000001 time 0.9209 (1.0083) model_time 0.9208 (1.0059) loss 0.7090 (0.8003) grad_norm 6.8938 (8.9109/2.2606) mem 68106MB [2022-12-20 20:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][340/1519] eta 0:19:48 lr 0.000001 time 0.9164 (1.0082) model_time 0.9163 (1.0059) loss 0.9750 (0.8009) grad_norm 9.4836 (8.8833/2.2420) mem 68106MB [2022-12-20 20:25:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][350/1519] eta 0:19:38 lr 0.000001 time 0.9213 (1.0085) model_time 0.9212 (1.0062) loss 0.6622 (0.7987) grad_norm 8.9316 (8.8738/2.2233) mem 68106MB [2022-12-20 20:25:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][360/1519] eta 0:19:28 lr 0.000001 time 0.9284 (1.0082) model_time 0.9283 (1.0060) loss 1.0069 (0.8011) grad_norm 5.9121 (8.8595/2.2236) mem 68106MB [2022-12-20 20:25:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][370/1519] eta 0:19:18 lr 0.000001 time 0.9276 (1.0083) model_time 0.9275 (1.0061) loss 0.7192 (0.8007) grad_norm 7.2665 (8.8627/2.2160) mem 68106MB [2022-12-20 20:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][380/1519] eta 0:19:08 lr 0.000001 time 0.9260 (1.0083) model_time 0.9258 (1.0062) loss 0.9367 (0.8024) grad_norm 8.4436 (8.8494/2.1970) mem 68106MB [2022-12-20 20:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][390/1519] eta 0:18:58 lr 0.000001 time 0.9210 (1.0080) model_time 0.9208 (1.0060) loss 0.7135 (0.8029) grad_norm 10.2969 (8.8455/2.1894) mem 68106MB [2022-12-20 20:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][400/1519] eta 0:18:48 lr 0.000001 time 0.9068 (1.0083) model_time 0.9067 (1.0063) loss 0.6814 (0.8029) grad_norm 7.7188 (8.8274/2.1715) mem 68106MB [2022-12-20 20:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][410/1519] eta 0:18:38 lr 0.000001 time 0.9189 (1.0086) model_time 0.9188 (1.0066) loss 0.7926 (0.8030) grad_norm 8.1035 (8.9228/2.6554) mem 68106MB [2022-12-20 20:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][420/1519] eta 0:18:28 lr 0.000001 time 0.9314 (1.0088) model_time 0.9313 (1.0069) loss 0.8749 (0.8035) grad_norm 8.6338 (8.9377/2.6445) mem 68106MB [2022-12-20 20:26:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][430/1519] eta 0:18:18 lr 0.000001 time 0.9257 (1.0090) model_time 0.9256 (1.0071) loss 0.6647 (0.8039) grad_norm 14.8497 (8.9695/2.6525) mem 68106MB [2022-12-20 20:26:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][440/1519] eta 0:18:08 lr 0.000001 time 0.9252 (1.0089) model_time 0.9250 (1.0070) loss 0.7797 (0.8038) grad_norm 5.5692 (8.9655/2.6989) mem 68106MB [2022-12-20 20:27:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][450/1519] eta 0:17:58 lr 0.000001 time 0.9274 (1.0088) model_time 0.9273 (1.0069) loss 1.0229 (0.8039) grad_norm 8.0250 (8.9632/2.6757) mem 68106MB [2022-12-20 20:27:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][460/1519] eta 0:17:48 lr 0.000001 time 0.9214 (1.0087) model_time 0.9212 (1.0069) loss 0.7090 (0.8031) grad_norm 12.8248 (8.9548/2.6678) mem 68106MB [2022-12-20 20:27:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][470/1519] eta 0:17:37 lr 0.000001 time 0.9248 (1.0086) model_time 0.9246 (1.0068) loss 0.7994 (0.8040) grad_norm 11.1319 (8.9443/2.6583) mem 68106MB [2022-12-20 20:27:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][480/1519] eta 0:17:27 lr 0.000001 time 0.9227 (1.0083) model_time 0.9226 (1.0066) loss 0.7910 (0.8040) grad_norm 8.9443 (8.9245/2.6363) mem 68106MB [2022-12-20 20:27:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][490/1519] eta 0:17:17 lr 0.000001 time 0.9208 (1.0083) model_time 0.9207 (1.0065) loss 0.8078 (0.8033) grad_norm 16.5057 (8.9410/2.6586) mem 68106MB [2022-12-20 20:27:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][500/1519] eta 0:17:07 lr 0.000001 time 0.9295 (1.0081) model_time 0.9292 (1.0064) loss 0.7293 (0.8021) grad_norm 10.4759 (8.9489/2.6349) mem 68106MB [2022-12-20 20:28:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][510/1519] eta 0:16:56 lr 0.000001 time 0.9286 (1.0079) model_time 0.9284 (1.0063) loss 0.7321 (0.8012) grad_norm 7.8821 (8.9159/2.6201) mem 68106MB [2022-12-20 20:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][520/1519] eta 0:16:46 lr 0.000001 time 0.9103 (1.0079) model_time 0.9102 (1.0062) loss 0.8920 (0.8026) grad_norm 7.3390 (8.8863/2.6066) mem 68106MB [2022-12-20 20:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][530/1519] eta 0:16:36 lr 0.000001 time 0.9233 (1.0077) model_time 0.9232 (1.0061) loss 0.9260 (0.8016) grad_norm 9.3779 (8.8906/2.5898) mem 68106MB [2022-12-20 20:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][540/1519] eta 0:16:26 lr 0.000001 time 0.9211 (1.0075) model_time 0.9210 (1.0059) loss 0.7744 (0.8014) grad_norm 6.7933 (8.8703/2.5733) mem 68106MB [2022-12-20 20:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][550/1519] eta 0:16:16 lr 0.000001 time 0.9265 (1.0076) model_time 0.9264 (1.0060) loss 0.7289 (0.8024) grad_norm 6.9142 (8.8666/2.5585) mem 68106MB [2022-12-20 20:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][560/1519] eta 0:16:06 lr 0.000001 time 0.9244 (1.0075) model_time 0.9243 (1.0060) loss 0.8256 (0.8019) grad_norm 7.0871 (8.8430/2.5474) mem 68106MB [2022-12-20 20:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][570/1519] eta 0:15:56 lr 0.000001 time 0.9211 (1.0074) model_time 0.9210 (1.0059) loss 0.7001 (0.8021) grad_norm 9.1182 (8.8535/2.5355) mem 68106MB [2022-12-20 20:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][580/1519] eta 0:15:45 lr 0.000001 time 0.9262 (1.0073) model_time 0.9261 (1.0058) loss 0.7672 (0.8020) grad_norm 7.2080 (8.8751/2.5293) mem 68106MB [2022-12-20 20:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][590/1519] eta 0:15:35 lr 0.000001 time 0.9184 (1.0071) model_time 0.9183 (1.0057) loss 0.8297 (0.8019) grad_norm 14.5649 (8.8855/2.5433) mem 68106MB [2022-12-20 20:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][600/1519] eta 0:15:25 lr 0.000001 time 0.9240 (1.0070) model_time 0.9239 (1.0056) loss 0.6853 (0.8020) grad_norm 5.8654 (8.8977/2.5602) mem 68106MB [2022-12-20 20:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][610/1519] eta 0:15:15 lr 0.000001 time 0.9243 (1.0070) model_time 0.9242 (1.0055) loss 0.7623 (0.8010) grad_norm 9.9372 (8.8986/2.5634) mem 68106MB [2022-12-20 20:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9303 (1.0069) model_time 0.9301 (1.0055) loss 0.6611 (0.8013) grad_norm 6.5661 (8.8980/2.5454) mem 68106MB [2022-12-20 20:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9261 (1.0068) model_time 0.9259 (1.0054) loss 0.7274 (0.8012) grad_norm 8.2252 (8.8568/2.5378) mem 68106MB [2022-12-20 20:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][640/1519] eta 0:14:44 lr 0.000001 time 0.9254 (1.0067) model_time 0.9252 (1.0053) loss 0.7272 (0.8007) grad_norm 11.4687 (8.9076/2.5430) mem 68106MB [2022-12-20 20:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][650/1519] eta 0:14:34 lr 0.000001 time 0.9300 (1.0066) model_time 0.9298 (1.0052) loss 0.8190 (0.8000) grad_norm 7.2167 (8.9272/2.5561) mem 68106MB [2022-12-20 20:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][660/1519] eta 0:14:24 lr 0.000001 time 0.9267 (1.0067) model_time 0.9266 (1.0054) loss 1.0077 (0.7998) grad_norm 9.5579 (8.9069/2.5347) mem 68106MB [2022-12-20 20:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][670/1519] eta 0:14:14 lr 0.000001 time 0.9232 (1.0066) model_time 0.9231 (1.0053) loss 0.8250 (0.7994) grad_norm 9.6054 (8.9153/2.5253) mem 68106MB [2022-12-20 20:30:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][680/1519] eta 0:14:04 lr 0.000001 time 0.9860 (1.0068) model_time 0.9859 (1.0054) loss 0.6905 (0.7992) grad_norm 8.0814 (8.9005/2.5229) mem 68106MB [2022-12-20 20:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][690/1519] eta 0:13:54 lr 0.000001 time 0.9330 (1.0068) model_time 0.9328 (1.0055) loss 0.8306 (0.8000) grad_norm 6.9212 (8.8627/2.5301) mem 68106MB [2022-12-20 20:31:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][700/1519] eta 0:13:44 lr 0.000001 time 0.9172 (1.0066) model_time 0.9171 (1.0053) loss 0.7325 (0.8008) grad_norm 7.3577 (8.8495/2.5267) mem 68106MB [2022-12-20 20:31:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][710/1519] eta 0:13:34 lr 0.000001 time 0.9278 (1.0067) model_time 0.9277 (1.0054) loss 0.6770 (0.8007) grad_norm 8.0488 (8.8144/2.5150) mem 68106MB [2022-12-20 20:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][720/1519] eta 0:13:24 lr 0.000001 time 0.9070 (1.0067) model_time 0.9069 (1.0055) loss 1.1184 (0.8006) grad_norm 8.0715 (8.8125/2.5162) mem 68106MB [2022-12-20 20:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][730/1519] eta 0:13:14 lr 0.000001 time 0.9266 (1.0068) model_time 0.9265 (1.0056) loss 1.0373 (0.8014) grad_norm 8.7586 (8.7929/2.5211) mem 68106MB [2022-12-20 20:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9288 (1.0068) model_time 0.9287 (1.0056) loss 0.7155 (0.8018) grad_norm 8.7179 (8.8269/2.5663) mem 68106MB [2022-12-20 20:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][750/1519] eta 0:12:54 lr 0.000001 time 0.9298 (1.0068) model_time 0.9296 (1.0056) loss 0.7066 (0.8010) grad_norm 9.6249 (8.8257/2.5621) mem 68106MB [2022-12-20 20:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9367 (1.0068) model_time 0.9366 (1.0056) loss 0.9390 (0.8016) grad_norm 6.1239 (8.8565/2.5569) mem 68106MB [2022-12-20 20:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][770/1519] eta 0:12:34 lr 0.000001 time 0.9304 (1.0068) model_time 0.9303 (1.0055) loss 0.6959 (0.8011) grad_norm 9.7458 (8.8461/2.5463) mem 68106MB [2022-12-20 20:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9292 (1.0067) model_time 0.9291 (1.0055) loss 1.0598 (0.8002) grad_norm 10.3655 (8.8185/2.5445) mem 68106MB [2022-12-20 20:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9226 (1.0066) model_time 0.9225 (1.0054) loss 0.7988 (0.8005) grad_norm 8.0026 (8.8242/2.4959) mem 68106MB [2022-12-20 20:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][800/1519] eta 0:12:03 lr 0.000001 time 0.9083 (1.0067) model_time 0.9081 (1.0055) loss 0.8097 (0.8010) grad_norm 7.3212 (8.8340/2.5312) mem 68106MB [2022-12-20 20:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][810/1519] eta 0:11:53 lr 0.000001 time 0.9283 (1.0067) model_time 0.9282 (1.0055) loss 0.8517 (0.8016) grad_norm 6.8645 (8.8218/2.5059) mem 68106MB [2022-12-20 20:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][820/1519] eta 0:11:43 lr 0.000001 time 0.9308 (1.0066) model_time 0.9306 (1.0054) loss 0.8745 (0.8021) grad_norm 7.9568 (8.8427/2.5822) mem 68106MB [2022-12-20 20:33:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9231 (1.0065) model_time 0.9230 (1.0054) loss 1.1145 (0.8029) grad_norm 6.9869 (8.8460/2.5835) mem 68106MB [2022-12-20 20:33:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9356 (1.0065) model_time 0.9355 (1.0054) loss 0.8073 (0.8024) grad_norm 8.9655 (8.8465/2.6084) mem 68106MB [2022-12-20 20:33:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][850/1519] eta 0:11:13 lr 0.000001 time 0.9310 (1.0065) model_time 0.9309 (1.0053) loss 0.6855 (0.8024) grad_norm 7.0323 (8.8310/2.5988) mem 68106MB [2022-12-20 20:33:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][860/1519] eta 0:11:03 lr 0.000001 time 1.0473 (1.0065) model_time 1.0471 (1.0054) loss 0.6962 (0.8024) grad_norm 14.0646 (8.8448/2.6226) mem 68106MB [2022-12-20 20:34:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][870/1519] eta 0:10:53 lr 0.000001 time 0.9313 (1.0065) model_time 0.9312 (1.0053) loss 0.8878 (0.8026) grad_norm 10.7051 (8.7838/2.5705) mem 68106MB [2022-12-20 20:34:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][880/1519] eta 0:10:43 lr 0.000001 time 0.9341 (1.0064) model_time 0.9339 (1.0052) loss 0.7737 (0.8026) grad_norm 8.7486 (8.7482/2.5034) mem 68106MB [2022-12-20 20:34:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][890/1519] eta 0:10:32 lr 0.000001 time 0.9223 (1.0063) model_time 0.9222 (1.0052) loss 0.9064 (0.8025) grad_norm 12.2019 (8.7446/2.5173) mem 68106MB [2022-12-20 20:34:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][900/1519] eta 0:10:22 lr 0.000001 time 0.9566 (1.0062) model_time 0.9564 (1.0052) loss 0.6749 (0.8026) grad_norm 5.0953 (8.7597/2.5295) mem 68106MB [2022-12-20 20:34:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][910/1519] eta 0:10:12 lr 0.000001 time 0.9246 (1.0062) model_time 0.9243 (1.0051) loss 0.9593 (0.8024) grad_norm 8.3509 (8.7719/2.5162) mem 68106MB [2022-12-20 20:34:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][920/1519] eta 0:10:02 lr 0.000001 time 0.9305 (1.0061) model_time 0.9304 (1.0051) loss 0.7349 (0.8019) grad_norm 11.0000 (8.7781/2.5179) mem 68106MB [2022-12-20 20:35:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][930/1519] eta 0:09:52 lr 0.000001 time 0.9584 (1.0061) model_time 0.9583 (1.0050) loss 0.7523 (0.8022) grad_norm 8.0446 (8.8123/2.5308) mem 68106MB [2022-12-20 20:35:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9236 (1.0061) model_time 0.9234 (1.0050) loss 0.6639 (0.8026) grad_norm 6.3839 (8.8546/2.5665) mem 68106MB [2022-12-20 20:35:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][950/1519] eta 0:09:32 lr 0.000001 time 0.9225 (1.0061) model_time 0.9224 (1.0050) loss 0.7488 (0.8032) grad_norm 7.0347 (8.8493/2.5654) mem 68106MB [2022-12-20 20:35:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9299 (1.0060) model_time 0.9298 (1.0049) loss 0.9210 (0.8032) grad_norm 7.6314 (8.8437/2.5550) mem 68106MB [2022-12-20 20:35:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][970/1519] eta 0:09:12 lr 0.000001 time 1.0333 (1.0061) model_time 1.0331 (1.0050) loss 1.0829 (0.8037) grad_norm 7.7474 (8.8134/2.5547) mem 68106MB [2022-12-20 20:35:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9271 (1.0060) model_time 0.9269 (1.0050) loss 0.8300 (0.8030) grad_norm 9.5169 (8.7855/2.5717) mem 68106MB [2022-12-20 20:36:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9635 (1.0061) model_time 0.9634 (1.0050) loss 0.6961 (0.8022) grad_norm 8.1875 (8.8047/2.5753) mem 68106MB [2022-12-20 20:36:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9277 (1.0062) model_time 0.9274 (1.0051) loss 0.7090 (0.8022) grad_norm 7.4539 (8.8197/2.5743) mem 68106MB [2022-12-20 20:36:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9342 (1.0061) model_time 0.9341 (1.0051) loss 0.7882 (0.8015) grad_norm 10.5393 (8.7643/2.2272) mem 68106MB [2022-12-20 20:36:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1020/1519] eta 0:08:22 lr 0.000001 time 0.8880 (1.0061) model_time 0.8878 (1.0051) loss 0.8502 (0.8018) grad_norm 8.0811 (8.7424/2.2331) mem 68106MB [2022-12-20 20:36:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1030/1519] eta 0:08:12 lr 0.000001 time 0.9808 (1.0061) model_time 0.9806 (1.0052) loss 0.8051 (0.8018) grad_norm 10.5077 (8.7154/2.2099) mem 68106MB [2022-12-20 20:36:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1040/1519] eta 0:08:01 lr 0.000001 time 0.9297 (1.0062) model_time 0.9296 (1.0053) loss 0.8952 (0.8012) grad_norm 7.5348 (8.6909/2.1561) mem 68106MB [2022-12-20 20:37:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1050/1519] eta 0:07:51 lr 0.000001 time 0.9240 (1.0063) model_time 0.9238 (1.0053) loss 0.8398 (0.8013) grad_norm 7.2008 (8.6944/2.1564) mem 68106MB [2022-12-20 20:37:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1060/1519] eta 0:07:41 lr 0.000001 time 0.9246 (1.0064) model_time 0.9245 (1.0054) loss 1.0667 (0.8009) grad_norm 12.9684 (8.7160/2.1667) mem 68106MB [2022-12-20 20:37:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1070/1519] eta 0:07:31 lr 0.000001 time 0.9229 (1.0063) model_time 0.9228 (1.0053) loss 0.6716 (0.8009) grad_norm 4.8189 (8.7115/2.1707) mem 68106MB [2022-12-20 20:37:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1080/1519] eta 0:07:21 lr 0.000001 time 0.9211 (1.0062) model_time 0.9210 (1.0053) loss 0.8370 (0.8014) grad_norm 11.5682 (8.7264/2.1746) mem 68106MB [2022-12-20 20:37:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9221 (1.0061) model_time 0.9220 (1.0052) loss 0.8665 (0.8018) grad_norm 7.3330 (8.7086/2.1333) mem 68106MB [2022-12-20 20:37:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9281 (1.0061) model_time 0.9280 (1.0051) loss 0.6770 (0.8017) grad_norm 8.4405 (8.7039/2.1343) mem 68106MB [2022-12-20 20:38:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9242 (1.0061) model_time 0.9240 (1.0051) loss 0.7725 (0.8017) grad_norm 9.4598 (8.7461/2.1579) mem 68106MB [2022-12-20 20:38:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9237 (1.0061) model_time 0.9235 (1.0051) loss 0.8477 (0.8018) grad_norm 6.9604 (8.7672/2.1567) mem 68106MB [2022-12-20 20:38:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9203 (1.0061) model_time 0.9201 (1.0052) loss 0.7029 (0.8020) grad_norm 6.4786 (8.7419/2.1611) mem 68106MB [2022-12-20 20:38:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9244 (1.0061) model_time 0.9242 (1.0051) loss 0.7052 (0.8021) grad_norm 9.0225 (8.7473/2.1617) mem 68106MB [2022-12-20 20:38:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1150/1519] eta 0:06:11 lr 0.000001 time 1.1959 (1.0063) model_time 1.1958 (1.0054) loss 0.7881 (0.8019) grad_norm 5.3835 (8.7464/2.1734) mem 68106MB [2022-12-20 20:38:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9230 (1.0063) model_time 0.9229 (1.0054) loss 0.6785 (0.8018) grad_norm 9.5652 (8.7523/2.1769) mem 68106MB [2022-12-20 20:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9185 (1.0063) model_time 0.9183 (1.0053) loss 0.9547 (0.8016) grad_norm 11.7244 (8.7787/2.1882) mem 68106MB [2022-12-20 20:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9206 (1.0063) model_time 0.9205 (1.0054) loss 0.6647 (0.8012) grad_norm 9.1898 (8.7571/2.1784) mem 68106MB [2022-12-20 20:39:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1190/1519] eta 0:05:31 lr 0.000001 time 0.9255 (1.0062) model_time 0.9253 (1.0053) loss 0.7336 (0.8011) grad_norm 13.3803 (8.7487/2.1567) mem 68106MB [2022-12-20 20:39:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1200/1519] eta 0:05:20 lr 0.000001 time 0.9811 (1.0062) model_time 0.9809 (1.0053) loss 0.9169 (0.8011) grad_norm 9.2388 (8.7221/2.1216) mem 68106MB [2022-12-20 20:39:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1210/1519] eta 0:05:10 lr 0.000001 time 1.0094 (1.0063) model_time 1.0092 (1.0054) loss 0.9722 (0.8014) grad_norm 9.9580 (8.7250/2.1163) mem 68106MB [2022-12-20 20:39:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1220/1519] eta 0:05:00 lr 0.000001 time 0.9222 (1.0063) model_time 0.9221 (1.0054) loss 0.7884 (0.8014) grad_norm 8.4007 (8.7026/2.1090) mem 68106MB [2022-12-20 20:40:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1230/1519] eta 0:04:50 lr 0.000001 time 0.9312 (1.0063) model_time 0.9311 (1.0054) loss 0.6737 (0.8016) grad_norm 8.5029 (8.7072/2.1052) mem 68106MB [2022-12-20 20:40:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9279 (1.0063) model_time 0.9277 (1.0054) loss 0.7292 (0.8017) grad_norm 5.9009 (8.6706/2.0966) mem 68106MB [2022-12-20 20:40:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9076 (1.0063) model_time 0.9075 (1.0054) loss 0.6659 (0.8015) grad_norm 10.8643 (8.6744/2.0814) mem 68106MB [2022-12-20 20:40:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9423 (1.0063) model_time 0.9421 (1.0054) loss 0.7058 (0.8013) grad_norm 9.1938 (8.7048/2.1537) mem 68106MB [2022-12-20 20:40:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9204 (1.0062) model_time 0.9203 (1.0053) loss 0.9245 (0.8012) grad_norm 7.6804 (8.7144/2.1710) mem 68106MB [2022-12-20 20:40:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9187 (1.0062) model_time 0.9186 (1.0053) loss 0.6761 (0.8010) grad_norm 8.1848 (8.7256/2.1720) mem 68106MB [2022-12-20 20:41:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9210 (1.0062) model_time 0.9209 (1.0053) loss 0.8749 (0.8015) grad_norm 11.2592 (8.7606/2.1681) mem 68106MB [2022-12-20 20:41:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9760 (1.0062) model_time 0.9759 (1.0053) loss 0.7738 (0.8014) grad_norm 9.9720 (8.7831/2.1633) mem 68106MB [2022-12-20 20:41:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9845 (1.0062) model_time 0.9844 (1.0053) loss 0.9069 (0.8014) grad_norm 11.7684 (8.8063/2.1714) mem 68106MB [2022-12-20 20:41:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9309 (1.0061) model_time 0.9308 (1.0053) loss 0.6885 (0.8010) grad_norm 6.8150 (8.8102/2.1773) mem 68106MB [2022-12-20 20:41:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9204 (1.0061) model_time 0.9203 (1.0053) loss 0.8785 (0.8012) grad_norm 8.0079 (8.8614/2.1939) mem 68106MB [2022-12-20 20:41:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9265 (1.0061) model_time 0.9264 (1.0053) loss 0.8196 (0.8009) grad_norm 7.4471 (8.8349/2.1462) mem 68106MB [2022-12-20 20:42:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9308 (1.0062) model_time 0.9306 (1.0054) loss 0.7779 (0.8010) grad_norm 8.1667 (8.8342/2.1497) mem 68106MB [2022-12-20 20:42:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1360/1519] eta 0:02:39 lr 0.000001 time 0.9337 (1.0063) model_time 0.9336 (1.0054) loss 0.7945 (0.8006) grad_norm 11.7703 (8.8262/2.1575) mem 68106MB [2022-12-20 20:42:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1370/1519] eta 0:02:29 lr 0.000001 time 0.9317 (1.0063) model_time 0.9316 (1.0055) loss 0.7589 (0.8010) grad_norm 9.1364 (8.8343/2.1617) mem 68106MB [2022-12-20 20:42:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9347 (1.0063) model_time 0.9345 (1.0054) loss 1.0820 (0.8012) grad_norm 10.4042 (8.8477/2.1595) mem 68106MB [2022-12-20 20:42:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9191 (1.0062) model_time 0.9190 (1.0054) loss 0.6566 (0.8010) grad_norm 7.6543 (8.8481/2.1600) mem 68106MB [2022-12-20 20:42:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9272 (1.0062) model_time 0.9271 (1.0054) loss 0.8534 (0.8012) grad_norm 11.1360 (8.8271/2.1256) mem 68106MB [2022-12-20 20:43:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9319 (1.0062) model_time 0.9318 (1.0054) loss 0.6905 (0.8013) grad_norm 9.6164 (8.8520/2.1308) mem 68106MB [2022-12-20 20:43:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9562 (1.0062) model_time 0.9561 (1.0054) loss 0.6770 (0.8009) grad_norm 7.8865 (8.8296/2.0092) mem 68106MB [2022-12-20 20:43:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9292 (1.0062) model_time 0.9290 (1.0054) loss 0.7519 (0.8018) grad_norm 11.8450 (8.8391/2.0123) mem 68106MB [2022-12-20 20:43:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9116 (1.0062) model_time 0.9115 (1.0054) loss 0.6688 (0.8018) grad_norm 7.6039 (8.8064/1.9861) mem 68106MB [2022-12-20 20:43:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9403 (1.0062) model_time 0.9402 (1.0054) loss 0.6849 (0.8014) grad_norm 8.1596 (8.8002/1.9867) mem 68106MB [2022-12-20 20:43:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1460/1519] eta 0:00:59 lr 0.000001 time 1.0023 (1.0062) model_time 1.0021 (1.0054) loss 0.6965 (0.8011) grad_norm 6.5307 (8.8022/1.9953) mem 68106MB [2022-12-20 20:44:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9254 (1.0062) model_time 0.9253 (1.0054) loss 1.1745 (0.8011) grad_norm 6.5637 (8.7847/1.9994) mem 68106MB [2022-12-20 20:44:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9852 (1.0062) model_time 0.9851 (1.0054) loss 0.6641 (0.8014) grad_norm 10.3988 (8.7776/1.9969) mem 68106MB [2022-12-20 20:44:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9815 (1.0063) model_time 0.9812 (1.0055) loss 0.6651 (0.8011) grad_norm 6.9564 (8.7864/1.9856) mem 68106MB [2022-12-20 20:44:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9294 (1.0062) model_time 0.9293 (1.0054) loss 0.7339 (0.8009) grad_norm 8.3089 (8.7700/1.9690) mem 68106MB [2022-12-20 20:44:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [91/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9214 (1.0061) model_time 0.9213 (1.0054) loss 0.9619 (0.8009) grad_norm 7.8258 (8.7503/1.9855) mem 68106MB [2022-12-20 20:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 91 training takes 0:25:28 [2022-12-20 20:44:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_91.pth saving...... [2022-12-20 20:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_91.pth saved !!! [2022-12-20 20:45:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.637 (0.637) Loss 0.5403 (0.5403) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 20:45:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.296 (0.327) Loss 0.5347 (0.5081) Acc@1 92.361 (92.771) Acc@5 98.264 (98.485) Mem 68106MB [2022-12-20 20:45:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.294 (0.313) Loss 0.4851 (0.5036) Acc@1 90.972 (92.708) Acc@5 99.306 (98.413) Mem 68106MB [2022-12-20 20:45:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.308) Loss 0.6368 (0.5107) Acc@1 90.972 (92.518) Acc@5 97.917 (98.410) Mem 68106MB [2022-12-20 20:45:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.306) Loss 0.4599 (0.5013) Acc@1 93.750 (92.598) Acc@5 99.306 (98.518) Mem 68106MB [2022-12-20 20:45:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.305) Loss 0.4908 (0.4985) Acc@1 92.361 (92.681) Acc@5 99.653 (98.563) Mem 68106MB [2022-12-20 20:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.302 (0.304) Loss 0.5097 (0.4983) Acc@1 90.972 (92.600) Acc@5 98.264 (98.549) Mem 68106MB [2022-12-20 20:45:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.304 (0.303) Loss 0.5444 (0.4995) Acc@1 93.056 (92.562) Acc@5 98.264 (98.538) Mem 68106MB [2022-12-20 20:45:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.303 (0.302) Loss 0.4310 (0.4982) Acc@1 93.403 (92.597) Acc@5 98.264 (98.568) Mem 68106MB [2022-12-20 20:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:91] * Acc@1 92.571 Acc@5 98.572 [2022-12-20 20:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 20:45:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 20:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 20:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.57% [2022-12-20 20:46:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][0/1519] eta 0:35:01 lr 0.000001 time 1.3838 (1.3838) model_time 0.9363 (0.9363) loss 0.6855 (0.6855) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 20:46:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][10/1519] eta 0:26:28 lr 0.000001 time 0.9933 (1.0525) model_time 0.9931 (1.0115) loss 0.9738 (0.7832) grad_norm 8.5266 (8.6020/1.4671) mem 68106MB [2022-12-20 20:46:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][20/1519] eta 0:25:54 lr 0.000001 time 1.0202 (1.0372) model_time 1.0201 (1.0155) loss 0.8094 (0.8074) grad_norm 6.4131 (8.5279/1.9229) mem 68106MB [2022-12-20 20:46:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][30/1519] eta 0:25:28 lr 0.000001 time 0.9287 (1.0268) model_time 0.9285 (1.0120) loss 0.6951 (0.8080) grad_norm 5.9100 (7.9648/1.8958) mem 68106MB [2022-12-20 20:46:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][40/1519] eta 0:25:09 lr 0.000001 time 0.9206 (1.0206) model_time 0.9205 (1.0094) loss 1.1202 (0.7981) grad_norm 13.6013 (8.1156/2.1204) mem 68106MB [2022-12-20 20:47:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][50/1519] eta 0:24:54 lr 0.000001 time 0.9275 (1.0174) model_time 0.9274 (1.0083) loss 0.9467 (0.8008) grad_norm 9.2130 (8.4024/2.0725) mem 68106MB [2022-12-20 20:47:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][60/1519] eta 0:24:41 lr 0.000001 time 0.9253 (1.0153) model_time 0.9252 (1.0076) loss 0.7261 (0.7944) grad_norm 8.6914 (8.3957/1.9316) mem 68106MB [2022-12-20 20:47:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][70/1519] eta 0:24:27 lr 0.000001 time 0.9311 (1.0130) model_time 0.9309 (1.0064) loss 0.9159 (0.8044) grad_norm 9.3126 (8.2494/1.8986) mem 68106MB [2022-12-20 20:47:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][80/1519] eta 0:24:15 lr 0.000001 time 0.9300 (1.0114) model_time 0.9299 (1.0056) loss 0.7232 (0.8011) grad_norm 6.7811 (8.2068/1.8334) mem 68106MB [2022-12-20 20:47:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][90/1519] eta 0:24:03 lr 0.000001 time 0.9200 (1.0101) model_time 0.9198 (1.0049) loss 0.7557 (0.8029) grad_norm 6.9275 (8.2560/1.7920) mem 68106MB [2022-12-20 20:47:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][100/1519] eta 0:23:52 lr 0.000001 time 0.9392 (1.0093) model_time 0.9391 (1.0045) loss 0.6783 (0.8012) grad_norm 8.3967 (8.3354/1.7281) mem 68106MB [2022-12-20 20:48:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][110/1519] eta 0:23:40 lr 0.000001 time 0.9272 (1.0085) model_time 0.9271 (1.0042) loss 0.7400 (0.7967) grad_norm 8.4577 (8.3722/1.7264) mem 68106MB [2022-12-20 20:48:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][120/1519] eta 0:23:33 lr 0.000001 time 0.9229 (1.0104) model_time 0.9228 (1.0064) loss 0.7192 (0.7995) grad_norm 8.1595 (8.5524/1.9260) mem 68106MB [2022-12-20 20:48:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][130/1519] eta 0:23:22 lr 0.000001 time 0.9775 (1.0099) model_time 0.9773 (1.0062) loss 0.7918 (0.7987) grad_norm 8.5697 (8.5392/1.8687) mem 68106MB [2022-12-20 20:48:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][140/1519] eta 0:23:11 lr 0.000001 time 0.9213 (1.0092) model_time 0.9210 (1.0057) loss 0.6896 (0.7983) grad_norm 7.8194 (8.6151/1.9623) mem 68106MB [2022-12-20 20:48:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][150/1519] eta 0:23:02 lr 0.000001 time 0.9210 (1.0099) model_time 0.9208 (1.0066) loss 0.6800 (0.7941) grad_norm 9.1842 (8.6088/1.9778) mem 68106MB [2022-12-20 20:48:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][160/1519] eta 0:22:51 lr 0.000001 time 0.9231 (1.0095) model_time 0.9229 (1.0064) loss 0.7177 (0.7931) grad_norm 7.7711 (8.5134/1.9666) mem 68106MB [2022-12-20 20:49:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][170/1519] eta 0:22:40 lr 0.000001 time 0.9242 (1.0089) model_time 0.9241 (1.0059) loss 0.7961 (0.7923) grad_norm 6.0424 (8.5575/2.0253) mem 68106MB [2022-12-20 20:49:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][180/1519] eta 0:22:30 lr 0.000001 time 0.9261 (1.0084) model_time 0.9259 (1.0056) loss 0.6939 (0.7973) grad_norm 8.5519 (8.5487/1.9698) mem 68106MB [2022-12-20 20:49:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][190/1519] eta 0:22:20 lr 0.000001 time 1.0093 (1.0086) model_time 1.0092 (1.0060) loss 0.9886 (0.7998) grad_norm 10.3409 (8.5933/1.9748) mem 68106MB [2022-12-20 20:49:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][200/1519] eta 0:22:10 lr 0.000001 time 1.0135 (1.0088) model_time 1.0134 (1.0063) loss 0.7773 (0.7973) grad_norm 9.2180 (8.6201/1.9646) mem 68106MB [2022-12-20 20:49:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][210/1519] eta 0:21:59 lr 0.000001 time 0.9177 (1.0084) model_time 0.9175 (1.0059) loss 0.7314 (0.7999) grad_norm 7.9234 (8.6607/1.9792) mem 68106MB [2022-12-20 20:49:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][220/1519] eta 0:21:49 lr 0.000001 time 0.9177 (1.0081) model_time 0.9175 (1.0058) loss 0.7693 (0.8010) grad_norm 12.7909 (8.6440/1.9999) mem 68106MB [2022-12-20 20:50:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][230/1519] eta 0:21:38 lr 0.000001 time 0.9446 (1.0077) model_time 0.9445 (1.0055) loss 0.6744 (0.7988) grad_norm 11.8252 (8.6643/1.9828) mem 68106MB [2022-12-20 20:50:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][240/1519] eta 0:21:28 lr 0.000001 time 0.9255 (1.0074) model_time 0.9254 (1.0052) loss 0.7981 (0.7970) grad_norm 9.0037 (8.7716/2.1368) mem 68106MB [2022-12-20 20:50:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][250/1519] eta 0:21:18 lr 0.000001 time 0.9747 (1.0072) model_time 0.9746 (1.0051) loss 0.8983 (0.7964) grad_norm 10.9360 (8.7798/2.1181) mem 68106MB [2022-12-20 20:50:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][260/1519] eta 0:21:08 lr 0.000001 time 0.9262 (1.0072) model_time 0.9261 (1.0051) loss 0.8899 (0.7989) grad_norm 7.2619 (8.8858/2.2632) mem 68106MB [2022-12-20 20:50:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][270/1519] eta 0:20:57 lr 0.000001 time 0.9235 (1.0068) model_time 0.9234 (1.0049) loss 0.7582 (0.7981) grad_norm 10.3904 (8.8900/2.2323) mem 68106MB [2022-12-20 20:50:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][280/1519] eta 0:20:47 lr 0.000001 time 0.9170 (1.0067) model_time 0.9168 (1.0048) loss 0.8192 (0.7969) grad_norm 6.7343 (8.8767/2.2141) mem 68106MB [2022-12-20 20:51:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][290/1519] eta 0:20:36 lr 0.000001 time 0.9215 (1.0064) model_time 0.9214 (1.0046) loss 0.7491 (0.7969) grad_norm 8.3351 (8.8706/2.1875) mem 68106MB [2022-12-20 20:51:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][300/1519] eta 0:20:26 lr 0.000001 time 0.9222 (1.0065) model_time 0.9220 (1.0047) loss 0.7330 (0.7969) grad_norm 10.0719 (8.8606/2.1656) mem 68106MB [2022-12-20 20:51:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][310/1519] eta 0:20:17 lr 0.000001 time 1.0074 (1.0068) model_time 1.0072 (1.0050) loss 0.8943 (0.7972) grad_norm 4.9612 (8.8614/2.1917) mem 68106MB [2022-12-20 20:51:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][320/1519] eta 0:20:07 lr 0.000001 time 0.9274 (1.0067) model_time 0.9272 (1.0050) loss 0.8200 (0.7974) grad_norm 7.5748 (8.8714/2.1962) mem 68106MB [2022-12-20 20:51:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][330/1519] eta 0:19:57 lr 0.000001 time 0.9238 (1.0067) model_time 0.9237 (1.0051) loss 0.7348 (0.7976) grad_norm 8.5277 (8.8667/2.1695) mem 68106MB [2022-12-20 20:51:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][340/1519] eta 0:19:46 lr 0.000001 time 0.9304 (1.0067) model_time 0.9303 (1.0051) loss 0.8928 (0.7987) grad_norm 9.3204 (8.9169/2.1772) mem 68106MB [2022-12-20 20:52:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][350/1519] eta 0:19:36 lr 0.000001 time 0.9297 (1.0068) model_time 0.9295 (1.0052) loss 0.7610 (0.8007) grad_norm 8.2354 (8.8832/2.1601) mem 68106MB [2022-12-20 20:52:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][360/1519] eta 0:19:26 lr 0.000001 time 0.9201 (1.0067) model_time 0.9200 (1.0052) loss 0.6627 (0.8017) grad_norm 10.5756 (8.8695/2.1417) mem 68106MB [2022-12-20 20:52:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][370/1519] eta 0:19:16 lr 0.000001 time 0.9239 (1.0065) model_time 0.9238 (1.0050) loss 0.8489 (0.8019) grad_norm 8.2551 (8.8548/2.1231) mem 68106MB [2022-12-20 20:52:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][380/1519] eta 0:19:06 lr 0.000001 time 1.0014 (1.0065) model_time 1.0012 (1.0050) loss 0.6708 (0.8025) grad_norm 8.1116 (8.8528/2.0972) mem 68106MB [2022-12-20 20:52:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][390/1519] eta 0:18:56 lr 0.000001 time 0.9229 (1.0063) model_time 0.9227 (1.0049) loss 0.9187 (0.8025) grad_norm 10.5189 (8.8436/2.0836) mem 68106MB [2022-12-20 20:52:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][400/1519] eta 0:18:45 lr 0.000001 time 0.9239 (1.0061) model_time 0.9237 (1.0047) loss 1.0208 (0.8029) grad_norm 12.7761 (8.8275/2.0966) mem 68106MB [2022-12-20 20:53:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][410/1519] eta 0:18:35 lr 0.000001 time 0.8888 (1.0060) model_time 0.8887 (1.0047) loss 0.6619 (0.8023) grad_norm 8.2825 (8.8218/2.0746) mem 68106MB [2022-12-20 20:53:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][420/1519] eta 0:18:25 lr 0.000001 time 0.9338 (1.0059) model_time 0.9336 (1.0045) loss 0.7101 (0.8023) grad_norm 7.5684 (8.8331/2.0879) mem 68106MB [2022-12-20 20:53:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][430/1519] eta 0:18:15 lr 0.000001 time 0.9872 (1.0059) model_time 0.9870 (1.0046) loss 0.8975 (0.8026) grad_norm 8.9482 (8.8205/2.0780) mem 68106MB [2022-12-20 20:53:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][440/1519] eta 0:18:05 lr 0.000001 time 0.9279 (1.0058) model_time 0.9278 (1.0045) loss 0.7654 (0.8026) grad_norm 8.3827 (8.8096/2.0680) mem 68106MB [2022-12-20 20:53:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][450/1519] eta 0:17:55 lr 0.000001 time 0.9215 (1.0058) model_time 0.9214 (1.0045) loss 0.7348 (0.8022) grad_norm 7.0633 (8.8087/2.0535) mem 68106MB [2022-12-20 20:53:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][460/1519] eta 0:17:45 lr 0.000001 time 0.9136 (1.0063) model_time 0.9134 (1.0050) loss 0.6806 (0.8012) grad_norm 8.5977 (8.8115/2.0727) mem 68106MB [2022-12-20 20:54:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][470/1519] eta 0:17:35 lr 0.000001 time 0.9323 (1.0065) model_time 0.9322 (1.0053) loss 0.7945 (0.8008) grad_norm 7.3643 (8.7847/2.0592) mem 68106MB [2022-12-20 20:54:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][480/1519] eta 0:17:25 lr 0.000001 time 0.9218 (1.0065) model_time 0.9216 (1.0053) loss 0.6960 (0.7999) grad_norm 8.6219 (8.7815/2.0420) mem 68106MB [2022-12-20 20:54:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][490/1519] eta 0:17:15 lr 0.000001 time 0.9961 (1.0067) model_time 0.9959 (1.0055) loss 0.8502 (0.7989) grad_norm 8.9788 (8.7553/2.0386) mem 68106MB [2022-12-20 20:54:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][500/1519] eta 0:17:05 lr 0.000001 time 0.9227 (1.0069) model_time 0.9225 (1.0057) loss 0.7821 (0.7989) grad_norm 6.8898 (8.7539/2.0465) mem 68106MB [2022-12-20 20:54:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][510/1519] eta 0:16:56 lr 0.000001 time 0.9221 (1.0070) model_time 0.9219 (1.0059) loss 0.7087 (0.8003) grad_norm 7.0402 (8.7424/2.0331) mem 68106MB [2022-12-20 20:54:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][520/1519] eta 0:16:46 lr 0.000001 time 0.9209 (1.0070) model_time 0.9208 (1.0059) loss 0.7816 (0.7999) grad_norm 7.4681 (8.7463/2.0219) mem 68106MB [2022-12-20 20:55:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][530/1519] eta 0:16:35 lr 0.000001 time 0.9234 (1.0069) model_time 0.9233 (1.0057) loss 0.7650 (0.7993) grad_norm 12.4792 (8.7353/2.0314) mem 68106MB [2022-12-20 20:55:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][540/1519] eta 0:16:25 lr 0.000001 time 0.9310 (1.0069) model_time 0.9309 (1.0058) loss 0.6616 (0.7975) grad_norm 9.3661 (8.7341/2.0182) mem 68106MB [2022-12-20 20:55:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][550/1519] eta 0:16:15 lr 0.000001 time 0.9273 (1.0068) model_time 0.9272 (1.0056) loss 0.7044 (0.7977) grad_norm 13.3778 (8.7618/2.0285) mem 68106MB [2022-12-20 20:55:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][560/1519] eta 0:16:05 lr 0.000001 time 0.9296 (1.0067) model_time 0.9295 (1.0056) loss 0.7216 (0.7965) grad_norm 6.5128 (8.7311/2.0264) mem 68106MB [2022-12-20 20:55:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][570/1519] eta 0:15:55 lr 0.000001 time 0.9227 (1.0067) model_time 0.9226 (1.0056) loss 0.8079 (0.7960) grad_norm 9.2794 (8.7344/2.0096) mem 68106MB [2022-12-20 20:55:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][580/1519] eta 0:15:45 lr 0.000001 time 0.9440 (1.0066) model_time 0.9439 (1.0055) loss 0.6540 (0.7965) grad_norm 6.0653 (8.7400/2.0245) mem 68106MB [2022-12-20 20:56:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][590/1519] eta 0:15:35 lr 0.000001 time 0.9235 (1.0065) model_time 0.9233 (1.0055) loss 0.6934 (0.7964) grad_norm 7.0726 (8.7377/2.0176) mem 68106MB [2022-12-20 20:56:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][600/1519] eta 0:15:24 lr 0.000001 time 0.9326 (1.0064) model_time 0.9324 (1.0054) loss 0.6620 (0.7967) grad_norm 10.7420 (8.7472/2.0162) mem 68106MB [2022-12-20 20:56:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][610/1519] eta 0:15:14 lr 0.000001 time 0.9211 (1.0063) model_time 0.9210 (1.0053) loss 0.7292 (0.7979) grad_norm 8.9337 (8.7337/2.0140) mem 68106MB [2022-12-20 20:56:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][620/1519] eta 0:15:04 lr 0.000001 time 0.9338 (1.0064) model_time 0.9336 (1.0054) loss 1.0056 (0.7994) grad_norm 10.0627 (8.7522/2.0203) mem 68106MB [2022-12-20 20:56:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][630/1519] eta 0:14:54 lr 0.000001 time 0.9238 (1.0063) model_time 0.9237 (1.0053) loss 0.8321 (0.7996) grad_norm 7.6079 (8.7712/2.0115) mem 68106MB [2022-12-20 20:56:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][640/1519] eta 0:14:44 lr 0.000001 time 0.9201 (1.0065) model_time 0.9199 (1.0055) loss 0.7773 (0.8009) grad_norm 8.4995 (8.7877/2.0052) mem 68106MB [2022-12-20 20:57:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][650/1519] eta 0:14:34 lr 0.000001 time 0.9867 (1.0065) model_time 0.9866 (1.0055) loss 0.7146 (0.8004) grad_norm 6.0888 (8.7835/2.0154) mem 68106MB [2022-12-20 20:57:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][660/1519] eta 0:14:24 lr 0.000001 time 0.9277 (1.0064) model_time 0.9275 (1.0054) loss 0.8396 (0.7993) grad_norm 9.1735 (8.7970/2.0361) mem 68106MB [2022-12-20 20:57:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][670/1519] eta 0:14:14 lr 0.000001 time 0.9505 (1.0063) model_time 0.9502 (1.0053) loss 0.7206 (0.7985) grad_norm 8.5826 (8.8192/2.0240) mem 68106MB [2022-12-20 20:57:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][680/1519] eta 0:14:04 lr 0.000001 time 0.9182 (1.0062) model_time 0.9179 (1.0052) loss 0.7360 (0.7979) grad_norm 8.7941 (8.8180/2.0222) mem 68106MB [2022-12-20 20:57:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][690/1519] eta 0:13:54 lr 0.000001 time 0.9233 (1.0060) model_time 0.9232 (1.0051) loss 0.8073 (0.7986) grad_norm 8.8329 (8.8231/2.0194) mem 68106MB [2022-12-20 20:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][700/1519] eta 0:13:44 lr 0.000001 time 0.9501 (1.0061) model_time 0.9500 (1.0052) loss 0.7669 (0.7990) grad_norm 13.5741 (8.8333/2.0563) mem 68106MB [2022-12-20 20:58:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][710/1519] eta 0:13:33 lr 0.000001 time 0.9201 (1.0060) model_time 0.9200 (1.0051) loss 0.7938 (0.7985) grad_norm 9.3809 (8.8384/2.0505) mem 68106MB [2022-12-20 20:58:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][720/1519] eta 0:13:23 lr 0.000001 time 0.9258 (1.0059) model_time 0.9256 (1.0050) loss 0.8368 (0.7975) grad_norm 8.8389 (8.7881/2.0177) mem 68106MB [2022-12-20 20:58:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][730/1519] eta 0:13:13 lr 0.000001 time 0.9225 (1.0058) model_time 0.9223 (1.0049) loss 0.8242 (0.7974) grad_norm 7.6701 (8.7788/2.0219) mem 68106MB [2022-12-20 20:58:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][740/1519] eta 0:13:03 lr 0.000001 time 0.9284 (1.0057) model_time 0.9281 (1.0048) loss 0.7421 (0.7973) grad_norm 8.7745 (8.7596/2.0066) mem 68106MB [2022-12-20 20:58:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][750/1519] eta 0:12:53 lr 0.000001 time 0.9243 (1.0058) model_time 0.9241 (1.0049) loss 0.7038 (0.7982) grad_norm 8.7497 (8.7583/1.9892) mem 68106MB [2022-12-20 20:58:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][760/1519] eta 0:12:43 lr 0.000001 time 0.9212 (1.0057) model_time 0.9210 (1.0048) loss 0.7935 (0.7982) grad_norm 6.5490 (8.7717/1.9818) mem 68106MB [2022-12-20 20:59:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][770/1519] eta 0:12:33 lr 0.000001 time 0.9252 (1.0056) model_time 0.9251 (1.0048) loss 0.7595 (0.7983) grad_norm 8.0567 (8.7450/1.9564) mem 68106MB [2022-12-20 20:59:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9230 (1.0058) model_time 0.9228 (1.0049) loss 0.6672 (0.7973) grad_norm 12.4450 (8.7481/1.9740) mem 68106MB [2022-12-20 20:59:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9224 (1.0058) model_time 0.9223 (1.0049) loss 0.6711 (0.7977) grad_norm 12.2710 (8.7478/1.9708) mem 68106MB [2022-12-20 20:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][800/1519] eta 0:12:03 lr 0.000001 time 0.9164 (1.0057) model_time 0.9162 (1.0049) loss 0.7721 (0.7975) grad_norm 7.6753 (8.7272/1.9628) mem 68106MB [2022-12-20 20:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][810/1519] eta 0:11:53 lr 0.000001 time 0.9473 (1.0058) model_time 0.9471 (1.0049) loss 0.8020 (0.7973) grad_norm 6.7040 (8.6951/1.9507) mem 68106MB [2022-12-20 20:59:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][820/1519] eta 0:11:43 lr 0.000001 time 0.8984 (1.0060) model_time 0.8982 (1.0051) loss 0.6742 (0.7982) grad_norm 7.8013 (8.7078/1.9398) mem 68106MB [2022-12-20 21:00:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9252 (1.0060) model_time 0.9250 (1.0052) loss 0.8427 (0.7983) grad_norm 6.6439 (8.7243/1.9717) mem 68106MB [2022-12-20 21:00:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9243 (1.0060) model_time 0.9242 (1.0052) loss 0.6985 (0.7985) grad_norm 7.4163 (8.6763/1.9087) mem 68106MB [2022-12-20 21:00:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][850/1519] eta 0:11:12 lr 0.000001 time 0.9292 (1.0059) model_time 0.9290 (1.0051) loss 1.0369 (0.7998) grad_norm 11.4868 (8.6848/1.9059) mem 68106MB [2022-12-20 21:00:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][860/1519] eta 0:11:02 lr 0.000001 time 0.9280 (1.0060) model_time 0.9277 (1.0052) loss 0.7307 (0.7998) grad_norm 9.2622 (8.6143/1.8232) mem 68106MB [2022-12-20 21:00:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][870/1519] eta 0:10:52 lr 0.000001 time 0.9208 (1.0059) model_time 0.9206 (1.0051) loss 0.7472 (0.7992) grad_norm 7.2720 (8.6446/1.9023) mem 68106MB [2022-12-20 21:00:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][880/1519] eta 0:10:42 lr 0.000001 time 0.9522 (1.0062) model_time 0.9521 (1.0054) loss 0.9422 (0.7999) grad_norm 10.5539 (8.6329/1.9149) mem 68106MB [2022-12-20 21:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][890/1519] eta 0:10:32 lr 0.000001 time 0.9341 (1.0061) model_time 0.9339 (1.0053) loss 0.6743 (0.7999) grad_norm 8.6237 (8.6446/1.9235) mem 68106MB [2022-12-20 21:01:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][900/1519] eta 0:10:22 lr 0.000001 time 0.9285 (1.0060) model_time 0.9283 (1.0052) loss 0.7179 (0.8007) grad_norm 9.0990 (8.6487/1.9373) mem 68106MB [2022-12-20 21:01:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][910/1519] eta 0:10:12 lr 0.000001 time 0.9420 (1.0060) model_time 0.9419 (1.0052) loss 0.6776 (0.8000) grad_norm 8.3827 (8.6279/1.9065) mem 68106MB [2022-12-20 21:01:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][920/1519] eta 0:10:02 lr 0.000001 time 0.9397 (1.0060) model_time 0.9396 (1.0052) loss 0.9726 (0.8002) grad_norm 7.2044 (8.6450/1.9153) mem 68106MB [2022-12-20 21:01:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][930/1519] eta 0:09:52 lr 0.000001 time 0.9243 (1.0060) model_time 0.9241 (1.0052) loss 0.6837 (0.7999) grad_norm 7.1087 (8.6588/1.9439) mem 68106MB [2022-12-20 21:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9250 (1.0060) model_time 0.9247 (1.0052) loss 0.7147 (0.7993) grad_norm 12.9418 (8.6654/1.9802) mem 68106MB [2022-12-20 21:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][950/1519] eta 0:09:32 lr 0.000001 time 0.9343 (1.0060) model_time 0.9342 (1.0052) loss 0.7611 (0.7991) grad_norm 15.6907 (8.6978/2.0195) mem 68106MB [2022-12-20 21:02:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9228 (1.0061) model_time 0.9221 (1.0053) loss 0.6683 (0.7988) grad_norm 10.0891 (8.6972/2.0209) mem 68106MB [2022-12-20 21:02:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][970/1519] eta 0:09:12 lr 0.000001 time 0.9291 (1.0061) model_time 0.9289 (1.0053) loss 0.9957 (0.7982) grad_norm 12.8497 (8.7013/2.0369) mem 68106MB [2022-12-20 21:02:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9284 (1.0060) model_time 0.9281 (1.0052) loss 0.7475 (0.7991) grad_norm 6.0991 (8.7282/2.1272) mem 68106MB [2022-12-20 21:02:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9539 (1.0060) model_time 0.9538 (1.0052) loss 0.7775 (0.7993) grad_norm 7.3229 (8.7440/2.1286) mem 68106MB [2022-12-20 21:02:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9354 (1.0060) model_time 0.9352 (1.0052) loss 0.9425 (0.8000) grad_norm 14.7268 (8.7843/2.1507) mem 68106MB [2022-12-20 21:03:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9277 (1.0059) model_time 0.9273 (1.0052) loss 0.9327 (0.7995) grad_norm 8.7830 (8.7714/2.1603) mem 68106MB [2022-12-20 21:03:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1020/1519] eta 0:08:21 lr 0.000001 time 0.9391 (1.0060) model_time 0.9389 (1.0052) loss 0.8605 (0.7996) grad_norm 8.7461 (8.7708/2.1478) mem 68106MB [2022-12-20 21:03:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1030/1519] eta 0:08:11 lr 0.000001 time 0.9341 (1.0059) model_time 0.9339 (1.0052) loss 0.6705 (0.7999) grad_norm 8.6227 (8.7710/2.1431) mem 68106MB [2022-12-20 21:03:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1040/1519] eta 0:08:01 lr 0.000001 time 0.9254 (1.0059) model_time 0.9252 (1.0051) loss 0.7081 (0.7997) grad_norm 8.3018 (8.7666/2.1437) mem 68106MB [2022-12-20 21:03:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1050/1519] eta 0:07:51 lr 0.000001 time 0.9415 (1.0059) model_time 0.9413 (1.0051) loss 0.8582 (0.7996) grad_norm 13.5543 (8.7897/2.1650) mem 68106MB [2022-12-20 21:03:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1060/1519] eta 0:07:41 lr 0.000001 time 0.9392 (1.0059) model_time 0.9390 (1.0052) loss 0.7993 (0.7994) grad_norm 13.2135 (8.8094/2.1692) mem 68106MB [2022-12-20 21:04:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1070/1519] eta 0:07:31 lr 0.000001 time 0.9344 (1.0059) model_time 0.9342 (1.0052) loss 0.7121 (0.7991) grad_norm 8.1045 (8.8283/2.1897) mem 68106MB [2022-12-20 21:04:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1080/1519] eta 0:07:21 lr 0.000001 time 0.9662 (1.0059) model_time 0.9660 (1.0052) loss 1.0920 (0.7994) grad_norm 7.5885 (8.8376/2.1938) mem 68106MB [2022-12-20 21:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9564 (1.0059) model_time 0.9562 (1.0052) loss 0.7699 (0.8002) grad_norm 9.4940 (8.8763/2.1910) mem 68106MB [2022-12-20 21:04:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9305 (1.0059) model_time 0.9303 (1.0052) loss 0.8432 (0.7998) grad_norm 8.6875 (8.8660/2.1763) mem 68106MB [2022-12-20 21:04:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9437 (1.0059) model_time 0.9435 (1.0052) loss 1.0972 (0.8002) grad_norm 8.5300 (8.8919/2.2050) mem 68106MB [2022-12-20 21:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9315 (1.0060) model_time 0.9313 (1.0052) loss 0.9603 (0.8001) grad_norm 9.4527 (8.8729/2.2146) mem 68106MB [2022-12-20 21:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9944 (1.0060) model_time 0.9942 (1.0053) loss 0.6681 (0.7997) grad_norm 6.4519 (8.8863/2.2000) mem 68106MB [2022-12-20 21:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9310 (1.0061) model_time 0.9308 (1.0054) loss 0.7108 (0.8002) grad_norm 8.6227 (8.8901/2.2058) mem 68106MB [2022-12-20 21:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9134 (1.0061) model_time 0.9133 (1.0054) loss 0.7402 (0.8003) grad_norm 10.6929 (8.8949/2.2124) mem 68106MB [2022-12-20 21:05:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9260 (1.0061) model_time 0.9258 (1.0054) loss 0.7281 (0.7996) grad_norm 7.7351 (8.8993/2.2070) mem 68106MB [2022-12-20 21:05:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9283 (1.0061) model_time 0.9282 (1.0054) loss 0.9074 (0.7996) grad_norm 5.9642 (8.9170/2.2489) mem 68106MB [2022-12-20 21:06:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9328 (1.0061) model_time 0.9324 (1.0053) loss 0.9477 (0.7998) grad_norm 9.0344 (8.9186/2.2388) mem 68106MB [2022-12-20 21:06:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1190/1519] eta 0:05:30 lr 0.000001 time 0.9253 (1.0060) model_time 0.9252 (1.0053) loss 0.8356 (0.7993) grad_norm 8.6922 (8.9051/2.2426) mem 68106MB [2022-12-20 21:06:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1200/1519] eta 0:05:20 lr 0.000001 time 0.9320 (1.0059) model_time 0.9319 (1.0052) loss 0.7799 (0.7994) grad_norm 9.8695 (8.8807/2.2393) mem 68106MB [2022-12-20 21:06:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1210/1519] eta 0:05:10 lr 0.000001 time 0.9411 (1.0060) model_time 0.9409 (1.0053) loss 0.6779 (0.7993) grad_norm 6.6697 (8.8795/2.2392) mem 68106MB [2022-12-20 21:06:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1220/1519] eta 0:05:00 lr 0.000001 time 0.9295 (1.0059) model_time 0.9294 (1.0052) loss 1.1534 (0.7993) grad_norm 8.1898 (8.8636/2.2195) mem 68106MB [2022-12-20 21:06:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1230/1519] eta 0:04:50 lr 0.000001 time 0.9284 (1.0059) model_time 0.9283 (1.0052) loss 0.9234 (0.7992) grad_norm 5.9376 (8.8701/2.2195) mem 68106MB [2022-12-20 21:07:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9285 (1.0059) model_time 0.9284 (1.0052) loss 0.6830 (0.7992) grad_norm 8.4614 (8.9371/2.5530) mem 68106MB [2022-12-20 21:07:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9287 (1.0059) model_time 0.9286 (1.0052) loss 0.8482 (0.7999) grad_norm 6.0669 (8.9115/2.5448) mem 68106MB [2022-12-20 21:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9331 (1.0059) model_time 0.9329 (1.0052) loss 0.6967 (0.7997) grad_norm 10.9251 (8.9231/2.5521) mem 68106MB [2022-12-20 21:07:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9316 (1.0061) model_time 0.9310 (1.0054) loss 0.9802 (0.7997) grad_norm 8.1551 (8.9488/2.5670) mem 68106MB [2022-12-20 21:07:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9298 (1.0061) model_time 0.9297 (1.0054) loss 0.6893 (0.7994) grad_norm 8.3030 (8.9400/2.5719) mem 68106MB [2022-12-20 21:07:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9311 (1.0060) model_time 0.9310 (1.0053) loss 1.0212 (0.7993) grad_norm 8.9278 (8.9115/2.5799) mem 68106MB [2022-12-20 21:08:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9284 (1.0060) model_time 0.9282 (1.0053) loss 0.6668 (0.7989) grad_norm 14.8133 (8.9302/2.5861) mem 68106MB [2022-12-20 21:08:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9908 (1.0060) model_time 0.9905 (1.0054) loss 0.9660 (0.7987) grad_norm 8.5484 (8.9315/2.5966) mem 68106MB [2022-12-20 21:08:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9331 (1.0060) model_time 0.9329 (1.0054) loss 0.7049 (0.7981) grad_norm 8.3233 (8.9460/2.5913) mem 68106MB [2022-12-20 21:08:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9427 (1.0060) model_time 0.9425 (1.0053) loss 0.7190 (0.7979) grad_norm 10.3536 (8.9609/2.5930) mem 68106MB [2022-12-20 21:08:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9321 (1.0060) model_time 0.9316 (1.0053) loss 0.7032 (0.7983) grad_norm 7.1442 (8.9633/2.5969) mem 68106MB [2022-12-20 21:08:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9309 (1.0060) model_time 0.9308 (1.0053) loss 0.7084 (0.7983) grad_norm 7.2342 (8.9552/2.6002) mem 68106MB [2022-12-20 21:09:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1360/1519] eta 0:02:39 lr 0.000001 time 0.9295 (1.0059) model_time 0.9294 (1.0053) loss 0.9343 (0.7985) grad_norm 9.1370 (8.9857/2.6026) mem 68106MB [2022-12-20 21:09:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1370/1519] eta 0:02:29 lr 0.000001 time 0.9302 (1.0059) model_time 0.9301 (1.0053) loss 0.8979 (0.7993) grad_norm 6.8323 (8.9917/2.6019) mem 68106MB [2022-12-20 21:09:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9195 (1.0059) model_time 0.9190 (1.0053) loss 1.0874 (0.7994) grad_norm 7.8663 (8.9850/2.5951) mem 68106MB [2022-12-20 21:09:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9242 (1.0059) model_time 0.9239 (1.0052) loss 0.8774 (0.7996) grad_norm 9.4136 (8.9883/2.5921) mem 68106MB [2022-12-20 21:09:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9270 (1.0059) model_time 0.9268 (1.0052) loss 1.0029 (0.7995) grad_norm 5.8270 (8.9974/2.5945) mem 68106MB [2022-12-20 21:09:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9324 (1.0059) model_time 0.9322 (1.0052) loss 0.8554 (0.7994) grad_norm 9.2704 (9.0383/2.5982) mem 68106MB [2022-12-20 21:10:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9300 (1.0060) model_time 0.9297 (1.0053) loss 0.8126 (0.7993) grad_norm 7.2692 (9.0209/2.5939) mem 68106MB [2022-12-20 21:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9308 (1.0062) model_time 0.9306 (1.0055) loss 0.6665 (0.7990) grad_norm 6.9950 (8.9571/2.5893) mem 68106MB [2022-12-20 21:10:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9347 (1.0062) model_time 0.9346 (1.0055) loss 0.8211 (0.7991) grad_norm 10.3490 (9.0071/2.6252) mem 68106MB [2022-12-20 21:10:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9318 (1.0062) model_time 0.9316 (1.0055) loss 0.6631 (0.7997) grad_norm 8.4220 (8.9852/2.6237) mem 68106MB [2022-12-20 21:10:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9274 (1.0062) model_time 0.9272 (1.0055) loss 0.7787 (0.8001) grad_norm 7.1738 (9.0043/2.6145) mem 68106MB [2022-12-20 21:10:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9340 (1.0061) model_time 0.9339 (1.0055) loss 0.8720 (0.7996) grad_norm 12.6502 (8.9663/2.5768) mem 68106MB [2022-12-20 21:11:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9327 (1.0061) model_time 0.9326 (1.0055) loss 0.9709 (0.7995) grad_norm 9.8324 (8.9897/2.5756) mem 68106MB [2022-12-20 21:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9276 (1.0061) model_time 0.9274 (1.0054) loss 0.7499 (0.7991) grad_norm 6.7372 (8.9769/2.5799) mem 68106MB [2022-12-20 21:11:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9154 (1.0061) model_time 0.9152 (1.0054) loss 0.7990 (0.7990) grad_norm 9.7747 (8.9693/2.5691) mem 68106MB [2022-12-20 21:11:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [92/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9372 (1.0060) model_time 0.9371 (1.0054) loss 0.8111 (0.7985) grad_norm 10.1815 (8.9829/2.5698) mem 68106MB [2022-12-20 21:11:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 92 training takes 0:25:28 [2022-12-20 21:11:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_92.pth saving...... [2022-12-20 21:12:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_92.pth saved !!! [2022-12-20 21:12:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.657 (0.657) Loss 0.5377 (0.5377) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 21:12:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.290 (0.330) Loss 0.5347 (0.5072) Acc@1 92.361 (92.803) Acc@5 98.611 (98.516) Mem 68106MB [2022-12-20 21:12:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.304 (0.315) Loss 0.4840 (0.5025) Acc@1 90.972 (92.791) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-20 21:12:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.6329 (0.5093) Acc@1 90.972 (92.574) Acc@5 97.917 (98.466) Mem 68106MB [2022-12-20 21:12:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.306) Loss 0.4594 (0.5002) Acc@1 93.750 (92.641) Acc@5 99.306 (98.552) Mem 68106MB [2022-12-20 21:12:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.303 (0.306) Loss 0.4893 (0.4976) Acc@1 92.361 (92.708) Acc@5 99.653 (98.611) Mem 68106MB [2022-12-20 21:12:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.304) Loss 0.5056 (0.4973) Acc@1 90.972 (92.629) Acc@5 98.264 (98.583) Mem 68106MB [2022-12-20 21:12:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5430 (0.4985) Acc@1 93.056 (92.591) Acc@5 98.264 (98.572) Mem 68106MB [2022-12-20 21:12:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4304 (0.4972) Acc@1 93.403 (92.618) Acc@5 98.264 (98.598) Mem 68106MB [2022-12-20 21:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:92] * Acc@1 92.592 Acc@5 98.600 [2022-12-20 21:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 21:12:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 21:12:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 21:12:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.59% [2022-12-20 21:12:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][0/1519] eta 0:37:40 lr 0.000001 time 1.4883 (1.4883) model_time 0.9923 (0.9923) loss 1.0543 (1.0543) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 21:13:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][10/1519] eta 0:26:25 lr 0.000001 time 0.9290 (1.0507) model_time 0.9289 (1.0053) loss 0.8332 (0.9133) grad_norm 8.4493 (9.4317/1.8449) mem 68106MB [2022-12-20 21:13:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][20/1519] eta 0:25:38 lr 0.000001 time 0.9230 (1.0266) model_time 0.9228 (1.0026) loss 0.8163 (0.8869) grad_norm 7.2553 (8.8643/1.9354) mem 68106MB [2022-12-20 21:13:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][30/1519] eta 0:25:16 lr 0.000001 time 0.9254 (1.0184) model_time 0.9253 (1.0021) loss 1.3335 (0.9092) grad_norm 7.7078 (8.6322/1.6633) mem 68106MB [2022-12-20 21:13:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][40/1519] eta 0:25:03 lr 0.000001 time 0.9219 (1.0163) model_time 0.9218 (1.0039) loss 0.6585 (0.8802) grad_norm 6.5686 (8.6942/2.7168) mem 68106MB [2022-12-20 21:13:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][50/1519] eta 0:24:50 lr 0.000001 time 0.9184 (1.0146) model_time 0.9183 (1.0045) loss 1.2149 (0.8700) grad_norm 8.7108 (8.7642/2.6733) mem 68106MB [2022-12-20 21:13:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][60/1519] eta 0:24:39 lr 0.000001 time 0.9279 (1.0139) model_time 0.9278 (1.0054) loss 0.6707 (0.8587) grad_norm 7.1504 (8.6818/2.5263) mem 68106MB [2022-12-20 21:14:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][70/1519] eta 0:24:27 lr 0.000001 time 0.9258 (1.0128) model_time 0.9256 (1.0055) loss 0.7084 (0.8394) grad_norm 8.5082 (8.6067/2.5761) mem 68106MB [2022-12-20 21:14:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][80/1519] eta 0:24:15 lr 0.000001 time 0.9227 (1.0112) model_time 0.9226 (1.0048) loss 0.7090 (0.8297) grad_norm 8.6952 (8.7330/2.5500) mem 68106MB [2022-12-20 21:14:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][90/1519] eta 0:24:04 lr 0.000001 time 0.9080 (1.0106) model_time 0.9079 (1.0049) loss 0.6799 (0.8213) grad_norm 5.4815 (8.5525/2.4765) mem 68106MB [2022-12-20 21:14:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][100/1519] eta 0:23:52 lr 0.000001 time 0.9222 (1.0097) model_time 0.9220 (1.0045) loss 0.8435 (0.8182) grad_norm 8.3725 (8.5440/2.3851) mem 68106MB [2022-12-20 21:14:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][110/1519] eta 0:23:42 lr 0.000001 time 0.9339 (1.0097) model_time 0.9337 (1.0050) loss 0.7780 (0.8181) grad_norm 7.6820 (8.5768/2.3507) mem 68106MB [2022-12-20 21:14:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][120/1519] eta 0:23:32 lr 0.000001 time 0.9227 (1.0095) model_time 0.9226 (1.0051) loss 0.8166 (0.8140) grad_norm 8.7564 (8.5549/2.2789) mem 68106MB [2022-12-20 21:15:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][130/1519] eta 0:23:21 lr 0.000001 time 0.9221 (1.0091) model_time 0.9220 (1.0051) loss 0.6794 (0.8129) grad_norm 8.5393 (8.5867/2.2456) mem 68106MB [2022-12-20 21:15:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][140/1519] eta 0:23:10 lr 0.000001 time 0.9264 (1.0086) model_time 0.9262 (1.0048) loss 0.7642 (0.8138) grad_norm 16.3571 (8.6156/2.4135) mem 68106MB [2022-12-20 21:15:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][150/1519] eta 0:23:00 lr 0.000001 time 0.9035 (1.0087) model_time 0.9034 (1.0051) loss 0.7016 (0.8135) grad_norm 8.6973 (8.6400/2.3607) mem 68106MB [2022-12-20 21:15:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][160/1519] eta 0:22:50 lr 0.000001 time 0.9247 (1.0082) model_time 0.9245 (1.0048) loss 0.6737 (0.8090) grad_norm 9.2750 (8.6203/2.2999) mem 68106MB [2022-12-20 21:15:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][170/1519] eta 0:22:42 lr 0.000001 time 0.9083 (1.0101) model_time 0.9082 (1.0069) loss 0.7391 (0.8067) grad_norm 9.0643 (8.6495/2.3228) mem 68106MB [2022-12-20 21:15:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][180/1519] eta 0:22:31 lr 0.000001 time 0.9230 (1.0095) model_time 0.9229 (1.0065) loss 0.8104 (0.8059) grad_norm 7.5564 (8.7447/2.3242) mem 68106MB [2022-12-20 21:16:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][190/1519] eta 0:22:21 lr 0.000001 time 0.9296 (1.0094) model_time 0.9294 (1.0065) loss 0.8576 (0.8101) grad_norm 6.6913 (8.6644/2.2979) mem 68106MB [2022-12-20 21:16:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][200/1519] eta 0:22:10 lr 0.000001 time 0.9195 (1.0090) model_time 0.9194 (1.0062) loss 0.8028 (0.8081) grad_norm 6.2567 (8.5766/2.2894) mem 68106MB [2022-12-20 21:16:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][210/1519] eta 0:22:00 lr 0.000001 time 0.9499 (1.0086) model_time 0.9497 (1.0059) loss 0.6808 (0.8061) grad_norm 7.4872 (8.5461/2.2547) mem 68106MB [2022-12-20 21:16:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][220/1519] eta 0:21:49 lr 0.000001 time 0.9270 (1.0084) model_time 0.9269 (1.0058) loss 1.0414 (0.8054) grad_norm 21.4099 (8.6394/2.5396) mem 68106MB [2022-12-20 21:16:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][230/1519] eta 0:21:39 lr 0.000001 time 0.9251 (1.0081) model_time 0.9250 (1.0056) loss 0.7350 (0.8066) grad_norm 6.0265 (8.6268/2.5040) mem 68106MB [2022-12-20 21:17:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][240/1519] eta 0:21:30 lr 0.000001 time 0.9217 (1.0092) model_time 0.9215 (1.0069) loss 0.6649 (0.8107) grad_norm 7.6282 (8.6456/2.4862) mem 68106MB [2022-12-20 21:17:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][250/1519] eta 0:21:20 lr 0.000001 time 0.9159 (1.0091) model_time 0.9158 (1.0069) loss 0.9509 (0.8100) grad_norm 7.7759 (8.6310/2.4446) mem 68106MB [2022-12-20 21:17:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][260/1519] eta 0:21:10 lr 0.000001 time 0.9256 (1.0090) model_time 0.9254 (1.0068) loss 0.6807 (0.8069) grad_norm 8.5765 (8.6217/2.4089) mem 68106MB [2022-12-20 21:17:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][270/1519] eta 0:20:59 lr 0.000001 time 0.9247 (1.0087) model_time 0.9246 (1.0066) loss 0.7132 (0.8080) grad_norm 8.9172 (8.6212/2.3725) mem 68106MB [2022-12-20 21:17:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][280/1519] eta 0:20:49 lr 0.000001 time 0.9363 (1.0084) model_time 0.9362 (1.0063) loss 0.9100 (0.8093) grad_norm 9.2469 (8.6342/2.3330) mem 68106MB [2022-12-20 21:17:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][290/1519] eta 0:20:39 lr 0.000001 time 0.9216 (1.0081) model_time 0.9214 (1.0061) loss 0.8184 (0.8118) grad_norm 7.3757 (8.6097/2.3219) mem 68106MB [2022-12-20 21:18:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][300/1519] eta 0:20:29 lr 0.000001 time 0.9250 (1.0082) model_time 0.9249 (1.0063) loss 0.8453 (0.8123) grad_norm 4.8344 (8.6290/2.3686) mem 68106MB [2022-12-20 21:18:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][310/1519] eta 0:20:18 lr 0.000001 time 0.9240 (1.0081) model_time 0.9238 (1.0062) loss 0.7552 (0.8118) grad_norm 8.0639 (8.6244/2.3382) mem 68106MB [2022-12-20 21:18:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][320/1519] eta 0:20:08 lr 0.000001 time 0.9278 (1.0078) model_time 0.9277 (1.0060) loss 1.1305 (0.8135) grad_norm 9.8246 (8.6142/2.3197) mem 68106MB [2022-12-20 21:18:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][330/1519] eta 0:19:57 lr 0.000001 time 0.9203 (1.0075) model_time 0.9201 (1.0057) loss 0.9265 (0.8142) grad_norm 8.5894 (8.6367/2.3279) mem 68106MB [2022-12-20 21:18:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][340/1519] eta 0:19:47 lr 0.000001 time 0.9232 (1.0072) model_time 0.9230 (1.0055) loss 0.6587 (0.8134) grad_norm 8.1321 (8.6404/2.3079) mem 68106MB [2022-12-20 21:18:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][350/1519] eta 0:19:37 lr 0.000001 time 0.9282 (1.0075) model_time 0.9280 (1.0058) loss 0.7715 (0.8129) grad_norm 7.1788 (8.6430/2.3012) mem 68106MB [2022-12-20 21:19:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][360/1519] eta 0:19:27 lr 0.000001 time 0.9723 (1.0075) model_time 0.9722 (1.0059) loss 0.7489 (0.8107) grad_norm 8.9103 (8.6186/2.2758) mem 68106MB [2022-12-20 21:19:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][370/1519] eta 0:19:17 lr 0.000001 time 0.9315 (1.0076) model_time 0.9313 (1.0059) loss 0.6597 (0.8086) grad_norm 11.0890 (8.6210/2.2559) mem 68106MB [2022-12-20 21:19:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][380/1519] eta 0:19:07 lr 0.000001 time 0.9310 (1.0075) model_time 0.9308 (1.0059) loss 0.6905 (0.8066) grad_norm 6.3209 (8.5797/2.2430) mem 68106MB [2022-12-20 21:19:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][390/1519] eta 0:18:57 lr 0.000001 time 0.9261 (1.0073) model_time 0.9260 (1.0057) loss 0.6977 (0.8059) grad_norm 8.0782 (8.5800/2.2221) mem 68106MB [2022-12-20 21:19:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][400/1519] eta 0:18:46 lr 0.000001 time 0.9253 (1.0071) model_time 0.9251 (1.0055) loss 0.8458 (0.8052) grad_norm 7.6931 (8.5985/2.2278) mem 68106MB [2022-12-20 21:19:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][410/1519] eta 0:18:36 lr 0.000001 time 0.9275 (1.0071) model_time 0.9273 (1.0056) loss 0.7547 (0.8042) grad_norm 9.6973 (8.5931/2.2115) mem 68106MB [2022-12-20 21:20:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][420/1519] eta 0:18:26 lr 0.000001 time 0.9205 (1.0071) model_time 0.9204 (1.0057) loss 0.7840 (0.8035) grad_norm 6.7365 (8.6223/2.2676) mem 68106MB [2022-12-20 21:20:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][430/1519] eta 0:18:16 lr 0.000001 time 0.9297 (1.0073) model_time 0.9296 (1.0058) loss 0.8459 (0.8034) grad_norm 8.7056 (8.6055/2.2458) mem 68106MB [2022-12-20 21:20:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][440/1519] eta 0:18:06 lr 0.000001 time 0.9368 (1.0073) model_time 0.9366 (1.0059) loss 0.9781 (0.8031) grad_norm 9.5656 (8.5953/2.2323) mem 68106MB [2022-12-20 21:20:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][450/1519] eta 0:17:56 lr 0.000001 time 0.9227 (1.0072) model_time 0.9225 (1.0058) loss 0.6836 (0.8024) grad_norm 9.0989 (8.5741/2.2213) mem 68106MB [2022-12-20 21:20:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][460/1519] eta 0:17:46 lr 0.000001 time 0.9274 (1.0072) model_time 0.9273 (1.0058) loss 0.8096 (0.8024) grad_norm 18.0450 (8.6531/2.3033) mem 68106MB [2022-12-20 21:20:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][470/1519] eta 0:17:36 lr 0.000001 time 0.9227 (1.0071) model_time 0.9225 (1.0057) loss 0.6774 (0.8034) grad_norm 6.9071 (8.6368/2.2865) mem 68106MB [2022-12-20 21:21:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][480/1519] eta 0:17:26 lr 0.000001 time 0.9306 (1.0069) model_time 0.9304 (1.0055) loss 0.7338 (0.8029) grad_norm 13.1350 (8.6393/2.2944) mem 68106MB [2022-12-20 21:21:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][490/1519] eta 0:17:16 lr 0.000001 time 0.9211 (1.0068) model_time 0.9210 (1.0055) loss 0.8844 (0.8035) grad_norm 7.9213 (8.6485/2.2978) mem 68106MB [2022-12-20 21:21:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][500/1519] eta 0:17:05 lr 0.000001 time 0.9500 (1.0068) model_time 0.9497 (1.0055) loss 0.7604 (0.8032) grad_norm 13.5230 (8.6568/2.3009) mem 68106MB [2022-12-20 21:21:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][510/1519] eta 0:16:55 lr 0.000001 time 0.9341 (1.0067) model_time 0.9339 (1.0054) loss 0.8518 (0.8049) grad_norm 7.7635 (8.6726/2.2861) mem 68106MB [2022-12-20 21:21:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][520/1519] eta 0:16:45 lr 0.000001 time 0.9299 (1.0066) model_time 0.9297 (1.0053) loss 0.9389 (0.8055) grad_norm 9.0401 (8.7200/2.3237) mem 68106MB [2022-12-20 21:21:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][530/1519] eta 0:16:35 lr 0.000001 time 0.9400 (1.0066) model_time 0.9399 (1.0053) loss 0.7071 (0.8051) grad_norm 7.2501 (8.7063/2.3070) mem 68106MB [2022-12-20 21:22:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][540/1519] eta 0:16:25 lr 0.000001 time 0.9249 (1.0066) model_time 0.9248 (1.0053) loss 0.8383 (0.8054) grad_norm 6.5282 (8.6810/2.2966) mem 68106MB [2022-12-20 21:22:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][550/1519] eta 0:16:15 lr 0.000001 time 1.1622 (1.0071) model_time 1.1620 (1.0059) loss 0.6770 (0.8052) grad_norm 7.6675 (8.6684/2.2790) mem 68106MB [2022-12-20 21:22:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][560/1519] eta 0:16:05 lr 0.000001 time 0.9203 (1.0072) model_time 0.9202 (1.0059) loss 0.8298 (0.8051) grad_norm 8.9006 (8.6796/2.2777) mem 68106MB [2022-12-20 21:22:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][570/1519] eta 0:15:55 lr 0.000001 time 0.9316 (1.0070) model_time 0.9314 (1.0058) loss 0.7154 (0.8056) grad_norm 6.7922 (8.6972/2.2725) mem 68106MB [2022-12-20 21:22:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][580/1519] eta 0:15:45 lr 0.000001 time 0.9268 (1.0069) model_time 0.9267 (1.0057) loss 0.6744 (0.8049) grad_norm 7.2805 (8.7049/2.2625) mem 68106MB [2022-12-20 21:22:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][590/1519] eta 0:15:35 lr 0.000001 time 0.9086 (1.0069) model_time 0.9085 (1.0057) loss 0.6809 (0.8044) grad_norm 7.1816 (8.6819/2.2506) mem 68106MB [2022-12-20 21:23:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][600/1519] eta 0:15:25 lr 0.000001 time 0.9286 (1.0069) model_time 0.9284 (1.0058) loss 0.8577 (0.8039) grad_norm 8.8738 (8.6659/2.2440) mem 68106MB [2022-12-20 21:23:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][610/1519] eta 0:15:15 lr 0.000001 time 0.9343 (1.0070) model_time 0.9341 (1.0058) loss 0.8502 (0.8027) grad_norm 9.2525 (8.6606/2.2578) mem 68106MB [2022-12-20 21:23:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9359 (1.0070) model_time 0.9357 (1.0059) loss 0.6963 (0.8030) grad_norm 9.9690 (8.6947/2.2632) mem 68106MB [2022-12-20 21:23:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9248 (1.0069) model_time 0.9246 (1.0058) loss 0.6986 (0.8033) grad_norm 7.4415 (8.7098/2.2793) mem 68106MB [2022-12-20 21:23:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][640/1519] eta 0:14:44 lr 0.000001 time 0.9223 (1.0068) model_time 0.9220 (1.0057) loss 0.9991 (0.8036) grad_norm 12.0013 (8.7265/2.2239) mem 68106MB [2022-12-20 21:23:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][650/1519] eta 0:14:34 lr 0.000001 time 0.9291 (1.0067) model_time 0.9290 (1.0056) loss 0.7802 (0.8030) grad_norm 7.2923 (8.7145/2.2169) mem 68106MB [2022-12-20 21:24:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][660/1519] eta 0:14:25 lr 0.000001 time 0.9314 (1.0070) model_time 0.9312 (1.0060) loss 0.7983 (0.8027) grad_norm 9.7447 (8.7035/2.2168) mem 68106MB [2022-12-20 21:24:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][670/1519] eta 0:14:14 lr 0.000001 time 0.9412 (1.0070) model_time 0.9410 (1.0059) loss 0.6871 (0.8028) grad_norm 10.3192 (8.7886/2.5662) mem 68106MB [2022-12-20 21:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][680/1519] eta 0:14:04 lr 0.000001 time 0.9204 (1.0069) model_time 0.9203 (1.0059) loss 0.8232 (0.8021) grad_norm 8.1270 (8.7823/2.5583) mem 68106MB [2022-12-20 21:24:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][690/1519] eta 0:13:54 lr 0.000001 time 0.9215 (1.0068) model_time 0.9214 (1.0057) loss 0.6772 (0.8022) grad_norm 7.7376 (8.8030/2.5500) mem 68106MB [2022-12-20 21:24:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][700/1519] eta 0:13:44 lr 0.000001 time 0.9217 (1.0067) model_time 0.9216 (1.0057) loss 0.6747 (0.8014) grad_norm 7.8979 (8.7901/2.5493) mem 68106MB [2022-12-20 21:24:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][710/1519] eta 0:13:34 lr 0.000001 time 0.9234 (1.0066) model_time 0.9232 (1.0056) loss 0.6832 (0.8009) grad_norm 7.7197 (8.7807/2.5470) mem 68106MB [2022-12-20 21:25:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][720/1519] eta 0:13:24 lr 0.000001 time 0.9200 (1.0066) model_time 0.9198 (1.0056) loss 0.6589 (0.8013) grad_norm 7.4784 (8.8399/2.6280) mem 68106MB [2022-12-20 21:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][730/1519] eta 0:13:14 lr 0.000001 time 1.0625 (1.0067) model_time 1.0623 (1.0057) loss 0.9047 (0.8013) grad_norm 10.2439 (8.8280/2.6242) mem 68106MB [2022-12-20 21:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9208 (1.0067) model_time 0.9206 (1.0057) loss 0.7718 (0.8024) grad_norm 8.9423 (8.8215/2.5847) mem 68106MB [2022-12-20 21:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][750/1519] eta 0:12:54 lr 0.000001 time 0.8873 (1.0068) model_time 0.8871 (1.0058) loss 0.8379 (0.8024) grad_norm 7.6636 (8.8148/2.5837) mem 68106MB [2022-12-20 21:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9257 (1.0067) model_time 0.9256 (1.0057) loss 0.7187 (0.8019) grad_norm 6.8816 (8.7999/2.5887) mem 68106MB [2022-12-20 21:25:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][770/1519] eta 0:12:34 lr 0.000001 time 1.0116 (1.0067) model_time 1.0115 (1.0058) loss 0.6843 (0.8025) grad_norm 8.0879 (8.8022/2.5802) mem 68106MB [2022-12-20 21:26:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9234 (1.0068) model_time 0.9233 (1.0058) loss 0.8040 (0.8024) grad_norm 7.2155 (8.7848/2.5992) mem 68106MB [2022-12-20 21:26:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9233 (1.0068) model_time 0.9231 (1.0059) loss 0.7780 (0.8033) grad_norm 8.7214 (8.8161/2.5941) mem 68106MB [2022-12-20 21:26:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][800/1519] eta 0:12:04 lr 0.000001 time 0.9278 (1.0071) model_time 0.9277 (1.0061) loss 0.9139 (0.8026) grad_norm 7.2638 (8.8444/2.5809) mem 68106MB [2022-12-20 21:26:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][810/1519] eta 0:11:53 lr 0.000001 time 0.9271 (1.0070) model_time 0.9269 (1.0061) loss 0.8522 (0.8029) grad_norm 7.9345 (8.8380/2.5820) mem 68106MB [2022-12-20 21:26:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][820/1519] eta 0:11:43 lr 0.000001 time 0.9200 (1.0069) model_time 0.9198 (1.0060) loss 0.8396 (0.8026) grad_norm 10.7208 (8.8186/2.4741) mem 68106MB [2022-12-20 21:26:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9217 (1.0068) model_time 0.9216 (1.0059) loss 0.7013 (0.8019) grad_norm 7.2344 (8.8381/2.4872) mem 68106MB [2022-12-20 21:27:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9258 (1.0068) model_time 0.9257 (1.0058) loss 0.9490 (0.8018) grad_norm 9.6540 (8.8366/2.4751) mem 68106MB [2022-12-20 21:27:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][850/1519] eta 0:11:13 lr 0.000001 time 0.9227 (1.0067) model_time 0.9226 (1.0058) loss 0.6888 (0.8020) grad_norm 9.0437 (8.8397/2.4725) mem 68106MB [2022-12-20 21:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][860/1519] eta 0:11:03 lr 0.000001 time 0.9185 (1.0068) model_time 0.9184 (1.0059) loss 0.7873 (0.8018) grad_norm 8.8206 (8.8432/2.4760) mem 68106MB [2022-12-20 21:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][870/1519] eta 0:10:53 lr 0.000001 time 0.9053 (1.0068) model_time 0.9051 (1.0059) loss 0.8665 (0.8017) grad_norm 9.5366 (8.8517/2.4764) mem 68106MB [2022-12-20 21:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][880/1519] eta 0:10:43 lr 0.000001 time 0.9227 (1.0067) model_time 0.9226 (1.0058) loss 0.8055 (0.8018) grad_norm 8.3065 (8.8527/2.4848) mem 68106MB [2022-12-20 21:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][890/1519] eta 0:10:33 lr 0.000001 time 0.9188 (1.0066) model_time 0.9187 (1.0057) loss 0.7147 (0.8018) grad_norm 8.0286 (8.8748/2.5013) mem 68106MB [2022-12-20 21:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][900/1519] eta 0:10:23 lr 0.000001 time 0.9290 (1.0066) model_time 0.9288 (1.0057) loss 1.0277 (0.8022) grad_norm 6.9289 (8.8449/2.4726) mem 68106MB [2022-12-20 21:28:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][910/1519] eta 0:10:13 lr 0.000001 time 0.9895 (1.0068) model_time 0.9894 (1.0060) loss 0.8286 (0.8034) grad_norm 18.0886 (8.8593/2.5358) mem 68106MB [2022-12-20 21:28:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][920/1519] eta 0:10:03 lr 0.000001 time 0.9280 (1.0068) model_time 0.9279 (1.0060) loss 0.7752 (0.8037) grad_norm 10.6319 (8.8665/2.5329) mem 68106MB [2022-12-20 21:28:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][930/1519] eta 0:09:53 lr 0.000001 time 0.9662 (1.0068) model_time 0.9660 (1.0059) loss 0.9178 (0.8032) grad_norm 8.2773 (8.8782/2.5211) mem 68106MB [2022-12-20 21:28:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9223 (1.0067) model_time 0.9222 (1.0059) loss 0.6950 (0.8038) grad_norm 9.2164 (8.8963/2.5288) mem 68106MB [2022-12-20 21:28:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][950/1519] eta 0:09:32 lr 0.000001 time 1.2026 (1.0069) model_time 1.2024 (1.0061) loss 0.7914 (0.8036) grad_norm 7.7435 (8.8896/2.5180) mem 68106MB [2022-12-20 21:29:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9272 (1.0069) model_time 0.9270 (1.0061) loss 0.9046 (0.8039) grad_norm 5.7442 (8.8895/2.5258) mem 68106MB [2022-12-20 21:29:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][970/1519] eta 0:09:12 lr 0.000001 time 0.9841 (1.0069) model_time 0.9840 (1.0061) loss 0.8262 (0.8038) grad_norm 8.1424 (8.8927/2.5248) mem 68106MB [2022-12-20 21:29:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9256 (1.0069) model_time 0.9254 (1.0061) loss 0.8523 (0.8050) grad_norm 12.2218 (8.9280/2.5291) mem 68106MB [2022-12-20 21:29:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9300 (1.0069) model_time 0.9298 (1.0061) loss 0.6693 (0.8047) grad_norm 13.0229 (8.9393/2.5438) mem 68106MB [2022-12-20 21:29:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9229 (1.0068) model_time 0.9227 (1.0060) loss 0.6733 (0.8052) grad_norm 8.9989 (8.9160/2.5314) mem 68106MB [2022-12-20 21:29:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1010/1519] eta 0:08:32 lr 0.000001 time 0.9231 (1.0067) model_time 0.9230 (1.0059) loss 0.8291 (0.8050) grad_norm 8.4383 (8.9260/2.5335) mem 68106MB [2022-12-20 21:30:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1020/1519] eta 0:08:22 lr 0.000001 time 0.9308 (1.0067) model_time 0.9306 (1.0059) loss 0.8869 (0.8046) grad_norm 11.7657 (8.9083/2.4933) mem 68106MB [2022-12-20 21:30:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1030/1519] eta 0:08:12 lr 0.000001 time 0.9237 (1.0066) model_time 0.9235 (1.0058) loss 0.9572 (0.8053) grad_norm 9.8638 (8.9593/2.5785) mem 68106MB [2022-12-20 21:30:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1040/1519] eta 0:08:02 lr 0.000001 time 0.9216 (1.0066) model_time 0.9214 (1.0058) loss 0.7949 (0.8059) grad_norm 11.5082 (8.9622/2.5845) mem 68106MB [2022-12-20 21:30:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1050/1519] eta 0:07:52 lr 0.000001 time 0.9309 (1.0067) model_time 0.9307 (1.0059) loss 1.0024 (0.8060) grad_norm 12.6774 (9.0078/2.5954) mem 68106MB [2022-12-20 21:30:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1060/1519] eta 0:07:42 lr 0.000001 time 0.9311 (1.0067) model_time 0.9309 (1.0059) loss 0.8647 (0.8065) grad_norm 10.9710 (8.9610/2.5470) mem 68106MB [2022-12-20 21:30:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1070/1519] eta 0:07:32 lr 0.000001 time 0.9245 (1.0067) model_time 0.9244 (1.0059) loss 0.7065 (0.8060) grad_norm 8.5866 (8.9573/2.5491) mem 68106MB [2022-12-20 21:31:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1080/1519] eta 0:07:21 lr 0.000001 time 0.9335 (1.0066) model_time 0.9334 (1.0059) loss 0.9826 (0.8062) grad_norm 7.0992 (8.9569/2.5553) mem 68106MB [2022-12-20 21:31:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9259 (1.0066) model_time 0.9258 (1.0058) loss 0.6659 (0.8060) grad_norm 7.4512 (8.9362/2.5481) mem 68106MB [2022-12-20 21:31:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9307 (1.0066) model_time 0.9306 (1.0059) loss 0.7799 (0.8058) grad_norm 8.8542 (8.9194/2.5322) mem 68106MB [2022-12-20 21:31:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9311 (1.0066) model_time 0.9310 (1.0058) loss 0.9868 (0.8059) grad_norm 7.0138 (8.9018/2.5306) mem 68106MB [2022-12-20 21:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9245 (1.0065) model_time 0.9244 (1.0058) loss 0.9687 (0.8058) grad_norm 7.8297 (8.8458/2.4955) mem 68106MB [2022-12-20 21:31:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9307 (1.0065) model_time 0.9306 (1.0057) loss 1.1453 (0.8070) grad_norm 11.5378 (8.8715/2.5015) mem 68106MB [2022-12-20 21:32:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9331 (1.0065) model_time 0.9330 (1.0057) loss 0.7107 (0.8073) grad_norm 9.8239 (8.8875/2.4956) mem 68106MB [2022-12-20 21:32:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9808 (1.0065) model_time 0.9807 (1.0058) loss 0.7128 (0.8071) grad_norm 22.0198 (8.9371/2.6060) mem 68106MB [2022-12-20 21:32:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9566 (1.0065) model_time 0.9565 (1.0057) loss 0.9074 (0.8071) grad_norm 10.4685 (8.9277/2.5974) mem 68106MB [2022-12-20 21:32:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9322 (1.0065) model_time 0.9321 (1.0057) loss 0.6890 (0.8074) grad_norm 7.3133 (8.9137/2.6120) mem 68106MB [2022-12-20 21:32:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9304 (1.0066) model_time 0.9302 (1.0058) loss 0.6682 (0.8070) grad_norm 7.0924 (8.8971/2.6207) mem 68106MB [2022-12-20 21:32:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1190/1519] eta 0:05:31 lr 0.000001 time 0.9341 (1.0065) model_time 0.9339 (1.0058) loss 0.7690 (0.8067) grad_norm 6.3654 (8.8988/2.6225) mem 68106MB [2022-12-20 21:33:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1200/1519] eta 0:05:21 lr 0.000001 time 0.9271 (1.0065) model_time 0.9269 (1.0057) loss 0.6824 (0.8068) grad_norm 6.2486 (8.9034/2.6186) mem 68106MB [2022-12-20 21:33:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1210/1519] eta 0:05:11 lr 0.000001 time 0.9451 (1.0065) model_time 0.9449 (1.0058) loss 0.8071 (0.8072) grad_norm 8.7509 (8.8739/2.6040) mem 68106MB [2022-12-20 21:33:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1220/1519] eta 0:05:01 lr 0.000001 time 1.1760 (1.0067) model_time 1.1759 (1.0060) loss 1.1560 (0.8074) grad_norm 7.6400 (8.8349/2.5925) mem 68106MB [2022-12-20 21:33:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1230/1519] eta 0:04:50 lr 0.000001 time 1.0070 (1.0068) model_time 1.0068 (1.0061) loss 0.6602 (0.8070) grad_norm 8.5542 (8.8188/2.5856) mem 68106MB [2022-12-20 21:33:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9294 (1.0068) model_time 0.9292 (1.0060) loss 0.6745 (0.8067) grad_norm 7.1463 (8.8242/2.6032) mem 68106MB [2022-12-20 21:33:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9317 (1.0067) model_time 0.9316 (1.0060) loss 0.8053 (0.8064) grad_norm 10.4435 (8.8395/2.5930) mem 68106MB [2022-12-20 21:34:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9268 (1.0067) model_time 0.9266 (1.0060) loss 0.7109 (0.8062) grad_norm 10.4575 (8.8602/2.6008) mem 68106MB [2022-12-20 21:34:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9285 (1.0069) model_time 0.9284 (1.0062) loss 1.3727 (0.8067) grad_norm 5.5789 (8.7719/2.2406) mem 68106MB [2022-12-20 21:34:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9217 (1.0069) model_time 0.9216 (1.0062) loss 0.7411 (0.8061) grad_norm 8.1005 (8.7687/2.2424) mem 68106MB [2022-12-20 21:34:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9252 (1.0070) model_time 0.9250 (1.0063) loss 0.9406 (0.8067) grad_norm 12.0372 (8.7962/2.2707) mem 68106MB [2022-12-20 21:34:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9304 (1.0069) model_time 0.9302 (1.0062) loss 1.1089 (0.8069) grad_norm 13.6632 (8.8054/2.2969) mem 68106MB [2022-12-20 21:34:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9388 (1.0069) model_time 0.9387 (1.0062) loss 0.6787 (0.8066) grad_norm 9.5804 (8.8464/2.3246) mem 68106MB [2022-12-20 21:35:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9331 (1.0069) model_time 0.9329 (1.0062) loss 0.7067 (0.8065) grad_norm 7.6709 (8.7959/2.2400) mem 68106MB [2022-12-20 21:35:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9344 (1.0070) model_time 0.9343 (1.0063) loss 0.7182 (0.8064) grad_norm 9.2265 (8.8048/2.2376) mem 68106MB [2022-12-20 21:35:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9238 (1.0069) model_time 0.9237 (1.0062) loss 0.6853 (0.8064) grad_norm 7.3939 (8.8019/2.2322) mem 68106MB [2022-12-20 21:35:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9264 (1.0069) model_time 0.9263 (1.0062) loss 0.9124 (0.8060) grad_norm 10.3662 (8.8148/2.2299) mem 68106MB [2022-12-20 21:35:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1360/1519] eta 0:02:40 lr 0.000001 time 0.9311 (1.0070) model_time 0.9310 (1.0063) loss 0.7783 (0.8062) grad_norm 7.7881 (8.8222/2.2246) mem 68106MB [2022-12-20 21:35:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1370/1519] eta 0:02:30 lr 0.000001 time 0.9291 (1.0070) model_time 0.9290 (1.0063) loss 0.7429 (0.8059) grad_norm 10.8257 (8.8208/2.2141) mem 68106MB [2022-12-20 21:36:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9339 (1.0070) model_time 0.9338 (1.0063) loss 0.6841 (0.8058) grad_norm 8.7820 (8.8199/2.1760) mem 68106MB [2022-12-20 21:36:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9338 (1.0070) model_time 0.9336 (1.0063) loss 0.6825 (0.8054) grad_norm 9.2043 (8.8212/2.1922) mem 68106MB [2022-12-20 21:36:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9396 (1.0070) model_time 0.9394 (1.0063) loss 0.7375 (0.8054) grad_norm 10.7514 (8.8075/2.2043) mem 68106MB [2022-12-20 21:36:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9508 (1.0070) model_time 0.9507 (1.0063) loss 0.6650 (0.8054) grad_norm 6.8808 (8.8308/2.2176) mem 68106MB [2022-12-20 21:36:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9328 (1.0069) model_time 0.9327 (1.0062) loss 0.8187 (0.8051) grad_norm 8.6063 (8.8280/2.2270) mem 68106MB [2022-12-20 21:36:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9276 (1.0069) model_time 0.9275 (1.0062) loss 0.6669 (0.8050) grad_norm 11.5932 (8.8231/2.2101) mem 68106MB [2022-12-20 21:37:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9297 (1.0069) model_time 0.9296 (1.0062) loss 0.8747 (0.8050) grad_norm 9.9928 (8.8345/2.2153) mem 68106MB [2022-12-20 21:37:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9345 (1.0068) model_time 0.9343 (1.0062) loss 0.6875 (0.8053) grad_norm 8.3435 (8.8306/2.2181) mem 68106MB [2022-12-20 21:37:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9285 (1.0068) model_time 0.9284 (1.0061) loss 0.8284 (0.8051) grad_norm 9.1061 (8.8226/2.2337) mem 68106MB [2022-12-20 21:37:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9338 (1.0070) model_time 0.9337 (1.0064) loss 0.8671 (0.8051) grad_norm 7.8276 (8.8340/2.2872) mem 68106MB [2022-12-20 21:37:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9282 (1.0071) model_time 0.9281 (1.0064) loss 0.9345 (0.8053) grad_norm 7.9805 (8.8294/2.2886) mem 68106MB [2022-12-20 21:37:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9229 (1.0072) model_time 0.9228 (1.0065) loss 0.8432 (0.8056) grad_norm 8.9107 (8.8149/2.2588) mem 68106MB [2022-12-20 21:38:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9230 (1.0071) model_time 0.9228 (1.0065) loss 0.6688 (0.8057) grad_norm 8.9928 (8.8258/2.2581) mem 68106MB [2022-12-20 21:38:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [93/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9179 (1.0071) model_time 0.9178 (1.0064) loss 0.7578 (0.8056) grad_norm 7.2443 (8.7892/2.1975) mem 68106MB [2022-12-20 21:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 93 training takes 0:25:29 [2022-12-20 21:38:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_93.pth saving...... [2022-12-20 21:38:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_93.pth saved !!! [2022-12-20 21:38:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.660 (0.660) Loss 0.5408 (0.5408) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 21:38:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.293 (0.330) Loss 0.5363 (0.5090) Acc@1 92.014 (92.771) Acc@5 98.264 (98.485) Mem 68106MB [2022-12-20 21:38:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.302 (0.315) Loss 0.4880 (0.5044) Acc@1 90.972 (92.791) Acc@5 99.306 (98.479) Mem 68106MB [2022-12-20 21:39:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.297 (0.309) Loss 0.6335 (0.5112) Acc@1 90.972 (92.552) Acc@5 97.917 (98.466) Mem 68106MB [2022-12-20 21:39:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.296 (0.306) Loss 0.4630 (0.5021) Acc@1 93.750 (92.615) Acc@5 99.306 (98.569) Mem 68106MB [2022-12-20 21:39:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.305) Loss 0.4927 (0.4995) Acc@1 92.708 (92.681) Acc@5 99.653 (98.604) Mem 68106MB [2022-12-20 21:39:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.5080 (0.4992) Acc@1 90.972 (92.623) Acc@5 98.264 (98.583) Mem 68106MB [2022-12-20 21:39:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.303) Loss 0.5459 (0.5003) Acc@1 93.056 (92.576) Acc@5 97.917 (98.567) Mem 68106MB [2022-12-20 21:39:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.302) Loss 0.4305 (0.4990) Acc@1 93.056 (92.605) Acc@5 98.264 (98.594) Mem 68106MB [2022-12-20 21:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:93] * Acc@1 92.579 Acc@5 98.596 [2022-12-20 21:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 21:39:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.59% [2022-12-20 21:39:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][0/1519] eta 0:48:19 lr 0.000001 time 1.9091 (1.9091) model_time 1.2511 (1.2511) loss 0.6809 (0.6809) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 21:39:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][10/1519] eta 0:27:29 lr 0.000001 time 0.9303 (1.0930) model_time 0.9302 (1.0329) loss 0.6602 (0.8268) grad_norm 7.1284 (7.7323/0.7826) mem 68106MB [2022-12-20 21:39:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][20/1519] eta 0:26:21 lr 0.000001 time 0.9203 (1.0552) model_time 0.9201 (1.0235) loss 0.6573 (0.8031) grad_norm 5.4849 (8.5665/3.3518) mem 68106MB [2022-12-20 21:39:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][30/1519] eta 0:25:44 lr 0.000001 time 0.9258 (1.0371) model_time 0.9257 (1.0156) loss 0.7432 (0.8132) grad_norm 5.6235 (8.5314/3.1171) mem 68106MB [2022-12-20 21:39:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][40/1519] eta 0:25:20 lr 0.000001 time 0.9252 (1.0280) model_time 0.9251 (1.0116) loss 0.6622 (0.7980) grad_norm 8.2868 (8.6616/2.8195) mem 68106MB [2022-12-20 21:40:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][50/1519] eta 0:25:03 lr 0.000001 time 0.9196 (1.0233) model_time 0.9194 (1.0101) loss 0.7728 (0.7963) grad_norm 8.4389 (8.7669/2.5546) mem 68106MB [2022-12-20 21:40:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][60/1519] eta 0:24:51 lr 0.000001 time 1.0169 (1.0224) model_time 1.0167 (1.0113) loss 0.7004 (0.8012) grad_norm 9.6820 (8.7041/2.3948) mem 68106MB [2022-12-20 21:40:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][70/1519] eta 0:24:36 lr 0.000001 time 0.9234 (1.0190) model_time 0.9233 (1.0095) loss 0.7293 (0.8081) grad_norm 7.2039 (8.6999/2.2482) mem 68106MB [2022-12-20 21:40:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][80/1519] eta 0:24:24 lr 0.000001 time 0.9657 (1.0175) model_time 0.9655 (1.0091) loss 0.9736 (0.8124) grad_norm 7.4402 (8.6020/2.1586) mem 68106MB [2022-12-20 21:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][90/1519] eta 0:24:12 lr 0.000001 time 0.9267 (1.0167) model_time 0.9266 (1.0092) loss 0.7035 (0.8130) grad_norm 7.5455 (8.4772/2.0841) mem 68106MB [2022-12-20 21:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][100/1519] eta 0:24:01 lr 0.000001 time 0.9256 (1.0156) model_time 0.9254 (1.0088) loss 0.8777 (0.8063) grad_norm 7.5818 (8.4217/2.0046) mem 68106MB [2022-12-20 21:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][110/1519] eta 0:23:49 lr 0.000001 time 0.9251 (1.0144) model_time 0.9249 (1.0082) loss 0.8470 (0.8129) grad_norm 8.5594 (8.5330/2.0300) mem 68106MB [2022-12-20 21:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][120/1519] eta 0:23:37 lr 0.000001 time 0.9247 (1.0135) model_time 0.9245 (1.0078) loss 0.8730 (0.8149) grad_norm 8.6046 (8.5496/2.0164) mem 68106MB [2022-12-20 21:41:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][130/1519] eta 0:23:26 lr 0.000001 time 0.9244 (1.0128) model_time 0.9243 (1.0075) loss 0.7084 (0.8125) grad_norm 7.2079 (8.6806/2.0512) mem 68106MB [2022-12-20 21:41:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][140/1519] eta 0:23:15 lr 0.000001 time 0.9209 (1.0119) model_time 0.9207 (1.0069) loss 0.7146 (0.8114) grad_norm 9.2232 (8.6329/2.0054) mem 68106MB [2022-12-20 21:41:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][150/1519] eta 0:23:04 lr 0.000001 time 0.9242 (1.0114) model_time 0.9241 (1.0068) loss 0.8532 (0.8116) grad_norm 6.9623 (8.5593/1.9682) mem 68106MB [2022-12-20 21:42:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][160/1519] eta 0:22:54 lr 0.000001 time 0.9269 (1.0112) model_time 0.9268 (1.0068) loss 0.9238 (0.8132) grad_norm 7.9001 (8.5004/1.9362) mem 68106MB [2022-12-20 21:42:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][170/1519] eta 0:22:43 lr 0.000001 time 0.9234 (1.0105) model_time 0.9232 (1.0064) loss 0.9493 (0.8123) grad_norm 7.4199 (8.4992/1.8897) mem 68106MB [2022-12-20 21:42:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][180/1519] eta 0:22:32 lr 0.000001 time 1.0031 (1.0103) model_time 1.0029 (1.0063) loss 0.7312 (0.8120) grad_norm 5.3272 (8.4583/1.8878) mem 68106MB [2022-12-20 21:42:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][190/1519] eta 0:22:21 lr 0.000001 time 0.9249 (1.0096) model_time 0.9248 (1.0059) loss 0.7248 (0.8107) grad_norm 7.7180 (8.3781/1.8857) mem 68106MB [2022-12-20 21:42:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][200/1519] eta 0:22:11 lr 0.000001 time 0.9175 (1.0093) model_time 0.9173 (1.0057) loss 0.6675 (0.8105) grad_norm 7.0829 (8.3480/1.8445) mem 68106MB [2022-12-20 21:42:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][210/1519] eta 0:22:00 lr 0.000001 time 0.9241 (1.0089) model_time 0.9240 (1.0054) loss 0.8247 (0.8107) grad_norm 7.0330 (8.3899/1.8473) mem 68106MB [2022-12-20 21:43:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][220/1519] eta 0:21:50 lr 0.000001 time 0.9258 (1.0086) model_time 0.9257 (1.0053) loss 1.1827 (0.8108) grad_norm 7.2522 (8.4073/1.8321) mem 68106MB [2022-12-20 21:43:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][230/1519] eta 0:21:39 lr 0.000001 time 0.9053 (1.0085) model_time 0.9051 (1.0053) loss 1.0129 (0.8081) grad_norm 6.5942 (8.4009/1.8456) mem 68106MB [2022-12-20 21:43:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][240/1519] eta 0:21:30 lr 0.000001 time 1.1653 (1.0094) model_time 1.1652 (1.0063) loss 0.8998 (0.8063) grad_norm 10.8673 (8.4177/1.8407) mem 68106MB [2022-12-20 21:43:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][250/1519] eta 0:21:20 lr 0.000001 time 1.0168 (1.0092) model_time 1.0167 (1.0063) loss 0.8828 (0.8072) grad_norm 9.5852 (8.4121/1.8286) mem 68106MB [2022-12-20 21:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][260/1519] eta 0:21:11 lr 0.000001 time 1.0106 (1.0103) model_time 1.0104 (1.0074) loss 0.7096 (0.8073) grad_norm 8.5394 (8.3997/1.8037) mem 68106MB [2022-12-20 21:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][270/1519] eta 0:21:01 lr 0.000001 time 0.9083 (1.0100) model_time 0.9082 (1.0073) loss 0.6628 (0.8045) grad_norm 8.9091 (8.4398/1.8451) mem 68106MB [2022-12-20 21:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][280/1519] eta 0:20:51 lr 0.000001 time 0.9229 (1.0099) model_time 0.9228 (1.0072) loss 0.6813 (0.8039) grad_norm 6.4661 (8.4313/1.8281) mem 68106MB [2022-12-20 21:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][290/1519] eta 0:20:40 lr 0.000001 time 0.9327 (1.0095) model_time 0.9326 (1.0069) loss 0.7689 (0.8073) grad_norm 9.5155 (8.4475/1.8715) mem 68106MB [2022-12-20 21:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][300/1519] eta 0:20:30 lr 0.000001 time 0.8964 (1.0091) model_time 0.8962 (1.0066) loss 0.8599 (0.8089) grad_norm 7.3962 (8.4620/1.8796) mem 68106MB [2022-12-20 21:44:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][310/1519] eta 0:20:19 lr 0.000001 time 0.9237 (1.0087) model_time 0.9236 (1.0063) loss 0.6576 (0.8092) grad_norm 7.9508 (8.4907/1.8959) mem 68106MB [2022-12-20 21:44:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][320/1519] eta 0:20:09 lr 0.000001 time 0.9163 (1.0086) model_time 0.9162 (1.0063) loss 0.8081 (0.8105) grad_norm 12.5294 (8.5607/1.9155) mem 68106MB [2022-12-20 21:44:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][330/1519] eta 0:19:59 lr 0.000001 time 0.9097 (1.0090) model_time 0.9096 (1.0067) loss 0.6948 (0.8092) grad_norm 8.7295 (8.5840/1.9107) mem 68106MB [2022-12-20 21:45:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][340/1519] eta 0:19:49 lr 0.000001 time 0.9348 (1.0089) model_time 0.9346 (1.0066) loss 0.7574 (0.8072) grad_norm 9.7805 (8.5776/1.9297) mem 68106MB [2022-12-20 21:45:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][350/1519] eta 0:19:39 lr 0.000001 time 0.9296 (1.0088) model_time 0.9295 (1.0066) loss 0.7412 (0.8073) grad_norm 8.0178 (8.5650/1.9081) mem 68106MB [2022-12-20 21:45:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][360/1519] eta 0:19:29 lr 0.000001 time 0.9283 (1.0087) model_time 0.9282 (1.0066) loss 0.6687 (0.8073) grad_norm 7.5433 (8.5998/1.9032) mem 68106MB [2022-12-20 21:45:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][370/1519] eta 0:19:18 lr 0.000001 time 0.9297 (1.0085) model_time 0.9295 (1.0064) loss 0.6804 (0.8058) grad_norm 10.2413 (8.6072/1.8887) mem 68106MB [2022-12-20 21:45:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][380/1519] eta 0:19:08 lr 0.000001 time 0.9261 (1.0083) model_time 0.9260 (1.0062) loss 0.7128 (0.8050) grad_norm 8.4851 (8.6363/1.9196) mem 68106MB [2022-12-20 21:45:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][390/1519] eta 0:18:58 lr 0.000001 time 0.9392 (1.0082) model_time 0.9390 (1.0062) loss 0.6679 (0.8035) grad_norm 9.9612 (8.6648/1.9268) mem 68106MB [2022-12-20 21:46:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][400/1519] eta 0:18:47 lr 0.000001 time 0.9346 (1.0080) model_time 0.9344 (1.0061) loss 0.8349 (0.8029) grad_norm 9.0997 (8.6967/1.9493) mem 68106MB [2022-12-20 21:46:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][410/1519] eta 0:18:37 lr 0.000001 time 0.9253 (1.0081) model_time 0.9252 (1.0062) loss 0.6916 (0.8035) grad_norm 6.9754 (8.6849/1.9499) mem 68106MB [2022-12-20 21:46:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][420/1519] eta 0:18:27 lr 0.000001 time 0.9327 (1.0079) model_time 0.9326 (1.0061) loss 0.6558 (0.8018) grad_norm 7.0280 (8.6593/1.9409) mem 68106MB [2022-12-20 21:46:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][430/1519] eta 0:18:17 lr 0.000001 time 0.9237 (1.0078) model_time 0.9235 (1.0059) loss 0.6732 (0.8023) grad_norm 9.3764 (8.6650/1.9247) mem 68106MB [2022-12-20 21:46:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][440/1519] eta 0:18:07 lr 0.000001 time 0.9259 (1.0079) model_time 0.9257 (1.0061) loss 1.1224 (0.8036) grad_norm 8.7220 (8.6641/1.9076) mem 68106MB [2022-12-20 21:46:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][450/1519] eta 0:17:57 lr 0.000001 time 0.9298 (1.0077) model_time 0.9297 (1.0060) loss 1.0068 (0.8048) grad_norm 12.5219 (8.6761/1.9118) mem 68106MB [2022-12-20 21:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][460/1519] eta 0:17:47 lr 0.000001 time 0.9244 (1.0077) model_time 0.9243 (1.0060) loss 0.7200 (0.8037) grad_norm 6.3660 (8.6657/1.9048) mem 68106MB [2022-12-20 21:47:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][470/1519] eta 0:17:37 lr 0.000001 time 0.9153 (1.0078) model_time 0.9152 (1.0061) loss 0.7749 (0.8034) grad_norm 7.4323 (8.6722/1.9145) mem 68106MB [2022-12-20 21:47:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][480/1519] eta 0:17:26 lr 0.000001 time 0.9228 (1.0076) model_time 0.9227 (1.0059) loss 0.7364 (0.8037) grad_norm 9.0502 (8.6428/1.9108) mem 68106MB [2022-12-20 21:47:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][490/1519] eta 0:17:16 lr 0.000001 time 0.9337 (1.0075) model_time 0.9335 (1.0058) loss 0.6918 (0.8016) grad_norm 6.5837 (8.6207/1.9013) mem 68106MB [2022-12-20 21:47:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][500/1519] eta 0:17:06 lr 0.000001 time 0.9192 (1.0074) model_time 0.9190 (1.0058) loss 0.7090 (0.8025) grad_norm 7.6876 (8.6225/1.9058) mem 68106MB [2022-12-20 21:47:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][510/1519] eta 0:16:56 lr 0.000001 time 0.9041 (1.0074) model_time 0.9040 (1.0058) loss 0.6692 (0.8023) grad_norm 6.4282 (8.6438/1.9282) mem 68106MB [2022-12-20 21:48:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][520/1519] eta 0:16:46 lr 0.000001 time 0.9280 (1.0072) model_time 0.9278 (1.0056) loss 0.7737 (0.8022) grad_norm 10.4124 (8.6465/1.9184) mem 68106MB [2022-12-20 21:48:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][530/1519] eta 0:16:36 lr 0.000001 time 0.9236 (1.0071) model_time 0.9235 (1.0056) loss 0.7164 (0.8016) grad_norm 6.3283 (8.6489/1.9285) mem 68106MB [2022-12-20 21:48:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][540/1519] eta 0:16:25 lr 0.000001 time 0.9225 (1.0070) model_time 0.9224 (1.0055) loss 0.7110 (0.8016) grad_norm 7.8251 (8.6620/1.9593) mem 68106MB [2022-12-20 21:48:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][550/1519] eta 0:16:15 lr 0.000001 time 0.9089 (1.0070) model_time 0.9087 (1.0055) loss 0.6960 (0.8033) grad_norm 7.1186 (8.6346/1.9541) mem 68106MB [2022-12-20 21:48:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][560/1519] eta 0:16:05 lr 0.000001 time 0.9254 (1.0070) model_time 0.9253 (1.0055) loss 0.8125 (0.8028) grad_norm 10.4254 (8.6306/1.9447) mem 68106MB [2022-12-20 21:48:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][570/1519] eta 0:15:55 lr 0.000001 time 0.9220 (1.0071) model_time 0.9219 (1.0057) loss 0.7178 (0.8016) grad_norm 9.9984 (8.6873/2.1091) mem 68106MB [2022-12-20 21:49:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][580/1519] eta 0:15:45 lr 0.000001 time 0.9353 (1.0073) model_time 0.9352 (1.0058) loss 0.8714 (0.8010) grad_norm 9.5381 (8.6815/2.0962) mem 68106MB [2022-12-20 21:49:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][590/1519] eta 0:15:35 lr 0.000001 time 0.9297 (1.0073) model_time 0.9296 (1.0058) loss 0.7224 (0.8001) grad_norm 7.0167 (8.6841/2.0866) mem 68106MB [2022-12-20 21:49:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][600/1519] eta 0:15:25 lr 0.000001 time 0.9233 (1.0071) model_time 0.9232 (1.0057) loss 0.9375 (0.8001) grad_norm 8.5372 (8.6800/2.0709) mem 68106MB [2022-12-20 21:49:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][610/1519] eta 0:15:15 lr 0.000001 time 0.9235 (1.0070) model_time 0.9234 (1.0056) loss 0.9341 (0.8003) grad_norm 6.6698 (8.6844/2.0744) mem 68106MB [2022-12-20 21:49:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][620/1519] eta 0:15:05 lr 0.000001 time 0.9307 (1.0068) model_time 0.9306 (1.0055) loss 0.7494 (0.7997) grad_norm 10.4365 (8.6893/1.9947) mem 68106MB [2022-12-20 21:49:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][630/1519] eta 0:14:55 lr 0.000001 time 0.9180 (1.0068) model_time 0.9179 (1.0054) loss 0.7584 (0.7994) grad_norm 7.8686 (8.6979/1.9782) mem 68106MB [2022-12-20 21:50:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][640/1519] eta 0:14:45 lr 0.000001 time 0.9222 (1.0068) model_time 0.9220 (1.0055) loss 0.8307 (0.7998) grad_norm 8.3398 (8.7400/2.2310) mem 68106MB [2022-12-20 21:50:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][650/1519] eta 0:14:34 lr 0.000001 time 0.9299 (1.0068) model_time 0.9297 (1.0055) loss 0.9273 (0.7996) grad_norm 7.1743 (8.7287/2.2329) mem 68106MB [2022-12-20 21:50:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][660/1519] eta 0:14:24 lr 0.000001 time 0.9216 (1.0067) model_time 0.9215 (1.0054) loss 0.8188 (0.7997) grad_norm 10.3876 (8.7302/2.2380) mem 68106MB [2022-12-20 21:50:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][670/1519] eta 0:14:14 lr 0.000001 time 0.9882 (1.0067) model_time 0.9881 (1.0054) loss 0.6878 (0.7994) grad_norm 8.1927 (8.7167/2.2396) mem 68106MB [2022-12-20 21:50:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][680/1519] eta 0:14:04 lr 0.000001 time 0.9206 (1.0066) model_time 0.9205 (1.0053) loss 0.7990 (0.7994) grad_norm 9.7395 (8.7465/2.2383) mem 68106MB [2022-12-20 21:50:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][690/1519] eta 0:13:54 lr 0.000001 time 0.9192 (1.0064) model_time 0.9190 (1.0052) loss 0.6803 (0.7986) grad_norm 12.0382 (8.7796/2.2400) mem 68106MB [2022-12-20 21:51:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][700/1519] eta 0:13:44 lr 0.000001 time 0.9259 (1.0064) model_time 0.9258 (1.0051) loss 0.6959 (0.7983) grad_norm 8.9311 (8.7911/2.2408) mem 68106MB [2022-12-20 21:51:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][710/1519] eta 0:13:34 lr 0.000001 time 0.9228 (1.0067) model_time 0.9227 (1.0054) loss 0.7861 (0.7985) grad_norm 9.2306 (8.8036/2.2596) mem 68106MB [2022-12-20 21:51:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][720/1519] eta 0:13:24 lr 0.000001 time 0.9310 (1.0069) model_time 0.9308 (1.0057) loss 0.7622 (0.7983) grad_norm 6.8745 (8.7875/2.2532) mem 68106MB [2022-12-20 21:51:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][730/1519] eta 0:13:14 lr 0.000001 time 0.9278 (1.0068) model_time 0.9277 (1.0056) loss 0.7559 (0.7990) grad_norm 11.2390 (8.7593/2.2468) mem 68106MB [2022-12-20 21:51:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][740/1519] eta 0:13:04 lr 0.000001 time 0.9305 (1.0068) model_time 0.9304 (1.0056) loss 0.6991 (0.7999) grad_norm 6.7841 (8.7666/2.2503) mem 68106MB [2022-12-20 21:51:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][750/1519] eta 0:12:54 lr 0.000001 time 0.9060 (1.0069) model_time 0.9059 (1.0057) loss 0.7757 (0.7998) grad_norm 12.7604 (8.7919/2.2558) mem 68106MB [2022-12-20 21:52:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][760/1519] eta 0:12:44 lr 0.000001 time 0.9228 (1.0068) model_time 0.9227 (1.0056) loss 1.0480 (0.7995) grad_norm 9.0871 (8.7911/2.2725) mem 68106MB [2022-12-20 21:52:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][770/1519] eta 0:12:34 lr 0.000001 time 0.8866 (1.0067) model_time 0.8865 (1.0056) loss 0.8311 (0.8015) grad_norm 11.4605 (8.7886/2.2887) mem 68106MB [2022-12-20 21:52:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9248 (1.0067) model_time 0.9247 (1.0055) loss 0.7490 (0.8016) grad_norm 7.6333 (8.7941/2.2865) mem 68106MB [2022-12-20 21:52:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9278 (1.0066) model_time 0.9276 (1.0055) loss 0.8155 (0.8015) grad_norm 6.8994 (8.8116/2.2741) mem 68106MB [2022-12-20 21:52:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][800/1519] eta 0:12:03 lr 0.000001 time 0.9324 (1.0065) model_time 0.9323 (1.0054) loss 0.9357 (0.8025) grad_norm 9.2785 (8.8635/2.3153) mem 68106MB [2022-12-20 21:52:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][810/1519] eta 0:11:53 lr 0.000001 time 0.9261 (1.0065) model_time 0.9259 (1.0054) loss 0.6554 (0.8019) grad_norm 10.2888 (8.8545/2.3109) mem 68106MB [2022-12-20 21:53:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][820/1519] eta 0:11:43 lr 0.000001 time 0.9246 (1.0064) model_time 0.9244 (1.0053) loss 0.6709 (0.8018) grad_norm 8.3484 (8.8435/2.3090) mem 68106MB [2022-12-20 21:53:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][830/1519] eta 0:11:33 lr 0.000001 time 0.9211 (1.0064) model_time 0.9210 (1.0053) loss 0.7353 (0.8015) grad_norm 12.8699 (8.8496/2.3090) mem 68106MB [2022-12-20 21:53:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][840/1519] eta 0:11:23 lr 0.000001 time 0.9557 (1.0063) model_time 0.9556 (1.0052) loss 0.6922 (0.8020) grad_norm 9.0821 (8.8423/2.3050) mem 68106MB [2022-12-20 21:53:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][850/1519] eta 0:11:13 lr 0.000001 time 0.9940 (1.0063) model_time 0.9938 (1.0052) loss 0.8261 (0.8023) grad_norm 8.2494 (8.8592/2.3147) mem 68106MB [2022-12-20 21:53:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][860/1519] eta 0:11:03 lr 0.000001 time 0.9260 (1.0062) model_time 0.9258 (1.0052) loss 0.6885 (0.8015) grad_norm 8.2015 (8.8727/2.3298) mem 68106MB [2022-12-20 21:53:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][870/1519] eta 0:10:53 lr 0.000001 time 0.9238 (1.0062) model_time 0.9237 (1.0052) loss 0.7309 (0.8013) grad_norm 7.8797 (8.8596/2.3109) mem 68106MB [2022-12-20 21:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][880/1519] eta 0:10:43 lr 0.000001 time 0.9261 (1.0063) model_time 0.9260 (1.0053) loss 0.7032 (0.8021) grad_norm 7.8124 (8.8831/2.3134) mem 68106MB [2022-12-20 21:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][890/1519] eta 0:10:32 lr 0.000001 time 0.9225 (1.0063) model_time 0.9223 (1.0052) loss 0.6619 (0.8017) grad_norm 11.7003 (8.8760/2.3010) mem 68106MB [2022-12-20 21:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][900/1519] eta 0:10:22 lr 0.000001 time 0.9752 (1.0064) model_time 0.9751 (1.0053) loss 0.8237 (0.8020) grad_norm 8.7030 (8.8696/2.2948) mem 68106MB [2022-12-20 21:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][910/1519] eta 0:10:12 lr 0.000001 time 0.9317 (1.0063) model_time 0.9316 (1.0052) loss 0.6656 (0.8018) grad_norm 8.9090 (8.8484/2.2960) mem 68106MB [2022-12-20 21:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][920/1519] eta 0:10:02 lr 0.000001 time 0.9277 (1.0062) model_time 0.9276 (1.0052) loss 0.7245 (0.8022) grad_norm 9.5884 (8.8143/2.2863) mem 68106MB [2022-12-20 21:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][930/1519] eta 0:09:52 lr 0.000001 time 0.9288 (1.0062) model_time 0.9286 (1.0051) loss 0.8876 (0.8016) grad_norm 6.7384 (8.7855/2.2860) mem 68106MB [2022-12-20 21:55:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][940/1519] eta 0:09:42 lr 0.000001 time 0.9245 (1.0061) model_time 0.9244 (1.0051) loss 0.8835 (0.8015) grad_norm 6.7043 (8.7811/2.2751) mem 68106MB [2022-12-20 21:55:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][950/1519] eta 0:09:32 lr 0.000001 time 0.9788 (1.0062) model_time 0.9786 (1.0052) loss 0.6730 (0.8018) grad_norm 8.0032 (8.7948/2.2880) mem 68106MB [2022-12-20 21:55:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][960/1519] eta 0:09:22 lr 0.000001 time 0.9384 (1.0062) model_time 0.9382 (1.0052) loss 0.7680 (0.8011) grad_norm 6.9009 (8.7649/2.2855) mem 68106MB [2022-12-20 21:55:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][970/1519] eta 0:09:12 lr 0.000001 time 0.9275 (1.0061) model_time 0.9271 (1.0051) loss 0.6683 (0.8015) grad_norm 10.0347 (8.7719/2.2965) mem 68106MB [2022-12-20 21:55:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][980/1519] eta 0:09:02 lr 0.000001 time 0.9258 (1.0060) model_time 0.9252 (1.0050) loss 0.6698 (0.8014) grad_norm 6.5634 (8.7406/2.2767) mem 68106MB [2022-12-20 21:55:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][990/1519] eta 0:08:52 lr 0.000001 time 0.9288 (1.0060) model_time 0.9287 (1.0050) loss 0.8555 (0.8016) grad_norm 8.0324 (8.7252/2.2783) mem 68106MB [2022-12-20 21:56:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1000/1519] eta 0:08:42 lr 0.000001 time 0.9290 (1.0059) model_time 0.9289 (1.0049) loss 0.7964 (0.8022) grad_norm 12.4620 (8.7246/2.2673) mem 68106MB [2022-12-20 21:56:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1010/1519] eta 0:08:31 lr 0.000001 time 0.9767 (1.0059) model_time 0.9766 (1.0049) loss 0.6903 (0.8026) grad_norm 6.3135 (8.7276/2.2633) mem 68106MB [2022-12-20 21:56:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1020/1519] eta 0:08:21 lr 0.000001 time 0.9504 (1.0058) model_time 0.9503 (1.0048) loss 0.6779 (0.8023) grad_norm 8.1437 (8.7507/2.2657) mem 68106MB [2022-12-20 21:56:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1030/1519] eta 0:08:11 lr 0.000001 time 0.9335 (1.0059) model_time 0.9334 (1.0050) loss 0.9046 (0.8020) grad_norm 9.8298 (8.7622/2.2738) mem 68106MB [2022-12-20 21:56:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1040/1519] eta 0:08:01 lr 0.000001 time 0.9357 (1.0059) model_time 0.9355 (1.0050) loss 1.0040 (0.8016) grad_norm 8.8824 (8.7730/2.2743) mem 68106MB [2022-12-20 21:56:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1050/1519] eta 0:07:51 lr 0.000001 time 0.9290 (1.0058) model_time 0.9288 (1.0049) loss 0.8055 (0.8013) grad_norm 10.7839 (8.7557/2.2666) mem 68106MB [2022-12-20 21:57:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1060/1519] eta 0:07:41 lr 0.000001 time 0.9290 (1.0060) model_time 0.9289 (1.0051) loss 0.6897 (0.8009) grad_norm 11.0232 (8.7501/2.2774) mem 68106MB [2022-12-20 21:57:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1070/1519] eta 0:07:31 lr 0.000001 time 0.9426 (1.0061) model_time 0.9425 (1.0052) loss 0.6609 (0.8016) grad_norm 7.5033 (8.7444/2.2653) mem 68106MB [2022-12-20 21:57:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1080/1519] eta 0:07:21 lr 0.000001 time 0.9244 (1.0061) model_time 0.9243 (1.0052) loss 0.6781 (0.8015) grad_norm 8.9723 (8.7535/2.2620) mem 68106MB [2022-12-20 21:57:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1090/1519] eta 0:07:11 lr 0.000001 time 0.9312 (1.0061) model_time 0.9310 (1.0052) loss 0.7378 (0.8011) grad_norm 7.7249 (8.7625/2.2636) mem 68106MB [2022-12-20 21:57:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1100/1519] eta 0:07:01 lr 0.000001 time 0.9235 (1.0061) model_time 0.9230 (1.0052) loss 0.6942 (0.8007) grad_norm 6.3747 (8.7765/2.2737) mem 68106MB [2022-12-20 21:57:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1110/1519] eta 0:06:51 lr 0.000001 time 0.9236 (1.0060) model_time 0.9235 (1.0051) loss 0.7472 (0.8006) grad_norm 6.3252 (8.7531/2.2534) mem 68106MB [2022-12-20 21:58:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1120/1519] eta 0:06:41 lr 0.000001 time 0.9383 (1.0060) model_time 0.9381 (1.0051) loss 0.8249 (0.8003) grad_norm 6.6465 (8.7211/2.2601) mem 68106MB [2022-12-20 21:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1130/1519] eta 0:06:31 lr 0.000001 time 0.9800 (1.0061) model_time 0.9798 (1.0052) loss 0.7600 (0.8002) grad_norm 7.6699 (8.7123/2.2467) mem 68106MB [2022-12-20 21:58:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1140/1519] eta 0:06:21 lr 0.000001 time 0.9291 (1.0062) model_time 0.9289 (1.0053) loss 0.8408 (0.8010) grad_norm 9.6168 (8.6966/2.2114) mem 68106MB [2022-12-20 21:58:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1150/1519] eta 0:06:11 lr 0.000001 time 0.9349 (1.0061) model_time 0.9347 (1.0052) loss 0.7638 (0.8018) grad_norm 8.5993 (8.7237/2.2076) mem 68106MB [2022-12-20 21:58:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1160/1519] eta 0:06:01 lr 0.000001 time 0.9234 (1.0061) model_time 0.9232 (1.0052) loss 1.0021 (0.8018) grad_norm 7.9149 (8.7186/2.2117) mem 68106MB [2022-12-20 21:58:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1170/1519] eta 0:05:51 lr 0.000001 time 0.9278 (1.0061) model_time 0.9277 (1.0052) loss 0.7145 (0.8013) grad_norm 10.6966 (8.6742/2.0562) mem 68106MB [2022-12-20 21:59:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1180/1519] eta 0:05:41 lr 0.000001 time 0.9257 (1.0062) model_time 0.9255 (1.0053) loss 0.7121 (0.8013) grad_norm 8.7109 (8.6554/2.0621) mem 68106MB [2022-12-20 21:59:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1190/1519] eta 0:05:31 lr 0.000001 time 0.9739 (1.0062) model_time 0.9737 (1.0053) loss 0.9429 (0.8013) grad_norm 7.0969 (8.6728/2.1066) mem 68106MB [2022-12-20 21:59:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1200/1519] eta 0:05:20 lr 0.000001 time 0.9230 (1.0063) model_time 0.9228 (1.0054) loss 0.6997 (0.8015) grad_norm 8.0728 (8.6834/2.1352) mem 68106MB [2022-12-20 21:59:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1210/1519] eta 0:05:10 lr 0.000001 time 0.9370 (1.0063) model_time 0.9368 (1.0054) loss 0.8803 (0.8017) grad_norm 6.9900 (8.6677/2.1393) mem 68106MB [2022-12-20 21:59:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1220/1519] eta 0:05:00 lr 0.000001 time 0.9228 (1.0063) model_time 0.9227 (1.0054) loss 0.6905 (0.8019) grad_norm 9.1076 (8.6439/2.1428) mem 68106MB [2022-12-20 21:59:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1230/1519] eta 0:04:50 lr 0.000001 time 0.9323 (1.0063) model_time 0.9321 (1.0054) loss 0.6933 (0.8015) grad_norm 7.6081 (8.6223/2.1371) mem 68106MB [2022-12-20 22:00:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1240/1519] eta 0:04:40 lr 0.000001 time 0.9284 (1.0062) model_time 0.9283 (1.0054) loss 0.7668 (0.8011) grad_norm 11.7824 (8.5760/1.8741) mem 68106MB [2022-12-20 22:00:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1250/1519] eta 0:04:30 lr 0.000001 time 0.9287 (1.0062) model_time 0.9285 (1.0053) loss 1.0128 (0.8009) grad_norm 10.7495 (8.6163/1.9110) mem 68106MB [2022-12-20 22:00:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1260/1519] eta 0:04:20 lr 0.000001 time 0.9373 (1.0061) model_time 0.9371 (1.0053) loss 0.7445 (0.8005) grad_norm 7.7511 (8.5937/1.9078) mem 68106MB [2022-12-20 22:00:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1270/1519] eta 0:04:10 lr 0.000001 time 0.9287 (1.0061) model_time 0.9285 (1.0053) loss 0.7918 (0.8014) grad_norm 7.3732 (8.6028/1.9193) mem 68106MB [2022-12-20 22:00:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1280/1519] eta 0:04:00 lr 0.000001 time 0.9255 (1.0061) model_time 0.9254 (1.0053) loss 0.6769 (0.8013) grad_norm 7.5989 (8.5952/1.9184) mem 68106MB [2022-12-20 22:00:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1290/1519] eta 0:03:50 lr 0.000001 time 0.9328 (1.0061) model_time 0.9326 (1.0052) loss 0.6732 (0.8010) grad_norm 9.0505 (8.5805/1.9162) mem 68106MB [2022-12-20 22:01:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1300/1519] eta 0:03:40 lr 0.000001 time 0.9343 (1.0061) model_time 0.9342 (1.0052) loss 0.7585 (0.8013) grad_norm 9.0585 (8.5998/1.9478) mem 68106MB [2022-12-20 22:01:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1310/1519] eta 0:03:30 lr 0.000001 time 0.9221 (1.0060) model_time 0.9220 (1.0052) loss 0.8067 (0.8018) grad_norm 7.6106 (8.5669/1.9249) mem 68106MB [2022-12-20 22:01:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1320/1519] eta 0:03:20 lr 0.000001 time 0.9286 (1.0060) model_time 0.9285 (1.0052) loss 0.6893 (0.8020) grad_norm 8.8252 (8.5848/1.9335) mem 68106MB [2022-12-20 22:01:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1330/1519] eta 0:03:10 lr 0.000001 time 0.9311 (1.0059) model_time 0.9310 (1.0051) loss 0.8018 (0.8022) grad_norm 9.2329 (8.5914/1.9189) mem 68106MB [2022-12-20 22:01:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1340/1519] eta 0:03:00 lr 0.000001 time 0.9259 (1.0060) model_time 0.9258 (1.0052) loss 0.8456 (0.8026) grad_norm 7.6637 (8.5739/1.9159) mem 68106MB [2022-12-20 22:01:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1350/1519] eta 0:02:50 lr 0.000001 time 0.9297 (1.0060) model_time 0.9296 (1.0052) loss 0.8082 (0.8022) grad_norm 8.6842 (8.5470/1.9103) mem 68106MB [2022-12-20 22:02:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1360/1519] eta 0:02:39 lr 0.000001 time 0.9484 (1.0061) model_time 0.9483 (1.0053) loss 1.0341 (0.8027) grad_norm 10.6140 (8.5755/1.8946) mem 68106MB [2022-12-20 22:02:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1370/1519] eta 0:02:29 lr 0.000001 time 0.9318 (1.0061) model_time 0.9316 (1.0053) loss 1.0230 (0.8029) grad_norm 7.2869 (8.5819/1.8838) mem 68106MB [2022-12-20 22:02:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1380/1519] eta 0:02:19 lr 0.000001 time 0.9338 (1.0062) model_time 0.9336 (1.0054) loss 0.7014 (0.8030) grad_norm 7.8009 (8.6046/1.8783) mem 68106MB [2022-12-20 22:02:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1390/1519] eta 0:02:09 lr 0.000001 time 0.9335 (1.0062) model_time 0.9334 (1.0054) loss 1.1868 (0.8033) grad_norm 6.8053 (8.6007/1.8804) mem 68106MB [2022-12-20 22:02:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1400/1519] eta 0:01:59 lr 0.000001 time 0.9242 (1.0062) model_time 0.9240 (1.0054) loss 0.8681 (0.8031) grad_norm 7.7500 (8.5541/1.8303) mem 68106MB [2022-12-20 22:02:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1410/1519] eta 0:01:49 lr 0.000001 time 0.9325 (1.0061) model_time 0.9324 (1.0054) loss 0.6916 (0.8032) grad_norm 6.5387 (8.5615/1.8497) mem 68106MB [2022-12-20 22:03:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1420/1519] eta 0:01:39 lr 0.000001 time 0.9261 (1.0061) model_time 0.9259 (1.0053) loss 0.6534 (0.8030) grad_norm 11.0444 (8.5814/1.8533) mem 68106MB [2022-12-20 22:03:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1430/1519] eta 0:01:29 lr 0.000001 time 0.9293 (1.0061) model_time 0.9292 (1.0053) loss 0.6781 (0.8028) grad_norm 8.9222 (8.6021/1.8522) mem 68106MB [2022-12-20 22:03:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1440/1519] eta 0:01:19 lr 0.000001 time 0.9093 (1.0061) model_time 0.9091 (1.0053) loss 0.7270 (0.8027) grad_norm 9.8986 (8.5954/1.8534) mem 68106MB [2022-12-20 22:03:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1450/1519] eta 0:01:09 lr 0.000001 time 0.9224 (1.0063) model_time 0.9222 (1.0055) loss 0.8694 (0.8030) grad_norm 9.6454 (8.5795/1.8329) mem 68106MB [2022-12-20 22:03:47 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1460/1519] eta 0:00:59 lr 0.000001 time 0.9346 (1.0062) model_time 0.9345 (1.0054) loss 0.7100 (0.8028) grad_norm 7.0049 (8.5676/1.8119) mem 68106MB [2022-12-20 22:03:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1470/1519] eta 0:00:49 lr 0.000001 time 0.9192 (1.0062) model_time 0.9190 (1.0054) loss 0.6877 (0.8021) grad_norm 8.5636 (8.5669/1.8119) mem 68106MB [2022-12-20 22:04:07 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1480/1519] eta 0:00:39 lr 0.000001 time 0.9199 (1.0062) model_time 0.9197 (1.0054) loss 0.9369 (0.8020) grad_norm 8.4091 (8.5548/1.8049) mem 68106MB [2022-12-20 22:04:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1490/1519] eta 0:00:29 lr 0.000001 time 0.9248 (1.0061) model_time 0.9246 (1.0054) loss 0.7887 (0.8028) grad_norm 6.7694 (8.5616/1.8103) mem 68106MB [2022-12-20 22:04:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1500/1519] eta 0:00:19 lr 0.000001 time 0.9311 (1.0061) model_time 0.9310 (1.0053) loss 0.6927 (0.8026) grad_norm 9.3015 (8.5634/1.8034) mem 68106MB [2022-12-20 22:04:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [94/100][1510/1519] eta 0:00:09 lr 0.000001 time 0.9156 (1.0060) model_time 0.9155 (1.0053) loss 0.8066 (0.8020) grad_norm 7.7986 (8.5853/1.8339) mem 68106MB [2022-12-20 22:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 94 training takes 0:25:28 [2022-12-20 22:04:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_94.pth saving...... [2022-12-20 22:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_94.pth saved !!! [2022-12-20 22:05:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.660 (0.660) Loss 0.5400 (0.5400) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 22:05:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.329) Loss 0.5341 (0.5080) Acc@1 92.361 (92.835) Acc@5 98.611 (98.580) Mem 68106MB [2022-12-20 22:05:17 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4884 (0.5037) Acc@1 90.972 (92.774) Acc@5 99.306 (98.528) Mem 68106MB [2022-12-20 22:05:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.308 (0.313) Loss 0.6336 (0.5108) Acc@1 90.625 (92.563) Acc@5 97.917 (98.499) Mem 68106MB [2022-12-20 22:05:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.310) Loss 0.4615 (0.5016) Acc@1 93.750 (92.641) Acc@5 99.306 (98.594) Mem 68106MB [2022-12-20 22:05:27 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.308) Loss 0.4949 (0.4990) Acc@1 92.708 (92.715) Acc@5 99.653 (98.638) Mem 68106MB [2022-12-20 22:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.298 (0.307) Loss 0.5080 (0.4988) Acc@1 90.972 (92.629) Acc@5 98.264 (98.594) Mem 68106MB [2022-12-20 22:05:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.293 (0.306) Loss 0.5456 (0.5000) Acc@1 93.056 (92.591) Acc@5 97.917 (98.577) Mem 68106MB [2022-12-20 22:05:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.304) Loss 0.4318 (0.4987) Acc@1 93.056 (92.618) Acc@5 98.264 (98.603) Mem 68106MB [2022-12-20 22:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:94] * Acc@1 92.588 Acc@5 98.604 [2022-12-20 22:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 22:05:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.59% [2022-12-20 22:05:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][0/1519] eta 0:46:51 lr 0.000001 time 1.8507 (1.8507) model_time 1.1682 (1.1682) loss 0.8092 (0.8092) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 22:05:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][10/1519] eta 0:27:05 lr 0.000001 time 0.9215 (1.0774) model_time 0.9214 (1.0150) loss 0.8325 (0.7820) grad_norm 8.1960 (8.5293/1.8123) mem 68106MB [2022-12-20 22:05:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][20/1519] eta 0:25:59 lr 0.000001 time 0.9326 (1.0402) model_time 0.9322 (1.0074) loss 0.6706 (0.7855) grad_norm 9.5262 (8.3173/1.7306) mem 68106MB [2022-12-20 22:06:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][30/1519] eta 0:25:32 lr 0.000001 time 0.9900 (1.0290) model_time 0.9898 (1.0067) loss 0.9345 (0.7810) grad_norm 9.0677 (8.4276/1.6143) mem 68106MB [2022-12-20 22:06:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][40/1519] eta 0:25:10 lr 0.000001 time 0.9261 (1.0211) model_time 0.9259 (1.0041) loss 0.7065 (0.7935) grad_norm 6.7755 (8.1747/1.5171) mem 68106MB [2022-12-20 22:06:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][50/1519] eta 0:24:53 lr 0.000001 time 0.9222 (1.0166) model_time 0.9221 (1.0029) loss 0.7071 (0.7969) grad_norm 10.9200 (8.3846/1.5709) mem 68106MB [2022-12-20 22:06:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][60/1519] eta 0:24:41 lr 0.000001 time 0.9124 (1.0152) model_time 0.9122 (1.0037) loss 0.8514 (0.7915) grad_norm 8.8917 (8.3437/1.4971) mem 68106MB [2022-12-20 22:06:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][70/1519] eta 0:24:28 lr 0.000001 time 0.9294 (1.0136) model_time 0.9292 (1.0037) loss 0.7336 (0.7858) grad_norm 9.9464 (8.4202/1.4771) mem 68106MB [2022-12-20 22:06:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][80/1519] eta 0:24:17 lr 0.000001 time 0.9875 (1.0132) model_time 0.9874 (1.0044) loss 0.7303 (0.7827) grad_norm 6.9413 (8.6019/1.6029) mem 68106MB [2022-12-20 22:07:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][90/1519] eta 0:24:05 lr 0.000001 time 0.9244 (1.0115) model_time 0.9242 (1.0036) loss 0.8145 (0.7783) grad_norm 6.0582 (8.5585/1.6082) mem 68106MB [2022-12-20 22:07:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][100/1519] eta 0:23:55 lr 0.000001 time 0.9266 (1.0119) model_time 0.9265 (1.0048) loss 0.7076 (0.7764) grad_norm 5.5185 (8.5447/1.6610) mem 68106MB [2022-12-20 22:07:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][110/1519] eta 0:23:44 lr 0.000001 time 0.9271 (1.0111) model_time 0.9268 (1.0046) loss 0.6882 (0.7709) grad_norm 7.8350 (8.5172/1.6228) mem 68106MB [2022-12-20 22:07:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][120/1519] eta 0:23:33 lr 0.000001 time 0.9195 (1.0100) model_time 0.9194 (1.0041) loss 0.6834 (0.7707) grad_norm 7.3282 (8.4475/1.5839) mem 68106MB [2022-12-20 22:07:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][130/1519] eta 0:23:22 lr 0.000001 time 0.9207 (1.0094) model_time 0.9206 (1.0038) loss 0.6740 (0.7750) grad_norm 12.2096 (8.5084/1.6640) mem 68106MB [2022-12-20 22:07:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][140/1519] eta 0:23:11 lr 0.000001 time 0.9216 (1.0090) model_time 0.9211 (1.0039) loss 0.6864 (0.7746) grad_norm 6.9647 (8.5382/1.6367) mem 68106MB [2022-12-20 22:08:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][150/1519] eta 0:23:02 lr 0.000001 time 0.9273 (1.0096) model_time 0.9272 (1.0047) loss 1.0043 (0.7744) grad_norm 11.8276 (8.5731/1.6284) mem 68106MB [2022-12-20 22:08:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][160/1519] eta 0:22:51 lr 0.000001 time 0.9231 (1.0095) model_time 0.9229 (1.0049) loss 0.7741 (0.7734) grad_norm 7.0258 (8.5529/1.6097) mem 68106MB [2022-12-20 22:08:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][170/1519] eta 0:22:41 lr 0.000001 time 0.9359 (1.0095) model_time 0.9358 (1.0052) loss 0.8515 (0.7745) grad_norm 7.6090 (8.4846/1.5926) mem 68106MB [2022-12-20 22:08:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][180/1519] eta 0:22:31 lr 0.000001 time 0.9750 (1.0094) model_time 0.9749 (1.0053) loss 0.7988 (0.7763) grad_norm 8.6885 (8.4964/1.5634) mem 68106MB [2022-12-20 22:08:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][190/1519] eta 0:22:21 lr 0.000001 time 0.9324 (1.0092) model_time 0.9322 (1.0053) loss 0.7970 (0.7775) grad_norm 7.2468 (8.5346/1.5726) mem 68106MB [2022-12-20 22:08:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][200/1519] eta 0:22:10 lr 0.000001 time 0.9222 (1.0087) model_time 0.9221 (1.0049) loss 0.8022 (0.7774) grad_norm 8.3647 (8.5396/1.5676) mem 68106MB [2022-12-20 22:09:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][210/1519] eta 0:22:00 lr 0.000001 time 0.9790 (1.0085) model_time 0.9789 (1.0049) loss 0.8211 (0.7775) grad_norm 7.3864 (8.4813/1.5556) mem 68106MB [2022-12-20 22:09:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][220/1519] eta 0:21:49 lr 0.000001 time 0.9311 (1.0084) model_time 0.9310 (1.0049) loss 0.9774 (0.7774) grad_norm 8.5693 (8.4613/1.5407) mem 68106MB [2022-12-20 22:09:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][230/1519] eta 0:21:39 lr 0.000001 time 0.9739 (1.0083) model_time 0.9737 (1.0049) loss 0.7059 (0.7776) grad_norm 12.7173 (8.5122/1.6064) mem 68106MB [2022-12-20 22:09:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][240/1519] eta 0:21:29 lr 0.000001 time 0.9232 (1.0082) model_time 0.9230 (1.0050) loss 0.6812 (0.7772) grad_norm 7.6357 (8.4854/1.5964) mem 68106MB [2022-12-20 22:09:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][250/1519] eta 0:21:19 lr 0.000001 time 0.9286 (1.0082) model_time 0.9285 (1.0051) loss 0.6850 (0.7770) grad_norm 13.1061 (8.4946/1.6350) mem 68106MB [2022-12-20 22:10:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][260/1519] eta 0:21:09 lr 0.000001 time 0.9298 (1.0080) model_time 0.9297 (1.0051) loss 0.7395 (0.7766) grad_norm 7.5837 (8.4925/1.6111) mem 68106MB [2022-12-20 22:10:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][270/1519] eta 0:20:58 lr 0.000001 time 0.9242 (1.0077) model_time 0.9240 (1.0048) loss 0.6959 (0.7747) grad_norm 11.1471 (8.4954/1.6099) mem 68106MB [2022-12-20 22:10:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][280/1519] eta 0:20:48 lr 0.000001 time 0.9200 (1.0073) model_time 0.9198 (1.0045) loss 0.7902 (0.7742) grad_norm 11.5248 (8.5030/1.6313) mem 68106MB [2022-12-20 22:10:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][290/1519] eta 0:20:37 lr 0.000001 time 0.9356 (1.0071) model_time 0.9355 (1.0044) loss 0.6874 (0.7757) grad_norm 7.4752 (8.5015/1.6329) mem 68106MB [2022-12-20 22:10:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][300/1519] eta 0:20:27 lr 0.000001 time 0.9387 (1.0070) model_time 0.9385 (1.0044) loss 0.8533 (0.7747) grad_norm 6.1600 (8.4907/1.6376) mem 68106MB [2022-12-20 22:10:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][310/1519] eta 0:20:17 lr 0.000001 time 0.9211 (1.0071) model_time 0.9209 (1.0045) loss 0.7688 (0.7807) grad_norm 7.5420 (8.5013/1.6440) mem 68106MB [2022-12-20 22:11:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][320/1519] eta 0:20:07 lr 0.000001 time 0.9229 (1.0068) model_time 0.9227 (1.0043) loss 0.7855 (0.7790) grad_norm 10.9745 (8.5166/1.6311) mem 68106MB [2022-12-20 22:11:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][330/1519] eta 0:19:56 lr 0.000001 time 0.9205 (1.0066) model_time 0.9204 (1.0042) loss 0.7260 (0.7790) grad_norm 9.0968 (8.5883/1.7388) mem 68106MB [2022-12-20 22:11:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][340/1519] eta 0:19:46 lr 0.000001 time 0.9265 (1.0064) model_time 0.9263 (1.0041) loss 0.6718 (0.7801) grad_norm 7.0268 (8.5648/1.7231) mem 68106MB [2022-12-20 22:11:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][350/1519] eta 0:19:36 lr 0.000001 time 0.9293 (1.0066) model_time 0.9292 (1.0044) loss 0.8661 (0.7798) grad_norm 6.5634 (8.5645/1.7553) mem 68106MB [2022-12-20 22:11:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][360/1519] eta 0:19:26 lr 0.000001 time 1.0042 (1.0068) model_time 1.0041 (1.0045) loss 0.6970 (0.7802) grad_norm 7.9762 (8.6072/1.7696) mem 68106MB [2022-12-20 22:11:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][370/1519] eta 0:19:16 lr 0.000001 time 0.9289 (1.0065) model_time 0.9287 (1.0044) loss 0.6979 (0.7808) grad_norm 10.4202 (8.6030/1.7645) mem 68106MB [2022-12-20 22:12:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][380/1519] eta 0:19:06 lr 0.000001 time 0.9858 (1.0066) model_time 0.9857 (1.0044) loss 0.7618 (0.7811) grad_norm 10.1178 (8.6434/1.7913) mem 68106MB [2022-12-20 22:12:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][390/1519] eta 0:18:56 lr 0.000001 time 0.9244 (1.0067) model_time 0.9243 (1.0046) loss 0.7370 (0.7823) grad_norm 7.6859 (8.6289/1.7836) mem 68106MB [2022-12-20 22:12:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][400/1519] eta 0:18:46 lr 0.000001 time 0.9277 (1.0066) model_time 0.9274 (1.0046) loss 0.8410 (0.7814) grad_norm 7.4821 (8.6278/1.7701) mem 68106MB [2022-12-20 22:12:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][410/1519] eta 0:18:37 lr 0.000001 time 0.9805 (1.0075) model_time 0.9804 (1.0055) loss 0.7508 (0.7809) grad_norm 6.8928 (8.6038/1.7618) mem 68106MB [2022-12-20 22:12:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][420/1519] eta 0:18:27 lr 0.000001 time 0.9682 (1.0075) model_time 0.9681 (1.0055) loss 0.8899 (0.7808) grad_norm 10.1425 (8.5962/1.7637) mem 68106MB [2022-12-20 22:12:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][430/1519] eta 0:18:17 lr 0.000001 time 0.9278 (1.0074) model_time 0.9276 (1.0055) loss 0.6699 (0.7814) grad_norm 9.4800 (8.5945/1.7561) mem 68106MB [2022-12-20 22:13:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][440/1519] eta 0:18:06 lr 0.000001 time 0.9268 (1.0072) model_time 0.9266 (1.0053) loss 0.7064 (0.7814) grad_norm 7.0517 (8.5852/1.7517) mem 68106MB [2022-12-20 22:13:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][450/1519] eta 0:17:56 lr 0.000001 time 0.9338 (1.0071) model_time 0.9336 (1.0053) loss 0.7893 (0.7812) grad_norm 6.4950 (8.5750/1.7525) mem 68106MB [2022-12-20 22:13:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][460/1519] eta 0:17:46 lr 0.000001 time 0.9785 (1.0071) model_time 0.9783 (1.0052) loss 0.6853 (0.7813) grad_norm 7.7233 (8.5707/1.7484) mem 68106MB [2022-12-20 22:13:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][470/1519] eta 0:17:36 lr 0.000001 time 0.9231 (1.0071) model_time 0.9230 (1.0053) loss 0.6629 (0.7810) grad_norm 9.6126 (8.5583/1.7429) mem 68106MB [2022-12-20 22:13:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][480/1519] eta 0:17:26 lr 0.000001 time 0.9224 (1.0070) model_time 0.9223 (1.0052) loss 0.8828 (0.7807) grad_norm 7.8774 (8.5766/1.7453) mem 68106MB [2022-12-20 22:13:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][490/1519] eta 0:17:16 lr 0.000001 time 0.9207 (1.0068) model_time 0.9205 (1.0051) loss 1.1296 (0.7811) grad_norm 9.2081 (8.5724/1.7546) mem 68106MB [2022-12-20 22:14:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][500/1519] eta 0:17:05 lr 0.000001 time 0.9243 (1.0068) model_time 0.9242 (1.0051) loss 0.7209 (0.7823) grad_norm 8.9801 (8.5754/1.7404) mem 68106MB [2022-12-20 22:14:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][510/1519] eta 0:16:55 lr 0.000001 time 0.9224 (1.0067) model_time 0.9223 (1.0050) loss 0.6867 (0.7826) grad_norm 10.6482 (8.5726/1.7393) mem 68106MB [2022-12-20 22:14:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][520/1519] eta 0:16:45 lr 0.000001 time 0.9257 (1.0065) model_time 0.9256 (1.0049) loss 0.7522 (0.7815) grad_norm 6.6988 (8.5944/1.7524) mem 68106MB [2022-12-20 22:14:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][530/1519] eta 0:16:35 lr 0.000001 time 0.9247 (1.0064) model_time 0.9244 (1.0048) loss 0.6791 (0.7822) grad_norm 7.5613 (8.6138/1.7847) mem 68106MB [2022-12-20 22:14:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][540/1519] eta 0:16:25 lr 0.000001 time 0.9254 (1.0063) model_time 0.9253 (1.0047) loss 0.8212 (0.7816) grad_norm 7.7335 (8.6114/1.7779) mem 68106MB [2022-12-20 22:14:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][550/1519] eta 0:16:15 lr 0.000001 time 0.9245 (1.0063) model_time 0.9243 (1.0048) loss 0.7521 (0.7816) grad_norm 9.7102 (8.6190/1.7986) mem 68106MB [2022-12-20 22:15:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][560/1519] eta 0:16:05 lr 0.000001 time 0.9967 (1.0063) model_time 0.9966 (1.0047) loss 1.1347 (0.7822) grad_norm 6.6747 (8.6282/1.8400) mem 68106MB [2022-12-20 22:15:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][570/1519] eta 0:15:54 lr 0.000001 time 0.9179 (1.0061) model_time 0.9177 (1.0046) loss 0.7887 (0.7820) grad_norm 11.5674 (8.6159/1.8459) mem 68106MB [2022-12-20 22:15:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][580/1519] eta 0:15:44 lr 0.000001 time 0.9298 (1.0060) model_time 0.9297 (1.0045) loss 0.8685 (0.7834) grad_norm 8.6452 (8.6194/1.8439) mem 68106MB [2022-12-20 22:15:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][590/1519] eta 0:15:34 lr 0.000001 time 0.9269 (1.0060) model_time 0.9267 (1.0045) loss 0.6596 (0.7837) grad_norm 9.2878 (8.6206/1.8316) mem 68106MB [2022-12-20 22:15:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][600/1519] eta 0:15:24 lr 0.000001 time 0.9730 (1.0060) model_time 0.9728 (1.0045) loss 0.7511 (0.7842) grad_norm 5.9236 (8.6255/1.8593) mem 68106MB [2022-12-20 22:15:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][610/1519] eta 0:15:14 lr 0.000001 time 0.9324 (1.0058) model_time 0.9322 (1.0044) loss 0.7253 (0.7834) grad_norm 6.9595 (8.6236/1.8513) mem 68106MB [2022-12-20 22:16:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][620/1519] eta 0:15:04 lr 0.000001 time 0.9288 (1.0059) model_time 0.9286 (1.0044) loss 0.6993 (0.7826) grad_norm 10.9472 (8.6585/1.8821) mem 68106MB [2022-12-20 22:16:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][630/1519] eta 0:14:54 lr 0.000001 time 0.9309 (1.0058) model_time 0.9307 (1.0044) loss 0.7086 (0.7831) grad_norm 8.9517 (8.6586/1.8862) mem 68106MB [2022-12-20 22:16:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][640/1519] eta 0:14:44 lr 0.000001 time 1.0256 (1.0058) model_time 1.0254 (1.0044) loss 0.6608 (0.7830) grad_norm 9.8806 (8.6677/1.8926) mem 68106MB [2022-12-20 22:16:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][650/1519] eta 0:14:33 lr 0.000001 time 0.9211 (1.0057) model_time 0.9210 (1.0043) loss 0.7248 (0.7839) grad_norm 7.8057 (8.6432/1.8866) mem 68106MB [2022-12-20 22:16:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][660/1519] eta 0:14:23 lr 0.000001 time 0.9239 (1.0058) model_time 0.9238 (1.0044) loss 0.7058 (0.7840) grad_norm 7.7110 (8.6407/1.9018) mem 68106MB [2022-12-20 22:16:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][670/1519] eta 0:14:13 lr 0.000001 time 0.9273 (1.0057) model_time 0.9272 (1.0044) loss 0.7046 (0.7837) grad_norm 6.4198 (8.6242/1.9016) mem 68106MB [2022-12-20 22:17:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][680/1519] eta 0:14:03 lr 0.000001 time 0.9261 (1.0058) model_time 0.9260 (1.0044) loss 0.8162 (0.7836) grad_norm 11.2752 (8.5963/1.8937) mem 68106MB [2022-12-20 22:17:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][690/1519] eta 0:13:53 lr 0.000001 time 0.9232 (1.0058) model_time 0.9230 (1.0044) loss 0.7026 (0.7831) grad_norm 7.9056 (8.5721/1.8982) mem 68106MB [2022-12-20 22:17:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][700/1519] eta 0:13:43 lr 0.000001 time 0.9273 (1.0057) model_time 0.9272 (1.0044) loss 0.7009 (0.7829) grad_norm 8.2105 (8.5617/1.8841) mem 68106MB [2022-12-20 22:17:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][710/1519] eta 0:13:33 lr 0.000001 time 0.9235 (1.0056) model_time 0.9233 (1.0043) loss 0.7732 (0.7838) grad_norm 15.5839 (8.6179/1.9547) mem 68106MB [2022-12-20 22:17:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][720/1519] eta 0:13:23 lr 0.000001 time 1.0101 (1.0056) model_time 1.0100 (1.0043) loss 0.7815 (0.7831) grad_norm 13.1676 (8.6455/1.9828) mem 68106MB [2022-12-20 22:17:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][730/1519] eta 0:13:13 lr 0.000001 time 0.9258 (1.0058) model_time 0.9257 (1.0045) loss 0.7004 (0.7840) grad_norm 13.2511 (8.6457/1.9812) mem 68106MB [2022-12-20 22:18:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][740/1519] eta 0:13:03 lr 0.000001 time 0.9098 (1.0058) model_time 0.9096 (1.0045) loss 0.6568 (0.7844) grad_norm 10.0123 (8.6352/1.9818) mem 68106MB [2022-12-20 22:18:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][750/1519] eta 0:12:53 lr 0.000001 time 0.9285 (1.0057) model_time 0.9283 (1.0044) loss 0.6878 (0.7840) grad_norm 7.9712 (8.6187/1.9839) mem 68106MB [2022-12-20 22:18:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][760/1519] eta 0:12:43 lr 0.000001 time 0.9724 (1.0057) model_time 0.9722 (1.0045) loss 0.8781 (0.7838) grad_norm 8.8274 (8.6143/1.9794) mem 68106MB [2022-12-20 22:18:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][770/1519] eta 0:12:33 lr 0.000001 time 0.9270 (1.0056) model_time 0.9269 (1.0044) loss 0.7936 (0.7847) grad_norm 6.0176 (8.6409/1.9831) mem 68106MB [2022-12-20 22:18:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][780/1519] eta 0:12:23 lr 0.000001 time 0.9887 (1.0056) model_time 0.9886 (1.0044) loss 0.6682 (0.7847) grad_norm 12.2803 (8.6758/2.0281) mem 68106MB [2022-12-20 22:18:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][790/1519] eta 0:12:13 lr 0.000001 time 0.9062 (1.0055) model_time 0.9060 (1.0044) loss 0.6624 (0.7849) grad_norm 8.7281 (8.6569/2.0214) mem 68106MB [2022-12-20 22:19:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][800/1519] eta 0:12:03 lr 0.000001 time 0.9279 (1.0056) model_time 0.9277 (1.0044) loss 0.6703 (0.7846) grad_norm 8.0646 (8.6612/2.0232) mem 68106MB [2022-12-20 22:19:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][810/1519] eta 0:11:52 lr 0.000001 time 0.9225 (1.0055) model_time 0.9223 (1.0044) loss 0.6715 (0.7844) grad_norm 8.2970 (8.6921/2.0185) mem 68106MB [2022-12-20 22:19:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][820/1519] eta 0:11:42 lr 0.000001 time 0.9182 (1.0054) model_time 0.9180 (1.0043) loss 0.9478 (0.7847) grad_norm 6.5482 (8.6757/2.0238) mem 68106MB [2022-12-20 22:19:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][830/1519] eta 0:11:32 lr 0.000001 time 0.9253 (1.0054) model_time 0.9251 (1.0042) loss 0.6523 (0.7848) grad_norm 6.5618 (8.6597/2.0044) mem 68106MB [2022-12-20 22:19:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][840/1519] eta 0:11:22 lr 0.000001 time 0.9318 (1.0054) model_time 0.9315 (1.0042) loss 1.0246 (0.7849) grad_norm 6.6877 (8.6595/2.0041) mem 68106MB [2022-12-20 22:19:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][850/1519] eta 0:11:12 lr 0.000000 time 0.9230 (1.0053) model_time 0.9228 (1.0042) loss 0.7538 (0.7843) grad_norm 9.7624 (8.6605/1.9881) mem 68106MB [2022-12-20 22:20:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][860/1519] eta 0:11:02 lr 0.000000 time 0.9225 (1.0053) model_time 0.9224 (1.0042) loss 0.7289 (0.7844) grad_norm 7.1565 (8.6381/2.0008) mem 68106MB [2022-12-20 22:20:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][870/1519] eta 0:10:52 lr 0.000000 time 0.9212 (1.0055) model_time 0.9209 (1.0044) loss 0.9175 (0.7848) grad_norm 6.1846 (8.6217/1.9989) mem 68106MB [2022-12-20 22:20:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][880/1519] eta 0:10:42 lr 0.000000 time 0.8873 (1.0056) model_time 0.8871 (1.0045) loss 0.7600 (0.7855) grad_norm 8.0387 (8.6249/1.9841) mem 68106MB [2022-12-20 22:20:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][890/1519] eta 0:10:32 lr 0.000000 time 0.9184 (1.0054) model_time 0.9183 (1.0044) loss 0.7937 (0.7855) grad_norm 11.2682 (8.6318/1.9823) mem 68106MB [2022-12-20 22:20:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][900/1519] eta 0:10:22 lr 0.000000 time 0.9306 (1.0054) model_time 0.9304 (1.0044) loss 0.6992 (0.7857) grad_norm 9.5626 (8.6356/1.9738) mem 68106MB [2022-12-20 22:20:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][910/1519] eta 0:10:12 lr 0.000000 time 0.9311 (1.0054) model_time 0.9309 (1.0043) loss 0.7493 (0.7863) grad_norm 7.9630 (8.6300/1.9695) mem 68106MB [2022-12-20 22:21:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][920/1519] eta 0:10:02 lr 0.000000 time 0.9285 (1.0054) model_time 0.9283 (1.0043) loss 0.7352 (0.7871) grad_norm 6.2700 (8.6025/1.9726) mem 68106MB [2022-12-20 22:21:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][930/1519] eta 0:09:52 lr 0.000000 time 0.9290 (1.0053) model_time 0.9288 (1.0042) loss 0.6686 (0.7872) grad_norm 9.0691 (8.5540/1.9148) mem 68106MB [2022-12-20 22:21:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][940/1519] eta 0:09:42 lr 0.000000 time 0.9213 (1.0052) model_time 0.9211 (1.0042) loss 0.6710 (0.7882) grad_norm 14.3029 (8.5879/1.9421) mem 68106MB [2022-12-20 22:21:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][950/1519] eta 0:09:31 lr 0.000000 time 0.9284 (1.0052) model_time 0.9282 (1.0041) loss 0.8385 (0.7887) grad_norm 9.3166 (8.6021/1.9470) mem 68106MB [2022-12-20 22:21:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][960/1519] eta 0:09:22 lr 0.000000 time 0.9392 (1.0055) model_time 0.9390 (1.0045) loss 0.9060 (0.7897) grad_norm 7.0645 (8.5834/1.9595) mem 68106MB [2022-12-20 22:21:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][970/1519] eta 0:09:12 lr 0.000000 time 1.0223 (1.0055) model_time 1.0221 (1.0045) loss 0.7475 (0.7901) grad_norm 9.5246 (8.6035/1.9708) mem 68106MB [2022-12-20 22:22:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][980/1519] eta 0:09:01 lr 0.000000 time 0.9254 (1.0055) model_time 0.9252 (1.0045) loss 0.7463 (0.7906) grad_norm 7.1401 (8.5522/1.9522) mem 68106MB [2022-12-20 22:22:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][990/1519] eta 0:08:51 lr 0.000000 time 0.9256 (1.0055) model_time 0.9254 (1.0045) loss 0.7301 (0.7899) grad_norm 8.1442 (8.5508/1.9501) mem 68106MB [2022-12-20 22:22:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1000/1519] eta 0:08:41 lr 0.000000 time 0.9271 (1.0055) model_time 0.9270 (1.0045) loss 0.8912 (0.7903) grad_norm 15.4698 (8.5859/1.9964) mem 68106MB [2022-12-20 22:22:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1010/1519] eta 0:08:31 lr 0.000000 time 1.0539 (1.0055) model_time 1.0538 (1.0045) loss 0.7764 (0.7912) grad_norm 8.7806 (8.5946/1.9978) mem 68106MB [2022-12-20 22:22:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1020/1519] eta 0:08:21 lr 0.000000 time 0.9188 (1.0055) model_time 0.9186 (1.0045) loss 0.7557 (0.7914) grad_norm 7.7187 (8.6030/2.0027) mem 68106MB [2022-12-20 22:22:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1030/1519] eta 0:08:11 lr 0.000000 time 0.9262 (1.0054) model_time 0.9260 (1.0044) loss 0.7183 (0.7914) grad_norm 6.0460 (8.6613/2.1858) mem 68106MB [2022-12-20 22:23:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1040/1519] eta 0:08:01 lr 0.000000 time 0.9208 (1.0055) model_time 0.9206 (1.0045) loss 0.7107 (0.7923) grad_norm 9.3760 (8.6522/2.1861) mem 68106MB [2022-12-20 22:23:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1050/1519] eta 0:07:51 lr 0.000000 time 0.9303 (1.0055) model_time 0.9301 (1.0045) loss 0.7612 (0.7922) grad_norm 7.9471 (8.6681/2.1775) mem 68106MB [2022-12-20 22:23:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1060/1519] eta 0:07:41 lr 0.000000 time 1.0019 (1.0056) model_time 1.0018 (1.0046) loss 0.6883 (0.7922) grad_norm 8.0746 (8.6742/2.1793) mem 68106MB [2022-12-20 22:23:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1070/1519] eta 0:07:31 lr 0.000000 time 0.9222 (1.0055) model_time 0.9221 (1.0045) loss 0.7938 (0.7923) grad_norm 7.9337 (8.6936/2.1852) mem 68106MB [2022-12-20 22:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1080/1519] eta 0:07:21 lr 0.000000 time 0.9205 (1.0055) model_time 0.9204 (1.0045) loss 0.7673 (0.7923) grad_norm 7.4963 (8.7016/2.2086) mem 68106MB [2022-12-20 22:23:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1090/1519] eta 0:07:11 lr 0.000000 time 0.9317 (1.0054) model_time 0.9316 (1.0045) loss 1.2160 (0.7924) grad_norm 9.1463 (8.7043/2.1914) mem 68106MB [2022-12-20 22:24:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1100/1519] eta 0:07:01 lr 0.000000 time 0.9265 (1.0055) model_time 0.9264 (1.0045) loss 0.6603 (0.7921) grad_norm 8.9675 (8.6985/2.1953) mem 68106MB [2022-12-20 22:24:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1110/1519] eta 0:06:51 lr 0.000000 time 0.9241 (1.0054) model_time 0.9239 (1.0045) loss 0.7988 (0.7924) grad_norm 8.8505 (8.7259/2.2167) mem 68106MB [2022-12-20 22:24:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1120/1519] eta 0:06:41 lr 0.000000 time 0.9213 (1.0054) model_time 0.9212 (1.0045) loss 0.7241 (0.7922) grad_norm 8.1863 (8.7213/2.2127) mem 68106MB [2022-12-20 22:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1130/1519] eta 0:06:31 lr 0.000000 time 0.9237 (1.0054) model_time 0.9235 (1.0045) loss 1.1218 (0.7922) grad_norm 8.8852 (8.6846/2.1907) mem 68106MB [2022-12-20 22:24:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1140/1519] eta 0:06:21 lr 0.000000 time 0.9280 (1.0054) model_time 0.9278 (1.0044) loss 0.6613 (0.7923) grad_norm 8.8002 (8.6790/2.1873) mem 68106MB [2022-12-20 22:24:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1150/1519] eta 0:06:10 lr 0.000000 time 0.9238 (1.0053) model_time 0.9237 (1.0044) loss 0.6649 (0.7924) grad_norm 13.7432 (8.6953/2.1999) mem 68106MB [2022-12-20 22:25:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1160/1519] eta 0:06:00 lr 0.000000 time 0.9389 (1.0053) model_time 0.9387 (1.0044) loss 0.8980 (0.7923) grad_norm 6.0037 (8.6888/2.1652) mem 68106MB [2022-12-20 22:25:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1170/1519] eta 0:05:50 lr 0.000000 time 0.9175 (1.0052) model_time 0.9173 (1.0043) loss 0.9635 (0.7928) grad_norm 8.1689 (8.6914/2.1486) mem 68106MB [2022-12-20 22:25:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1180/1519] eta 0:05:40 lr 0.000000 time 0.9221 (1.0052) model_time 0.9220 (1.0043) loss 0.7564 (0.7933) grad_norm 9.9274 (8.6807/2.1496) mem 68106MB [2022-12-20 22:25:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1190/1519] eta 0:05:30 lr 0.000000 time 1.0118 (1.0052) model_time 1.0116 (1.0043) loss 0.7197 (0.7931) grad_norm 8.1185 (8.6482/2.1633) mem 68106MB [2022-12-20 22:25:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1200/1519] eta 0:05:20 lr 0.000000 time 0.9254 (1.0052) model_time 0.9253 (1.0043) loss 0.6771 (0.7935) grad_norm 7.6643 (8.6379/2.1334) mem 68106MB [2022-12-20 22:25:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1210/1519] eta 0:05:10 lr 0.000000 time 0.9508 (1.0052) model_time 0.9507 (1.0043) loss 0.6699 (0.7933) grad_norm 10.4347 (8.6706/2.1535) mem 68106MB [2022-12-20 22:26:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1220/1519] eta 0:05:00 lr 0.000000 time 0.9199 (1.0053) model_time 0.9198 (1.0044) loss 0.8965 (0.7929) grad_norm 7.0947 (8.6592/2.1303) mem 68106MB [2022-12-20 22:26:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1230/1519] eta 0:04:50 lr 0.000000 time 0.9297 (1.0053) model_time 0.9295 (1.0044) loss 0.7814 (0.7933) grad_norm 10.3998 (8.6537/2.1385) mem 68106MB [2022-12-20 22:26:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1240/1519] eta 0:04:40 lr 0.000000 time 0.9252 (1.0052) model_time 0.9251 (1.0044) loss 0.6998 (0.7931) grad_norm 8.6825 (8.6819/2.1875) mem 68106MB [2022-12-20 22:26:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1250/1519] eta 0:04:30 lr 0.000000 time 0.9266 (1.0052) model_time 0.9265 (1.0043) loss 0.6589 (0.7932) grad_norm 6.4337 (8.6825/2.1997) mem 68106MB [2022-12-20 22:26:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1260/1519] eta 0:04:20 lr 0.000000 time 0.9208 (1.0053) model_time 0.9207 (1.0045) loss 0.6960 (0.7931) grad_norm 7.2204 (8.6995/2.1981) mem 68106MB [2022-12-20 22:26:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1270/1519] eta 0:04:10 lr 0.000000 time 0.9606 (1.0054) model_time 0.9605 (1.0045) loss 0.6777 (0.7929) grad_norm 8.0450 (8.7230/2.2015) mem 68106MB [2022-12-20 22:27:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1280/1519] eta 0:04:00 lr 0.000000 time 0.9246 (1.0054) model_time 0.9245 (1.0046) loss 0.6990 (0.7931) grad_norm 7.6931 (8.7185/2.1967) mem 68106MB [2022-12-20 22:27:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1290/1519] eta 0:03:50 lr 0.000000 time 0.9864 (1.0055) model_time 0.9862 (1.0047) loss 0.7451 (0.7931) grad_norm 6.2333 (8.7304/2.1890) mem 68106MB [2022-12-20 22:27:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1300/1519] eta 0:03:40 lr 0.000000 time 0.9261 (1.0055) model_time 0.9259 (1.0047) loss 0.9514 (0.7929) grad_norm 11.6117 (8.7424/2.1948) mem 68106MB [2022-12-20 22:27:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1310/1519] eta 0:03:30 lr 0.000000 time 0.9239 (1.0055) model_time 0.9238 (1.0047) loss 0.9433 (0.7927) grad_norm 5.8816 (8.6802/2.1478) mem 68106MB [2022-12-20 22:27:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1320/1519] eta 0:03:20 lr 0.000000 time 0.9218 (1.0054) model_time 0.9216 (1.0046) loss 0.7353 (0.7926) grad_norm 12.2086 (8.6604/2.1352) mem 68106MB [2022-12-20 22:27:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1330/1519] eta 0:03:10 lr 0.000000 time 0.9215 (1.0054) model_time 0.9213 (1.0046) loss 0.8999 (0.7927) grad_norm 10.3039 (8.6463/2.1205) mem 68106MB [2022-12-20 22:28:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1340/1519] eta 0:02:59 lr 0.000000 time 0.9280 (1.0054) model_time 0.9279 (1.0045) loss 0.6745 (0.7928) grad_norm 13.4848 (8.6723/2.1394) mem 68106MB [2022-12-20 22:28:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1350/1519] eta 0:02:49 lr 0.000000 time 0.9988 (1.0054) model_time 0.9987 (1.0045) loss 0.9114 (0.7927) grad_norm 10.8030 (8.6911/2.1427) mem 68106MB [2022-12-20 22:28:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1360/1519] eta 0:02:39 lr 0.000000 time 0.9203 (1.0055) model_time 0.9201 (1.0047) loss 0.6719 (0.7928) grad_norm 9.6732 (8.7057/2.1469) mem 68106MB [2022-12-20 22:28:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1370/1519] eta 0:02:29 lr 0.000000 time 0.9285 (1.0055) model_time 0.9284 (1.0047) loss 1.4833 (0.7930) grad_norm 11.0715 (8.7049/2.1433) mem 68106MB [2022-12-20 22:28:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1380/1519] eta 0:02:19 lr 0.000000 time 0.9282 (1.0055) model_time 0.9280 (1.0047) loss 0.7574 (0.7928) grad_norm 8.5387 (8.6539/2.1033) mem 68106MB [2022-12-20 22:28:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1390/1519] eta 0:02:09 lr 0.000000 time 0.9208 (1.0054) model_time 0.9207 (1.0046) loss 0.6731 (0.7925) grad_norm 8.2515 (8.6512/2.1012) mem 68106MB [2022-12-20 22:29:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1400/1519] eta 0:01:59 lr 0.000000 time 0.9323 (1.0054) model_time 0.9321 (1.0046) loss 0.8407 (0.7923) grad_norm 8.7661 (8.6490/2.1081) mem 68106MB [2022-12-20 22:29:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1410/1519] eta 0:01:49 lr 0.000000 time 0.9230 (1.0054) model_time 0.9229 (1.0045) loss 0.8731 (0.7918) grad_norm 6.1308 (8.6308/2.1147) mem 68106MB [2022-12-20 22:29:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1420/1519] eta 0:01:39 lr 0.000000 time 0.9249 (1.0053) model_time 0.9248 (1.0045) loss 1.0132 (0.7919) grad_norm 9.2677 (8.6658/2.1068) mem 68106MB [2022-12-20 22:29:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1430/1519] eta 0:01:29 lr 0.000000 time 0.9416 (1.0054) model_time 0.9415 (1.0045) loss 0.8000 (0.7919) grad_norm 8.1146 (8.6564/2.1023) mem 68106MB [2022-12-20 22:29:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1440/1519] eta 0:01:19 lr 0.000000 time 0.9390 (1.0053) model_time 0.9388 (1.0045) loss 0.7377 (0.7922) grad_norm 11.8187 (8.6548/2.1141) mem 68106MB [2022-12-20 22:29:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1450/1519] eta 0:01:09 lr 0.000000 time 0.9193 (1.0053) model_time 0.9192 (1.0045) loss 0.8266 (0.7930) grad_norm 7.5359 (8.6787/2.1580) mem 68106MB [2022-12-20 22:30:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1460/1519] eta 0:00:59 lr 0.000000 time 0.9123 (1.0052) model_time 0.9122 (1.0044) loss 1.0994 (0.7930) grad_norm 7.4129 (8.7084/2.1494) mem 68106MB [2022-12-20 22:30:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1470/1519] eta 0:00:49 lr 0.000000 time 0.9745 (1.0054) model_time 0.9744 (1.0046) loss 0.7373 (0.7935) grad_norm 12.4010 (8.7211/2.1581) mem 68106MB [2022-12-20 22:30:26 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1480/1519] eta 0:00:39 lr 0.000000 time 0.9186 (1.0054) model_time 0.9184 (1.0046) loss 0.6635 (0.7936) grad_norm 7.0917 (8.7096/2.1606) mem 68106MB [2022-12-20 22:30:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1490/1519] eta 0:00:29 lr 0.000000 time 0.9252 (1.0054) model_time 0.9250 (1.0046) loss 0.8530 (0.7941) grad_norm 10.1106 (8.7144/2.1597) mem 68106MB [2022-12-20 22:30:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1500/1519] eta 0:00:19 lr 0.000000 time 0.9278 (1.0054) model_time 0.9277 (1.0046) loss 0.6746 (0.7941) grad_norm 8.0215 (8.7067/2.1715) mem 68106MB [2022-12-20 22:30:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [95/100][1510/1519] eta 0:00:09 lr 0.000000 time 0.9245 (1.0054) model_time 0.9244 (1.0046) loss 0.7305 (0.7943) grad_norm 6.3641 (8.6923/2.1691) mem 68106MB [2022-12-20 22:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 95 training takes 0:25:27 [2022-12-20 22:31:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_95.pth saving...... [2022-12-20 22:31:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_95.pth saved !!! [2022-12-20 22:31:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.638 (0.638) Loss 0.5393 (0.5393) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 22:31:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.299 (0.328) Loss 0.5335 (0.5076) Acc@1 92.014 (92.740) Acc@5 98.611 (98.485) Mem 68106MB [2022-12-20 22:31:37 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.296 (0.314) Loss 0.4878 (0.5031) Acc@1 91.319 (92.741) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-20 22:31:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.303 (0.309) Loss 0.6338 (0.5103) Acc@1 90.972 (92.540) Acc@5 97.917 (98.477) Mem 68106MB [2022-12-20 22:31:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.297 (0.306) Loss 0.4610 (0.5011) Acc@1 93.750 (92.624) Acc@5 99.306 (98.577) Mem 68106MB [2022-12-20 22:31:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.298 (0.305) Loss 0.4921 (0.4985) Acc@1 92.708 (92.715) Acc@5 99.653 (98.632) Mem 68106MB [2022-12-20 22:31:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.304) Loss 0.5094 (0.4983) Acc@1 90.972 (92.646) Acc@5 98.264 (98.605) Mem 68106MB [2022-12-20 22:31:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.294 (0.303) Loss 0.5459 (0.4995) Acc@1 92.361 (92.601) Acc@5 97.917 (98.592) Mem 68106MB [2022-12-20 22:31:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.307 (0.303) Loss 0.4313 (0.4982) Acc@1 92.708 (92.623) Acc@5 98.611 (98.620) Mem 68106MB [2022-12-20 22:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:95] * Acc@1 92.588 Acc@5 98.621 [2022-12-20 22:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 22:31:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.59% [2022-12-20 22:31:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][0/1519] eta 0:47:22 lr 0.000000 time 1.8711 (1.8711) model_time 1.1315 (1.1315) loss 0.7187 (0.7187) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 22:32:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][10/1519] eta 0:27:22 lr 0.000000 time 0.9306 (1.0883) model_time 0.9305 (1.0207) loss 0.8011 (0.8074) grad_norm 12.1167 (10.1804/2.3268) mem 68106MB [2022-12-20 22:32:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][20/1519] eta 0:26:19 lr 0.000000 time 0.9228 (1.0540) model_time 0.9227 (1.0184) loss 0.8331 (0.7898) grad_norm 8.3817 (9.5830/1.9555) mem 68106MB [2022-12-20 22:32:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][30/1519] eta 0:25:42 lr 0.000000 time 0.9235 (1.0359) model_time 0.9233 (1.0117) loss 0.6620 (0.7831) grad_norm 7.3538 (9.7253/3.0192) mem 68106MB [2022-12-20 22:32:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][40/1519] eta 0:25:23 lr 0.000000 time 0.9241 (1.0300) model_time 0.9240 (1.0117) loss 0.7307 (0.7776) grad_norm 7.8555 (9.4394/3.0240) mem 68106MB [2022-12-20 22:32:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][50/1519] eta 0:25:06 lr 0.000000 time 0.9296 (1.0259) model_time 0.9295 (1.0110) loss 0.8951 (0.7841) grad_norm 7.5871 (9.1096/2.7974) mem 68106MB [2022-12-20 22:32:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][60/1519] eta 0:24:51 lr 0.000000 time 0.9812 (1.0225) model_time 0.9810 (1.0100) loss 0.9292 (0.7810) grad_norm 9.0583 (9.0662/2.5751) mem 68106MB [2022-12-20 22:33:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][70/1519] eta 0:24:38 lr 0.000000 time 0.9292 (1.0204) model_time 0.9291 (1.0096) loss 1.0634 (0.7857) grad_norm 9.0516 (8.8625/2.4732) mem 68106MB [2022-12-20 22:33:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][80/1519] eta 0:24:25 lr 0.000000 time 0.9863 (1.0185) model_time 0.9862 (1.0090) loss 0.8146 (0.7887) grad_norm 9.5044 (8.8654/2.3984) mem 68106MB [2022-12-20 22:33:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][90/1519] eta 0:24:12 lr 0.000000 time 0.9274 (1.0164) model_time 0.9273 (1.0080) loss 0.6760 (0.7870) grad_norm 8.2118 (8.7747/2.2766) mem 68106MB [2022-12-20 22:33:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][100/1519] eta 0:23:59 lr 0.000000 time 0.9344 (1.0147) model_time 0.9342 (1.0070) loss 0.7018 (0.7859) grad_norm 7.3084 (8.7649/2.4105) mem 68106MB [2022-12-20 22:33:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][110/1519] eta 0:23:47 lr 0.000000 time 0.9357 (1.0132) model_time 0.9356 (1.0062) loss 0.6673 (0.7885) grad_norm 7.7107 (8.6789/2.3562) mem 68106MB [2022-12-20 22:33:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][120/1519] eta 0:23:35 lr 0.000000 time 0.9411 (1.0121) model_time 0.9409 (1.0056) loss 1.1058 (0.7931) grad_norm 12.0138 (8.7772/2.3339) mem 68106MB [2022-12-20 22:34:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][130/1519] eta 0:23:26 lr 0.000000 time 0.9074 (1.0127) model_time 0.9073 (1.0067) loss 0.8606 (0.7901) grad_norm 8.1718 (8.7682/2.2483) mem 68106MB [2022-12-20 22:34:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][140/1519] eta 0:23:15 lr 0.000000 time 0.9173 (1.0118) model_time 0.9171 (1.0062) loss 0.6858 (0.7862) grad_norm 6.8933 (8.7491/2.1943) mem 68106MB [2022-12-20 22:34:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][150/1519] eta 0:23:04 lr 0.000000 time 0.9189 (1.0113) model_time 0.9187 (1.0060) loss 0.7334 (0.7860) grad_norm 10.5530 (8.7170/2.1508) mem 68106MB [2022-12-20 22:34:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][160/1519] eta 0:22:53 lr 0.000000 time 0.9193 (1.0106) model_time 0.9191 (1.0056) loss 0.6664 (0.7825) grad_norm 8.5434 (8.7376/2.1345) mem 68106MB [2022-12-20 22:34:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][170/1519] eta 0:22:43 lr 0.000000 time 0.9279 (1.0104) model_time 0.9277 (1.0057) loss 0.8810 (0.7867) grad_norm 8.2984 (8.6785/2.0969) mem 68106MB [2022-12-20 22:34:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][180/1519] eta 0:22:32 lr 0.000000 time 0.9254 (1.0104) model_time 0.9251 (1.0059) loss 0.7148 (0.7858) grad_norm 8.9787 (8.7087/2.0563) mem 68106MB [2022-12-20 22:35:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][190/1519] eta 0:22:22 lr 0.000000 time 0.9296 (1.0099) model_time 0.9294 (1.0057) loss 0.8620 (0.7883) grad_norm 12.5153 (8.7442/2.0726) mem 68106MB [2022-12-20 22:35:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][200/1519] eta 0:22:11 lr 0.000000 time 0.9240 (1.0096) model_time 0.9239 (1.0056) loss 0.7502 (0.7936) grad_norm 7.9815 (8.7042/2.0442) mem 68106MB [2022-12-20 22:35:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][210/1519] eta 0:22:01 lr 0.000000 time 0.9352 (1.0092) model_time 0.9350 (1.0054) loss 0.9236 (0.7930) grad_norm 7.8607 (8.6795/2.0059) mem 68106MB [2022-12-20 22:35:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][220/1519] eta 0:21:50 lr 0.000000 time 0.9259 (1.0088) model_time 0.9258 (1.0051) loss 0.8882 (0.7956) grad_norm 11.4754 (8.7203/1.9930) mem 68106MB [2022-12-20 22:35:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][230/1519] eta 0:21:39 lr 0.000000 time 0.9337 (1.0084) model_time 0.9335 (1.0049) loss 0.7249 (0.7947) grad_norm 9.0941 (8.7328/2.0238) mem 68106MB [2022-12-20 22:35:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][240/1519] eta 0:21:29 lr 0.000000 time 0.9357 (1.0081) model_time 0.9356 (1.0046) loss 0.8670 (0.7975) grad_norm 8.7299 (8.7483/2.0027) mem 68106MB [2022-12-20 22:36:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][250/1519] eta 0:21:18 lr 0.000000 time 0.9292 (1.0078) model_time 0.9291 (1.0045) loss 0.7432 (0.7974) grad_norm 8.0911 (8.7663/1.9777) mem 68106MB [2022-12-20 22:36:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][260/1519] eta 0:21:09 lr 0.000000 time 0.9203 (1.0080) model_time 0.9201 (1.0048) loss 0.8492 (0.7973) grad_norm 8.7453 (8.7515/1.9503) mem 68106MB [2022-12-20 22:36:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][270/1519] eta 0:20:58 lr 0.000000 time 0.9249 (1.0079) model_time 0.9247 (1.0048) loss 0.9433 (0.7968) grad_norm 12.7079 (8.7546/1.9732) mem 68106MB [2022-12-20 22:36:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][280/1519] eta 0:20:48 lr 0.000000 time 0.9084 (1.0081) model_time 0.9083 (1.0051) loss 0.6731 (0.7958) grad_norm 9.0670 (8.7460/1.9474) mem 68106MB [2022-12-20 22:36:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][290/1519] eta 0:20:38 lr 0.000000 time 0.9262 (1.0079) model_time 0.9261 (1.0050) loss 0.9875 (0.7952) grad_norm 5.3744 (8.7262/1.9468) mem 68106MB [2022-12-20 22:36:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][300/1519] eta 0:20:28 lr 0.000000 time 0.9242 (1.0078) model_time 0.9241 (1.0050) loss 0.9076 (0.7935) grad_norm 9.1131 (8.7043/1.9443) mem 68106MB [2022-12-20 22:37:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][310/1519] eta 0:20:18 lr 0.000000 time 0.9272 (1.0078) model_time 0.9270 (1.0051) loss 0.7481 (0.7930) grad_norm 9.9865 (8.6725/1.9344) mem 68106MB [2022-12-20 22:37:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][320/1519] eta 0:20:08 lr 0.000000 time 0.9323 (1.0078) model_time 0.9321 (1.0052) loss 0.7472 (0.7964) grad_norm 8.1559 (8.6803/1.9399) mem 68106MB [2022-12-20 22:37:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][330/1519] eta 0:19:58 lr 0.000000 time 0.9629 (1.0080) model_time 0.9628 (1.0054) loss 0.7129 (0.7972) grad_norm 9.2439 (8.6520/1.9259) mem 68106MB [2022-12-20 22:37:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][340/1519] eta 0:19:48 lr 0.000000 time 0.9223 (1.0078) model_time 0.9222 (1.0053) loss 0.9006 (0.7973) grad_norm 9.1613 (8.6674/1.9104) mem 68106MB [2022-12-20 22:37:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][350/1519] eta 0:19:37 lr 0.000000 time 0.9250 (1.0077) model_time 0.9248 (1.0053) loss 0.8383 (0.7984) grad_norm 9.0062 (8.6624/1.8867) mem 68106MB [2022-12-20 22:38:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][360/1519] eta 0:19:28 lr 0.000000 time 0.9260 (1.0080) model_time 0.9259 (1.0056) loss 0.8612 (0.7979) grad_norm 6.1301 (8.6485/1.8710) mem 68106MB [2022-12-20 22:38:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][370/1519] eta 0:19:17 lr 0.000000 time 0.9235 (1.0078) model_time 0.9234 (1.0055) loss 0.7207 (0.7968) grad_norm 7.3610 (8.6471/1.8637) mem 68106MB [2022-12-20 22:38:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][380/1519] eta 0:19:08 lr 0.000000 time 0.9270 (1.0080) model_time 0.9269 (1.0058) loss 0.6648 (0.7967) grad_norm 17.6001 (8.7026/1.9579) mem 68106MB [2022-12-20 22:38:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][390/1519] eta 0:18:57 lr 0.000000 time 0.9257 (1.0079) model_time 0.9255 (1.0056) loss 0.9033 (0.7994) grad_norm 10.6723 (8.7187/1.9720) mem 68106MB [2022-12-20 22:38:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][400/1519] eta 0:18:48 lr 0.000000 time 0.9201 (1.0084) model_time 0.9200 (1.0062) loss 1.0057 (0.7989) grad_norm 5.9541 (8.7616/2.0182) mem 68106MB [2022-12-20 22:38:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][410/1519] eta 0:18:37 lr 0.000000 time 0.9210 (1.0081) model_time 0.9209 (1.0060) loss 0.7745 (0.7985) grad_norm 6.6084 (8.7243/2.0091) mem 68106MB [2022-12-20 22:39:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][420/1519] eta 0:18:27 lr 0.000000 time 0.9390 (1.0080) model_time 0.9389 (1.0059) loss 0.8505 (0.7980) grad_norm 10.0640 (8.7420/1.9903) mem 68106MB [2022-12-20 22:39:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][430/1519] eta 0:18:17 lr 0.000000 time 0.9252 (1.0078) model_time 0.9250 (1.0057) loss 1.1836 (0.7991) grad_norm 7.8820 (8.7258/1.9846) mem 68106MB [2022-12-20 22:39:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][440/1519] eta 0:18:07 lr 0.000000 time 0.9193 (1.0079) model_time 0.9192 (1.0059) loss 0.8071 (0.7987) grad_norm 8.4309 (8.7252/1.9745) mem 68106MB [2022-12-20 22:39:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][450/1519] eta 0:17:57 lr 0.000000 time 0.9189 (1.0077) model_time 0.9188 (1.0058) loss 0.8764 (0.7987) grad_norm 7.1006 (8.7127/1.9658) mem 68106MB [2022-12-20 22:39:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][460/1519] eta 0:17:46 lr 0.000000 time 0.9311 (1.0076) model_time 0.9310 (1.0056) loss 0.6922 (0.7982) grad_norm 8.4156 (8.7052/1.9508) mem 68106MB [2022-12-20 22:39:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][470/1519] eta 0:17:36 lr 0.000000 time 0.9366 (1.0074) model_time 0.9364 (1.0055) loss 0.7255 (0.7992) grad_norm 6.3964 (8.6848/1.9495) mem 68106MB [2022-12-20 22:40:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][480/1519] eta 0:17:26 lr 0.000000 time 0.9267 (1.0075) model_time 0.9265 (1.0056) loss 0.6618 (0.8005) grad_norm 7.6267 (8.6602/1.9462) mem 68106MB [2022-12-20 22:40:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][490/1519] eta 0:17:16 lr 0.000000 time 0.9326 (1.0075) model_time 0.9324 (1.0057) loss 0.6796 (0.7989) grad_norm 7.1013 (8.6345/1.9399) mem 68106MB [2022-12-20 22:40:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][500/1519] eta 0:17:06 lr 0.000000 time 0.9317 (1.0075) model_time 0.9315 (1.0057) loss 0.6672 (0.7988) grad_norm 17.2168 (8.6630/2.0012) mem 68106MB [2022-12-20 22:40:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][510/1519] eta 0:16:56 lr 0.000000 time 0.9397 (1.0074) model_time 0.9395 (1.0057) loss 0.6757 (0.7983) grad_norm 9.0014 (8.6534/1.9872) mem 68106MB [2022-12-20 22:40:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][520/1519] eta 0:16:46 lr 0.000000 time 0.9233 (1.0073) model_time 0.9232 (1.0055) loss 0.6892 (0.7973) grad_norm 7.0996 (8.6366/1.9736) mem 68106MB [2022-12-20 22:40:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][530/1519] eta 0:16:36 lr 0.000000 time 0.9267 (1.0071) model_time 0.9266 (1.0054) loss 0.6637 (0.7965) grad_norm 7.2148 (8.6499/1.9775) mem 68106MB [2022-12-20 22:41:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][540/1519] eta 0:16:25 lr 0.000000 time 0.9304 (1.0070) model_time 0.9301 (1.0053) loss 1.2026 (0.7969) grad_norm 7.6099 (8.6362/1.9647) mem 68106MB [2022-12-20 22:41:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][550/1519] eta 0:16:15 lr 0.000000 time 0.9449 (1.0069) model_time 0.9447 (1.0052) loss 1.1250 (0.7982) grad_norm 7.3769 (8.6317/1.9553) mem 68106MB [2022-12-20 22:41:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][560/1519] eta 0:16:05 lr 0.000000 time 0.9212 (1.0069) model_time 0.9211 (1.0052) loss 0.6515 (0.7984) grad_norm 12.0711 (8.6305/1.9554) mem 68106MB [2022-12-20 22:41:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][570/1519] eta 0:15:55 lr 0.000000 time 0.9203 (1.0072) model_time 0.9201 (1.0056) loss 0.7717 (0.8001) grad_norm 14.4976 (8.6446/1.9780) mem 68106MB [2022-12-20 22:41:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][580/1519] eta 0:15:45 lr 0.000000 time 0.9236 (1.0074) model_time 0.9234 (1.0058) loss 0.7818 (0.7987) grad_norm 6.3028 (8.6350/1.9785) mem 68106MB [2022-12-20 22:41:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][590/1519] eta 0:15:35 lr 0.000000 time 0.9270 (1.0073) model_time 0.9269 (1.0057) loss 0.8908 (0.8001) grad_norm 9.2543 (8.6327/1.9656) mem 68106MB [2022-12-20 22:42:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][600/1519] eta 0:15:25 lr 0.000000 time 0.9245 (1.0073) model_time 0.9244 (1.0057) loss 0.9943 (0.8005) grad_norm 9.1021 (8.6578/1.9861) mem 68106MB [2022-12-20 22:42:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][610/1519] eta 0:15:15 lr 0.000000 time 0.9214 (1.0072) model_time 0.9213 (1.0057) loss 0.8122 (0.7996) grad_norm 8.9712 (8.6165/1.9649) mem 68106MB [2022-12-20 22:42:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][620/1519] eta 0:15:05 lr 0.000000 time 0.9273 (1.0072) model_time 0.9271 (1.0056) loss 0.7630 (0.7994) grad_norm 6.4201 (8.5991/1.9632) mem 68106MB [2022-12-20 22:42:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][630/1519] eta 0:14:55 lr 0.000000 time 0.9308 (1.0071) model_time 0.9306 (1.0056) loss 0.9206 (0.7987) grad_norm 8.1809 (8.5811/1.8993) mem 68106MB [2022-12-20 22:42:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][640/1519] eta 0:14:45 lr 0.000000 time 0.9727 (1.0070) model_time 0.9726 (1.0056) loss 0.7443 (0.7994) grad_norm 11.4306 (8.5917/1.8759) mem 68106MB [2022-12-20 22:42:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][650/1519] eta 0:14:35 lr 0.000000 time 0.9396 (1.0070) model_time 0.9395 (1.0056) loss 0.6700 (0.7985) grad_norm 12.7843 (8.6281/1.9022) mem 68106MB [2022-12-20 22:43:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][660/1519] eta 0:14:25 lr 0.000000 time 0.9973 (1.0070) model_time 0.9972 (1.0056) loss 0.8674 (0.7995) grad_norm 10.8752 (8.6311/1.9111) mem 68106MB [2022-12-20 22:43:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][670/1519] eta 0:14:15 lr 0.000000 time 0.9209 (1.0071) model_time 0.9208 (1.0057) loss 0.8940 (0.7995) grad_norm 9.6249 (8.6672/1.9141) mem 68106MB [2022-12-20 22:43:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][680/1519] eta 0:14:04 lr 0.000000 time 0.9254 (1.0070) model_time 0.9253 (1.0056) loss 0.7417 (0.7994) grad_norm 7.4133 (8.6589/1.9116) mem 68106MB [2022-12-20 22:43:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][690/1519] eta 0:13:54 lr 0.000000 time 0.9220 (1.0070) model_time 0.9219 (1.0056) loss 0.7075 (0.7982) grad_norm 6.9427 (8.6497/1.9172) mem 68106MB [2022-12-20 22:43:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][700/1519] eta 0:13:44 lr 0.000000 time 0.9199 (1.0068) model_time 0.9198 (1.0055) loss 0.6648 (0.7983) grad_norm 10.6835 (8.6549/1.8784) mem 68106MB [2022-12-20 22:43:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][710/1519] eta 0:13:34 lr 0.000000 time 0.9190 (1.0067) model_time 0.9189 (1.0054) loss 0.7082 (0.7979) grad_norm 8.5557 (8.7026/1.9458) mem 68106MB [2022-12-20 22:44:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][720/1519] eta 0:13:24 lr 0.000000 time 0.9213 (1.0066) model_time 0.9210 (1.0053) loss 0.8678 (0.7982) grad_norm 9.8365 (8.6797/1.9293) mem 68106MB [2022-12-20 22:44:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][730/1519] eta 0:13:14 lr 0.000000 time 0.9242 (1.0066) model_time 0.9241 (1.0053) loss 0.7412 (0.7974) grad_norm 7.6965 (8.6669/1.9337) mem 68106MB [2022-12-20 22:44:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][740/1519] eta 0:13:04 lr 0.000000 time 0.9285 (1.0065) model_time 0.9284 (1.0052) loss 0.6965 (0.7976) grad_norm 8.8940 (8.6750/1.9352) mem 68106MB [2022-12-20 22:44:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][750/1519] eta 0:12:54 lr 0.000000 time 0.9298 (1.0066) model_time 0.9297 (1.0053) loss 0.9259 (0.7978) grad_norm 10.6566 (8.6774/1.9354) mem 68106MB [2022-12-20 22:44:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][760/1519] eta 0:12:44 lr 0.000000 time 0.9238 (1.0069) model_time 0.9237 (1.0056) loss 0.9527 (0.7980) grad_norm 7.9136 (8.6622/1.9245) mem 68106MB [2022-12-20 22:44:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][770/1519] eta 0:12:34 lr 0.000000 time 0.9267 (1.0068) model_time 0.9265 (1.0055) loss 0.6921 (0.7980) grad_norm 9.1806 (8.7055/1.9566) mem 68106MB [2022-12-20 22:45:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][780/1519] eta 0:12:23 lr 0.000000 time 0.9219 (1.0067) model_time 0.9217 (1.0054) loss 0.8021 (0.7986) grad_norm 8.2873 (8.6833/1.9550) mem 68106MB [2022-12-20 22:45:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][790/1519] eta 0:12:13 lr 0.000000 time 0.8978 (1.0068) model_time 0.8976 (1.0055) loss 0.7366 (0.7991) grad_norm 8.6241 (8.6603/1.9356) mem 68106MB [2022-12-20 22:45:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][800/1519] eta 0:12:03 lr 0.000000 time 0.9276 (1.0068) model_time 0.9275 (1.0055) loss 0.7534 (0.7988) grad_norm 9.2346 (8.6604/1.9411) mem 68106MB [2022-12-20 22:45:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][810/1519] eta 0:11:53 lr 0.000000 time 0.9221 (1.0067) model_time 0.9220 (1.0054) loss 0.6831 (0.7985) grad_norm 6.5801 (8.6525/1.9463) mem 68106MB [2022-12-20 22:45:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][820/1519] eta 0:11:43 lr 0.000000 time 0.9742 (1.0067) model_time 0.9740 (1.0055) loss 0.6598 (0.7981) grad_norm 5.9912 (8.6301/1.9401) mem 68106MB [2022-12-20 22:45:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][830/1519] eta 0:11:33 lr 0.000000 time 0.9384 (1.0067) model_time 0.9382 (1.0055) loss 0.7066 (0.7972) grad_norm 11.3565 (8.6397/1.9407) mem 68106MB [2022-12-20 22:46:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][840/1519] eta 0:11:23 lr 0.000000 time 0.9308 (1.0067) model_time 0.9307 (1.0055) loss 0.8020 (0.7975) grad_norm 6.8828 (8.6150/1.9473) mem 68106MB [2022-12-20 22:46:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][850/1519] eta 0:11:13 lr 0.000000 time 0.9148 (1.0067) model_time 0.9146 (1.0055) loss 0.9651 (0.7981) grad_norm 8.0397 (8.6182/1.9534) mem 68106MB [2022-12-20 22:46:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][860/1519] eta 0:11:03 lr 0.000000 time 0.9299 (1.0067) model_time 0.9297 (1.0055) loss 0.7750 (0.7986) grad_norm 6.9820 (8.6107/1.9529) mem 68106MB [2022-12-20 22:46:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][870/1519] eta 0:10:53 lr 0.000000 time 0.9304 (1.0067) model_time 0.9303 (1.0055) loss 0.8657 (0.7984) grad_norm 9.0014 (8.6138/1.9305) mem 68106MB [2022-12-20 22:46:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][880/1519] eta 0:10:43 lr 0.000000 time 0.9274 (1.0067) model_time 0.9272 (1.0056) loss 0.9019 (0.7982) grad_norm 7.3686 (8.6169/1.9314) mem 68106MB [2022-12-20 22:46:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][890/1519] eta 0:10:33 lr 0.000000 time 0.9221 (1.0066) model_time 0.9220 (1.0055) loss 0.6895 (0.7979) grad_norm 10.0979 (8.6269/1.9221) mem 68106MB [2022-12-20 22:47:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][900/1519] eta 0:10:23 lr 0.000000 time 0.9188 (1.0065) model_time 0.9186 (1.0054) loss 0.6703 (0.7974) grad_norm 10.5987 (8.6335/1.9238) mem 68106MB [2022-12-20 22:47:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][910/1519] eta 0:10:12 lr 0.000000 time 0.9277 (1.0065) model_time 0.9276 (1.0054) loss 0.8812 (0.7973) grad_norm 15.0265 (8.6609/1.9542) mem 68106MB [2022-12-20 22:47:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][920/1519] eta 0:10:02 lr 0.000000 time 0.9275 (1.0064) model_time 0.9274 (1.0053) loss 0.7728 (0.7976) grad_norm 11.6084 (8.6705/1.9521) mem 68106MB [2022-12-20 22:47:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][930/1519] eta 0:09:52 lr 0.000000 time 0.9288 (1.0064) model_time 0.9286 (1.0053) loss 0.6874 (0.7973) grad_norm 6.9081 (8.7085/1.9909) mem 68106MB [2022-12-20 22:47:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][940/1519] eta 0:09:42 lr 0.000000 time 0.9730 (1.0064) model_time 0.9728 (1.0053) loss 0.7354 (0.7975) grad_norm 7.0246 (8.7136/2.0133) mem 68106MB [2022-12-20 22:47:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][950/1519] eta 0:09:32 lr 0.000000 time 0.9677 (1.0064) model_time 0.9676 (1.0053) loss 0.6771 (0.7974) grad_norm 6.3565 (8.6961/2.0202) mem 68106MB [2022-12-20 22:48:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][960/1519] eta 0:09:22 lr 0.000000 time 0.9260 (1.0064) model_time 0.9254 (1.0053) loss 0.9969 (0.7975) grad_norm 7.0316 (8.7139/2.0291) mem 68106MB [2022-12-20 22:48:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][970/1519] eta 0:09:12 lr 0.000000 time 0.9295 (1.0063) model_time 0.9293 (1.0052) loss 0.9231 (0.7979) grad_norm 8.4244 (8.7285/2.0610) mem 68106MB [2022-12-20 22:48:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][980/1519] eta 0:09:02 lr 0.000000 time 0.9218 (1.0063) model_time 0.9216 (1.0053) loss 0.7785 (0.7972) grad_norm 6.5011 (8.6717/2.0021) mem 68106MB [2022-12-20 22:48:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][990/1519] eta 0:08:52 lr 0.000000 time 0.9327 (1.0063) model_time 0.9325 (1.0052) loss 0.8779 (0.7972) grad_norm 6.6214 (8.6528/1.9891) mem 68106MB [2022-12-20 22:48:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1000/1519] eta 0:08:42 lr 0.000000 time 0.9406 (1.0063) model_time 0.9404 (1.0052) loss 0.8377 (0.7973) grad_norm 11.4190 (8.6141/1.9537) mem 68106MB [2022-12-20 22:48:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1010/1519] eta 0:08:32 lr 0.000000 time 0.9200 (1.0062) model_time 0.9199 (1.0052) loss 0.6993 (0.7972) grad_norm 6.9986 (8.6326/1.9492) mem 68106MB [2022-12-20 22:49:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1020/1519] eta 0:08:22 lr 0.000000 time 0.9224 (1.0062) model_time 0.9223 (1.0052) loss 0.7465 (0.7973) grad_norm 11.7518 (8.6274/1.9553) mem 68106MB [2022-12-20 22:49:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1030/1519] eta 0:08:12 lr 0.000000 time 0.9204 (1.0062) model_time 0.9202 (1.0051) loss 0.6888 (0.7985) grad_norm 6.8190 (8.6516/1.9592) mem 68106MB [2022-12-20 22:49:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1040/1519] eta 0:08:01 lr 0.000000 time 0.9292 (1.0061) model_time 0.9290 (1.0051) loss 0.9062 (0.7988) grad_norm 8.0968 (8.6360/1.9539) mem 68106MB [2022-12-20 22:49:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1050/1519] eta 0:07:51 lr 0.000000 time 0.9323 (1.0061) model_time 0.9322 (1.0051) loss 0.8491 (0.7991) grad_norm 9.0337 (8.6363/1.9496) mem 68106MB [2022-12-20 22:49:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1060/1519] eta 0:07:41 lr 0.000000 time 0.9331 (1.0060) model_time 0.9329 (1.0050) loss 0.6669 (0.8001) grad_norm 12.9213 (8.6349/1.9692) mem 68106MB [2022-12-20 22:49:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1070/1519] eta 0:07:31 lr 0.000000 time 0.9466 (1.0063) model_time 0.9465 (1.0053) loss 0.8430 (0.8001) grad_norm 8.1390 (8.6651/2.0111) mem 68106MB [2022-12-20 22:50:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1080/1519] eta 0:07:21 lr 0.000000 time 0.9248 (1.0062) model_time 0.9247 (1.0052) loss 0.8448 (0.8000) grad_norm 12.5802 (8.6753/2.0218) mem 68106MB [2022-12-20 22:50:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1090/1519] eta 0:07:11 lr 0.000000 time 0.9216 (1.0062) model_time 0.9214 (1.0052) loss 0.8024 (0.8001) grad_norm 8.9072 (8.6954/2.0166) mem 68106MB [2022-12-20 22:50:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1100/1519] eta 0:07:01 lr 0.000000 time 0.9890 (1.0062) model_time 0.9889 (1.0052) loss 0.6813 (0.8002) grad_norm 5.7583 (8.6844/1.9791) mem 68106MB [2022-12-20 22:50:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1110/1519] eta 0:06:51 lr 0.000000 time 0.9649 (1.0062) model_time 0.9648 (1.0052) loss 0.7634 (0.8001) grad_norm 12.4378 (8.7284/2.0066) mem 68106MB [2022-12-20 22:50:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1120/1519] eta 0:06:41 lr 0.000000 time 0.9813 (1.0062) model_time 0.9812 (1.0052) loss 0.6691 (0.8003) grad_norm 10.3508 (8.7389/2.0127) mem 68106MB [2022-12-20 22:50:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1130/1519] eta 0:06:31 lr 0.000000 time 0.9835 (1.0062) model_time 0.9833 (1.0053) loss 0.7857 (0.8001) grad_norm 8.9975 (8.7330/2.0090) mem 68106MB [2022-12-20 22:51:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1140/1519] eta 0:06:21 lr 0.000000 time 0.9296 (1.0063) model_time 0.9294 (1.0053) loss 0.9103 (0.8004) grad_norm 7.0290 (8.7381/2.0146) mem 68106MB [2022-12-20 22:51:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1150/1519] eta 0:06:11 lr 0.000000 time 0.9250 (1.0062) model_time 0.9249 (1.0052) loss 0.6605 (0.8003) grad_norm 8.3413 (8.7281/2.0121) mem 68106MB [2022-12-20 22:51:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1160/1519] eta 0:06:01 lr 0.000000 time 0.8928 (1.0064) model_time 0.8926 (1.0054) loss 0.8140 (0.8004) grad_norm 5.7707 (8.7170/2.0067) mem 68106MB [2022-12-20 22:51:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1170/1519] eta 0:05:51 lr 0.000000 time 0.9262 (1.0064) model_time 0.9261 (1.0054) loss 0.8334 (0.8001) grad_norm 8.4545 (8.7298/1.9926) mem 68106MB [2022-12-20 22:51:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1180/1519] eta 0:05:41 lr 0.000000 time 0.9237 (1.0064) model_time 0.9235 (1.0055) loss 0.7273 (0.7995) grad_norm 10.1550 (8.7663/2.0204) mem 68106MB [2022-12-20 22:51:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1190/1519] eta 0:05:31 lr 0.000000 time 0.9253 (1.0064) model_time 0.9251 (1.0054) loss 0.8220 (0.8003) grad_norm 8.8682 (8.7754/2.0357) mem 68106MB [2022-12-20 22:52:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1200/1519] eta 0:05:20 lr 0.000000 time 0.9278 (1.0063) model_time 0.9276 (1.0053) loss 0.6737 (0.8011) grad_norm inf (8.7384/2.0092) mem 68106MB [2022-12-20 22:52:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1210/1519] eta 0:05:10 lr 0.000000 time 0.9346 (1.0062) model_time 0.9345 (1.0053) loss 1.0250 (0.8013) grad_norm 10.6400 (8.7580/2.0225) mem 68106MB [2022-12-20 22:52:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1220/1519] eta 0:05:00 lr 0.000000 time 0.9253 (1.0062) model_time 0.9252 (1.0052) loss 0.6773 (0.8013) grad_norm 12.0018 (8.7749/2.0360) mem 68106MB [2022-12-20 22:52:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1230/1519] eta 0:04:50 lr 0.000000 time 0.9219 (1.0061) model_time 0.9217 (1.0052) loss 0.6707 (0.8011) grad_norm 6.6079 (8.7726/2.0307) mem 68106MB [2022-12-20 22:52:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1240/1519] eta 0:04:40 lr 0.000000 time 0.9541 (1.0061) model_time 0.9540 (1.0052) loss 0.7664 (0.8009) grad_norm 9.1290 (8.7877/2.0332) mem 68106MB [2022-12-20 22:52:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1250/1519] eta 0:04:30 lr 0.000000 time 0.9340 (1.0062) model_time 0.9339 (1.0053) loss 0.9468 (0.8005) grad_norm 10.8586 (8.7774/2.0203) mem 68106MB [2022-12-20 22:53:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1260/1519] eta 0:04:20 lr 0.000000 time 0.9216 (1.0061) model_time 0.9215 (1.0052) loss 0.8364 (0.8002) grad_norm 9.5445 (8.7729/2.0125) mem 68106MB [2022-12-20 22:53:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1270/1519] eta 0:04:10 lr 0.000000 time 0.9198 (1.0061) model_time 0.9197 (1.0052) loss 0.6925 (0.8006) grad_norm 8.7653 (8.7437/2.0063) mem 68106MB [2022-12-20 22:53:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1280/1519] eta 0:04:00 lr 0.000000 time 0.9809 (1.0061) model_time 0.9807 (1.0052) loss 0.7585 (0.8005) grad_norm 6.2774 (8.7287/2.0100) mem 68106MB [2022-12-20 22:53:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1290/1519] eta 0:03:50 lr 0.000000 time 0.9174 (1.0061) model_time 0.9172 (1.0052) loss 0.7632 (0.8003) grad_norm 8.4720 (8.7552/2.0272) mem 68106MB [2022-12-20 22:53:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1300/1519] eta 0:03:40 lr 0.000000 time 0.9243 (1.0060) model_time 0.9242 (1.0051) loss 0.8039 (0.7998) grad_norm 10.0423 (8.7942/2.0322) mem 68106MB [2022-12-20 22:53:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1310/1519] eta 0:03:30 lr 0.000000 time 0.9247 (1.0060) model_time 0.9244 (1.0051) loss 0.7848 (0.7996) grad_norm 6.4262 (8.7345/1.9742) mem 68106MB [2022-12-20 22:54:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1320/1519] eta 0:03:20 lr 0.000000 time 0.9244 (1.0060) model_time 0.9242 (1.0051) loss 0.7365 (0.8000) grad_norm 6.7129 (8.7597/2.0385) mem 68106MB [2022-12-20 22:54:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1330/1519] eta 0:03:10 lr 0.000000 time 0.9161 (1.0059) model_time 0.9160 (1.0050) loss 0.7033 (0.7999) grad_norm 15.0909 (8.7731/2.0708) mem 68106MB [2022-12-20 22:54:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1340/1519] eta 0:03:00 lr 0.000000 time 0.9261 (1.0059) model_time 0.9260 (1.0050) loss 0.9078 (0.8006) grad_norm 7.6046 (8.7952/2.0837) mem 68106MB [2022-12-20 22:54:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1350/1519] eta 0:02:49 lr 0.000000 time 0.9213 (1.0058) model_time 0.9211 (1.0049) loss 0.8424 (0.8007) grad_norm 9.7533 (8.8268/2.1067) mem 68106MB [2022-12-20 22:54:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1360/1519] eta 0:02:39 lr 0.000000 time 0.9224 (1.0058) model_time 0.9222 (1.0049) loss 0.6734 (0.8009) grad_norm 7.3543 (8.8304/2.1366) mem 68106MB [2022-12-20 22:54:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1370/1519] eta 0:02:29 lr 0.000000 time 0.9244 (1.0057) model_time 0.9243 (1.0049) loss 0.7068 (0.8007) grad_norm 6.5464 (8.7724/2.1219) mem 68106MB [2022-12-20 22:55:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1380/1519] eta 0:02:19 lr 0.000000 time 0.9477 (1.0058) model_time 0.9475 (1.0049) loss 0.6882 (0.8011) grad_norm 7.2470 (8.7659/2.1269) mem 68106MB [2022-12-20 22:55:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1390/1519] eta 0:02:09 lr 0.000000 time 0.9497 (1.0059) model_time 0.9496 (1.0051) loss 0.7679 (0.8009) grad_norm 7.6251 (8.7887/2.1475) mem 68106MB [2022-12-20 22:55:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1400/1519] eta 0:01:59 lr 0.000000 time 0.9237 (1.0059) model_time 0.9236 (1.0051) loss 0.7653 (0.8010) grad_norm 6.8395 (8.7910/2.1469) mem 68106MB [2022-12-20 22:55:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1410/1519] eta 0:01:49 lr 0.000000 time 0.9341 (1.0059) model_time 0.9340 (1.0050) loss 0.8734 (0.8013) grad_norm 9.1919 (8.7927/2.1450) mem 68106MB [2022-12-20 22:55:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1420/1519] eta 0:01:39 lr 0.000000 time 0.9272 (1.0059) model_time 0.9270 (1.0051) loss 0.8847 (0.8018) grad_norm 11.8378 (8.8198/2.1570) mem 68106MB [2022-12-20 22:55:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1430/1519] eta 0:01:29 lr 0.000000 time 0.9217 (1.0059) model_time 0.9216 (1.0050) loss 0.7000 (0.8015) grad_norm 10.1302 (8.8405/2.1524) mem 68106MB [2022-12-20 22:56:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1440/1519] eta 0:01:19 lr 0.000000 time 0.9248 (1.0059) model_time 0.9246 (1.0051) loss 0.7873 (0.8017) grad_norm 8.1198 (8.8321/2.1419) mem 68106MB [2022-12-20 22:56:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1450/1519] eta 0:01:09 lr 0.000000 time 0.9366 (1.0059) model_time 0.9364 (1.0051) loss 0.7298 (0.8016) grad_norm 7.6727 (8.8119/2.1365) mem 68106MB [2022-12-20 22:56:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1460/1519] eta 0:00:59 lr 0.000000 time 0.9258 (1.0059) model_time 0.9257 (1.0051) loss 0.7433 (0.8019) grad_norm 7.5903 (8.8174/2.1374) mem 68106MB [2022-12-20 22:56:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1470/1519] eta 0:00:49 lr 0.000000 time 0.9096 (1.0060) model_time 0.9095 (1.0052) loss 0.9003 (0.8021) grad_norm 9.7744 (8.8223/2.1391) mem 68106MB [2022-12-20 22:56:46 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1480/1519] eta 0:00:39 lr 0.000000 time 0.9249 (1.0061) model_time 0.9247 (1.0052) loss 0.7851 (0.8023) grad_norm 7.0144 (8.8284/2.1564) mem 68106MB [2022-12-20 22:56:56 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1490/1519] eta 0:00:29 lr 0.000000 time 0.9231 (1.0060) model_time 0.9229 (1.0052) loss 0.6726 (0.8024) grad_norm 10.5803 (8.8408/2.1655) mem 68106MB [2022-12-20 22:57:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1500/1519] eta 0:00:19 lr 0.000000 time 0.9251 (1.0062) model_time 0.9249 (1.0054) loss 0.6623 (0.8026) grad_norm 7.3953 (8.8568/2.1570) mem 68106MB [2022-12-20 22:57:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [96/100][1510/1519] eta 0:00:09 lr 0.000000 time 0.9233 (1.0061) model_time 0.9232 (1.0053) loss 1.0063 (0.8026) grad_norm 8.8601 (8.8571/2.1564) mem 68106MB [2022-12-20 22:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 96 training takes 0:25:28 [2022-12-20 22:57:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_96.pth saving...... [2022-12-20 22:57:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_96.pth saved !!! [2022-12-20 22:57:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.668 (0.668) Loss 0.5394 (0.5394) Acc@1 93.056 (93.056) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 22:57:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.301 (0.332) Loss 0.5346 (0.5093) Acc@1 92.014 (92.771) Acc@5 98.611 (98.485) Mem 68106MB [2022-12-20 22:57:57 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.298 (0.316) Loss 0.4903 (0.5049) Acc@1 91.319 (92.758) Acc@5 99.306 (98.462) Mem 68106MB [2022-12-20 22:58:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.296 (0.310) Loss 0.6373 (0.5120) Acc@1 90.972 (92.563) Acc@5 98.264 (98.443) Mem 68106MB [2022-12-20 22:58:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.295 (0.307) Loss 0.4596 (0.5026) Acc@1 93.750 (92.641) Acc@5 99.306 (98.552) Mem 68106MB [2022-12-20 22:58:06 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.301 (0.306) Loss 0.4958 (0.5001) Acc@1 92.708 (92.729) Acc@5 99.653 (98.598) Mem 68106MB [2022-12-20 22:58:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.296 (0.305) Loss 0.5114 (0.4999) Acc@1 90.972 (92.651) Acc@5 98.264 (98.577) Mem 68106MB [2022-12-20 22:58:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.292 (0.304) Loss 0.5470 (0.5012) Acc@1 92.708 (92.606) Acc@5 97.917 (98.562) Mem 68106MB [2022-12-20 22:58:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.305 (0.303) Loss 0.4328 (0.4998) Acc@1 92.708 (92.631) Acc@5 98.264 (98.590) Mem 68106MB [2022-12-20 22:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:96] * Acc@1 92.592 Acc@5 98.592 [2022-12-20 22:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 22:58:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.59% [2022-12-20 22:58:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][0/1519] eta 0:46:02 lr 0.000000 time 1.8189 (1.8189) model_time 1.0762 (1.0762) loss 0.8220 (0.8220) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 22:58:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][10/1519] eta 0:27:00 lr 0.000000 time 0.9296 (1.0738) model_time 0.9295 (1.0060) loss 0.7541 (0.8050) grad_norm 6.5171 (8.3343/1.5285) mem 68106MB [2022-12-20 22:58:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][20/1519] eta 0:26:00 lr 0.000000 time 0.9843 (1.0413) model_time 0.9842 (1.0057) loss 0.8842 (0.7881) grad_norm 8.3160 (8.9992/1.4566) mem 68106MB [2022-12-20 22:58:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][30/1519] eta 0:25:31 lr 0.000000 time 0.9222 (1.0285) model_time 0.9220 (1.0043) loss 0.8085 (0.7983) grad_norm 11.4971 (9.7238/2.8379) mem 68106MB [2022-12-20 22:58:58 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][40/1519] eta 0:25:16 lr 0.000000 time 0.9749 (1.0253) model_time 0.9748 (1.0069) loss 0.7961 (0.8087) grad_norm 9.0075 (9.4560/2.6311) mem 68106MB [2022-12-20 22:59:08 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][50/1519] eta 0:24:58 lr 0.000000 time 0.9265 (1.0199) model_time 0.9264 (1.0051) loss 0.8246 (0.8096) grad_norm 10.1320 (10.0782/3.6648) mem 68106MB [2022-12-20 22:59:18 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][60/1519] eta 0:24:43 lr 0.000000 time 0.9252 (1.0165) model_time 0.9251 (1.0040) loss 0.6733 (0.8070) grad_norm 7.9624 (9.7082/3.4832) mem 68106MB [2022-12-20 22:59:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][70/1519] eta 0:24:28 lr 0.000000 time 0.9262 (1.0138) model_time 0.9260 (1.0031) loss 0.8805 (0.8047) grad_norm 7.2532 (9.6102/3.3292) mem 68106MB [2022-12-20 22:59:38 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][80/1519] eta 0:24:16 lr 0.000000 time 0.9238 (1.0122) model_time 0.9237 (1.0028) loss 0.7476 (0.8038) grad_norm 7.3872 (9.7428/3.2531) mem 68106MB [2022-12-20 22:59:48 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][90/1519] eta 0:24:04 lr 0.000000 time 0.9280 (1.0111) model_time 0.9278 (1.0026) loss 0.7865 (0.7991) grad_norm 8.9725 (9.5219/3.1386) mem 68106MB [2022-12-20 22:59:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][100/1519] eta 0:23:55 lr 0.000000 time 0.9349 (1.0118) model_time 0.9348 (1.0041) loss 0.7974 (0.8012) grad_norm 12.6375 (9.4295/3.0736) mem 68106MB [2022-12-20 23:00:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][110/1519] eta 0:23:45 lr 0.000000 time 0.9312 (1.0116) model_time 0.9311 (1.0046) loss 0.7610 (0.8027) grad_norm 10.7038 (9.5799/3.2585) mem 68106MB [2022-12-20 23:00:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][120/1519] eta 0:23:35 lr 0.000000 time 0.9249 (1.0116) model_time 0.9248 (1.0052) loss 0.9295 (0.8057) grad_norm 7.1789 (9.4740/3.2005) mem 68106MB [2022-12-20 23:00:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][130/1519] eta 0:23:26 lr 0.000000 time 1.1540 (1.0124) model_time 1.1539 (1.0064) loss 0.6767 (0.8021) grad_norm 11.1186 (9.4446/3.1057) mem 68106MB [2022-12-20 23:00:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][140/1519] eta 0:23:15 lr 0.000000 time 0.9202 (1.0118) model_time 0.9199 (1.0062) loss 0.8955 (0.8032) grad_norm 10.7955 (9.4087/3.0157) mem 68106MB [2022-12-20 23:00:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][150/1519] eta 0:23:04 lr 0.000000 time 0.9240 (1.0111) model_time 0.9238 (1.0059) loss 0.7065 (0.8036) grad_norm 11.3019 (9.3560/2.9462) mem 68106MB [2022-12-20 23:00:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][160/1519] eta 0:22:53 lr 0.000000 time 0.9256 (1.0108) model_time 0.9255 (1.0059) loss 1.2233 (0.8070) grad_norm 9.1261 (9.3253/2.8605) mem 68106MB [2022-12-20 23:01:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][170/1519] eta 0:22:43 lr 0.000000 time 0.9096 (1.0107) model_time 0.9095 (1.0060) loss 0.9312 (0.8052) grad_norm 11.2700 (9.4414/2.8431) mem 68106MB [2022-12-20 23:01:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][180/1519] eta 0:22:32 lr 0.000000 time 0.9264 (1.0099) model_time 0.9262 (1.0055) loss 0.7872 (0.8052) grad_norm 9.5089 (9.3235/2.8215) mem 68106MB [2022-12-20 23:01:29 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][190/1519] eta 0:22:22 lr 0.000000 time 1.0170 (1.0098) model_time 1.0169 (1.0056) loss 0.6640 (0.8039) grad_norm 6.3897 (9.3107/2.7804) mem 68106MB [2022-12-20 23:01:39 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][200/1519] eta 0:22:11 lr 0.000000 time 0.9298 (1.0092) model_time 0.9297 (1.0052) loss 0.7323 (0.8067) grad_norm 6.8699 (9.2035/2.7508) mem 68106MB [2022-12-20 23:01:49 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][210/1519] eta 0:22:00 lr 0.000000 time 0.9275 (1.0086) model_time 0.9274 (1.0048) loss 0.9339 (0.8071) grad_norm 7.5968 (9.1549/2.7045) mem 68106MB [2022-12-20 23:01:59 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][220/1519] eta 0:21:50 lr 0.000000 time 0.9760 (1.0085) model_time 0.9759 (1.0048) loss 1.1346 (0.8083) grad_norm 6.6955 (9.1781/2.7398) mem 68106MB [2022-12-20 23:02:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][230/1519] eta 0:21:39 lr 0.000000 time 0.9283 (1.0083) model_time 0.9282 (1.0047) loss 0.8479 (0.8082) grad_norm 6.5324 (9.2035/2.7077) mem 68106MB [2022-12-20 23:02:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][240/1519] eta 0:21:30 lr 0.000000 time 0.9167 (1.0090) model_time 0.9164 (1.0057) loss 0.6757 (0.8077) grad_norm 8.3011 (9.1746/2.6631) mem 68106MB [2022-12-20 23:02:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][250/1519] eta 0:21:19 lr 0.000000 time 0.9240 (1.0086) model_time 0.9239 (1.0053) loss 0.7270 (0.8057) grad_norm 9.6821 (9.1348/2.6224) mem 68106MB [2022-12-20 23:02:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][260/1519] eta 0:21:09 lr 0.000000 time 0.9256 (1.0084) model_time 0.9255 (1.0052) loss 0.9386 (0.8075) grad_norm 6.3106 (9.0705/2.5953) mem 68106MB [2022-12-20 23:02:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][270/1519] eta 0:20:59 lr 0.000000 time 0.9349 (1.0084) model_time 0.9346 (1.0053) loss 0.9496 (0.8103) grad_norm 6.8749 (9.0647/2.5605) mem 68106MB [2022-12-20 23:03:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][280/1519] eta 0:20:49 lr 0.000000 time 0.9318 (1.0082) model_time 0.9317 (1.0053) loss 0.8949 (0.8124) grad_norm 7.0305 (9.0183/2.5353) mem 68106MB [2022-12-20 23:03:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][290/1519] eta 0:20:39 lr 0.000000 time 0.9282 (1.0084) model_time 0.9280 (1.0055) loss 0.8373 (0.8116) grad_norm 7.7744 (8.9947/2.5033) mem 68106MB [2022-12-20 23:03:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][300/1519] eta 0:20:29 lr 0.000000 time 0.9257 (1.0082) model_time 0.9256 (1.0055) loss 0.6850 (0.8117) grad_norm 7.8258 (8.9583/2.4820) mem 68106MB [2022-12-20 23:03:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][310/1519] eta 0:20:18 lr 0.000000 time 0.9218 (1.0080) model_time 0.9216 (1.0053) loss 0.8716 (0.8102) grad_norm 7.3590 (8.9236/2.4554) mem 68106MB [2022-12-20 23:03:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][320/1519] eta 0:20:08 lr 0.000000 time 0.9225 (1.0077) model_time 0.9223 (1.0051) loss 0.7196 (0.8093) grad_norm 8.9694 (8.9630/2.4938) mem 68106MB [2022-12-20 23:03:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][330/1519] eta 0:19:57 lr 0.000000 time 0.9168 (1.0074) model_time 0.9167 (1.0048) loss 0.9437 (0.8111) grad_norm 8.9371 (8.9577/2.4889) mem 68106MB [2022-12-20 23:04:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][340/1519] eta 0:19:47 lr 0.000000 time 0.9331 (1.0074) model_time 0.9330 (1.0049) loss 0.9511 (0.8094) grad_norm 8.9465 (9.0108/2.5223) mem 68106MB [2022-12-20 23:04:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][350/1519] eta 0:19:37 lr 0.000000 time 1.0469 (1.0075) model_time 1.0468 (1.0051) loss 0.8036 (0.8081) grad_norm 11.0351 (8.9791/2.5075) mem 68106MB [2022-12-20 23:04:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][360/1519] eta 0:19:27 lr 0.000000 time 0.9274 (1.0074) model_time 0.9272 (1.0050) loss 0.8918 (0.8109) grad_norm 8.0987 (9.0081/2.5036) mem 68106MB [2022-12-20 23:04:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][370/1519] eta 0:19:17 lr 0.000000 time 0.9993 (1.0073) model_time 0.9991 (1.0050) loss 0.6986 (0.8094) grad_norm 7.9415 (9.0230/2.4769) mem 68106MB [2022-12-20 23:04:40 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][380/1519] eta 0:19:06 lr 0.000000 time 0.9199 (1.0070) model_time 0.9198 (1.0047) loss 0.6681 (0.8077) grad_norm 7.1046 (9.0628/2.5032) mem 68106MB [2022-12-20 23:04:50 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][390/1519] eta 0:18:56 lr 0.000000 time 0.9249 (1.0069) model_time 0.9248 (1.0047) loss 0.7738 (0.8086) grad_norm 7.5741 (9.0271/2.4992) mem 68106MB [2022-12-20 23:05:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][400/1519] eta 0:18:46 lr 0.000000 time 0.9190 (1.0067) model_time 0.9189 (1.0045) loss 1.2010 (0.8098) grad_norm 6.5057 (9.0185/2.4769) mem 68106MB [2022-12-20 23:05:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][410/1519] eta 0:18:36 lr 0.000000 time 0.9236 (1.0069) model_time 0.9235 (1.0048) loss 1.1564 (0.8100) grad_norm 10.1879 (9.0232/2.4561) mem 68106MB [2022-12-20 23:05:20 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][420/1519] eta 0:18:26 lr 0.000000 time 0.9317 (1.0072) model_time 0.9315 (1.0051) loss 0.6915 (0.8082) grad_norm 8.3239 (9.0425/2.4547) mem 68106MB [2022-12-20 23:05:30 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][430/1519] eta 0:18:16 lr 0.000000 time 0.9335 (1.0070) model_time 0.9333 (1.0050) loss 0.6702 (0.8077) grad_norm 21.0925 (9.0876/2.5781) mem 68106MB [2022-12-20 23:05:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][440/1519] eta 0:18:06 lr 0.000000 time 0.9322 (1.0069) model_time 0.9305 (1.0049) loss 0.6676 (0.8072) grad_norm 5.9794 (9.0697/2.5624) mem 68106MB [2022-12-20 23:05:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][450/1519] eta 0:17:56 lr 0.000000 time 0.9201 (1.0073) model_time 0.9200 (1.0053) loss 0.9099 (0.8061) grad_norm 7.8007 (9.0407/2.5474) mem 68106MB [2022-12-20 23:06:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][460/1519] eta 0:17:46 lr 0.000000 time 0.9327 (1.0073) model_time 0.9325 (1.0054) loss 0.6673 (0.8059) grad_norm 9.6284 (9.0389/2.5247) mem 68106MB [2022-12-20 23:06:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][470/1519] eta 0:17:37 lr 0.000000 time 0.9275 (1.0078) model_time 0.9273 (1.0060) loss 0.7414 (0.8054) grad_norm 6.4124 (9.0301/2.5567) mem 68106MB [2022-12-20 23:06:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][480/1519] eta 0:17:26 lr 0.000000 time 0.9208 (1.0077) model_time 0.9206 (1.0058) loss 0.8659 (0.8051) grad_norm 7.0521 (9.0012/2.5521) mem 68106MB [2022-12-20 23:06:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][490/1519] eta 0:17:17 lr 0.000000 time 0.9251 (1.0078) model_time 0.9249 (1.0060) loss 0.6931 (0.8046) grad_norm 8.7098 (8.9934/2.5323) mem 68106MB [2022-12-20 23:06:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][500/1519] eta 0:17:06 lr 0.000000 time 0.9235 (1.0076) model_time 0.9233 (1.0058) loss 0.9735 (0.8043) grad_norm 10.7765 (8.9773/2.5178) mem 68106MB [2022-12-20 23:06:51 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][510/1519] eta 0:16:56 lr 0.000000 time 0.9174 (1.0074) model_time 0.9172 (1.0057) loss 1.0544 (0.8047) grad_norm 6.4835 (8.9661/2.5008) mem 68106MB [2022-12-20 23:07:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][520/1519] eta 0:16:46 lr 0.000000 time 0.9299 (1.0073) model_time 0.9298 (1.0055) loss 0.6883 (0.8050) grad_norm 8.2527 (8.9694/2.4853) mem 68106MB [2022-12-20 23:07:11 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][530/1519] eta 0:16:36 lr 0.000000 time 1.0176 (1.0073) model_time 1.0174 (1.0056) loss 0.8809 (0.8051) grad_norm 7.8309 (8.9732/2.4716) mem 68106MB [2022-12-20 23:07:21 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][540/1519] eta 0:16:26 lr 0.000000 time 0.8917 (1.0072) model_time 0.8915 (1.0055) loss 0.6689 (0.8063) grad_norm 6.2399 (8.9796/2.4600) mem 68106MB [2022-12-20 23:07:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][550/1519] eta 0:16:15 lr 0.000000 time 0.9309 (1.0070) model_time 0.9308 (1.0054) loss 0.7940 (0.8063) grad_norm 9.1108 (8.9817/2.4399) mem 68106MB [2022-12-20 23:07:41 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][560/1519] eta 0:16:05 lr 0.000000 time 0.9263 (1.0071) model_time 0.9262 (1.0054) loss 0.9807 (0.8052) grad_norm 5.1882 (8.9637/2.4303) mem 68106MB [2022-12-20 23:07:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][570/1519] eta 0:15:55 lr 0.000000 time 0.9219 (1.0071) model_time 0.9217 (1.0055) loss 0.6919 (0.8047) grad_norm 6.9469 (8.9659/2.4560) mem 68106MB [2022-12-20 23:08:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][580/1519] eta 0:15:45 lr 0.000000 time 0.9362 (1.0070) model_time 0.9360 (1.0054) loss 0.7912 (0.8037) grad_norm 6.5640 (8.9904/2.5017) mem 68106MB [2022-12-20 23:08:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][590/1519] eta 0:15:35 lr 0.000000 time 0.9815 (1.0070) model_time 0.9813 (1.0054) loss 0.8405 (0.8040) grad_norm 6.0763 (8.9876/2.4941) mem 68106MB [2022-12-20 23:08:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][600/1519] eta 0:15:25 lr 0.000000 time 0.9175 (1.0069) model_time 0.9173 (1.0053) loss 0.7330 (0.8040) grad_norm 10.4786 (8.9998/2.4810) mem 68106MB [2022-12-20 23:08:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][610/1519] eta 0:15:15 lr 0.000000 time 0.8928 (1.0070) model_time 0.8926 (1.0054) loss 0.6764 (0.8032) grad_norm 16.1246 (9.0219/2.5150) mem 68106MB [2022-12-20 23:08:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][620/1519] eta 0:15:05 lr 0.000000 time 0.9399 (1.0069) model_time 0.9397 (1.0054) loss 0.7278 (0.8033) grad_norm 8.8085 (9.0108/2.5160) mem 68106MB [2022-12-20 23:08:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][630/1519] eta 0:14:55 lr 0.000000 time 0.9293 (1.0071) model_time 0.9291 (1.0056) loss 0.7215 (0.8028) grad_norm 9.2088 (8.9550/2.4553) mem 68106MB [2022-12-20 23:09:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][640/1519] eta 0:14:45 lr 0.000000 time 0.9332 (1.0070) model_time 0.9331 (1.0055) loss 0.6589 (0.8035) grad_norm 8.5607 (8.9427/2.4542) mem 68106MB [2022-12-20 23:09:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][650/1519] eta 0:14:35 lr 0.000000 time 0.9303 (1.0069) model_time 0.9301 (1.0054) loss 0.8187 (0.8038) grad_norm 9.8585 (8.8734/2.3026) mem 68106MB [2022-12-20 23:09:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][660/1519] eta 0:14:24 lr 0.000000 time 0.9244 (1.0068) model_time 0.9242 (1.0053) loss 0.6943 (0.8037) grad_norm 8.1811 (8.8840/2.3050) mem 68106MB [2022-12-20 23:09:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][670/1519] eta 0:14:14 lr 0.000000 time 0.9224 (1.0066) model_time 0.9222 (1.0052) loss 1.1266 (0.8041) grad_norm 7.1343 (8.8669/2.2965) mem 68106MB [2022-12-20 23:09:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][680/1519] eta 0:14:04 lr 0.000000 time 0.9264 (1.0065) model_time 0.9262 (1.0051) loss 0.8818 (0.8050) grad_norm 9.2536 (8.8269/2.2675) mem 68106MB [2022-12-20 23:09:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][690/1519] eta 0:13:54 lr 0.000000 time 0.9287 (1.0065) model_time 0.9285 (1.0050) loss 0.8539 (0.8040) grad_norm 12.5402 (8.8520/2.2837) mem 68106MB [2022-12-20 23:10:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][700/1519] eta 0:13:44 lr 0.000000 time 0.9298 (1.0063) model_time 0.9296 (1.0049) loss 0.7887 (0.8038) grad_norm 7.0814 (8.8854/2.2976) mem 68106MB [2022-12-20 23:10:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][710/1519] eta 0:13:34 lr 0.000000 time 0.9215 (1.0062) model_time 0.9213 (1.0048) loss 0.9487 (0.8039) grad_norm 12.4649 (8.8814/2.2397) mem 68106MB [2022-12-20 23:10:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][720/1519] eta 0:13:24 lr 0.000000 time 1.0096 (1.0063) model_time 1.0095 (1.0049) loss 0.7079 (0.8031) grad_norm 7.2902 (8.8776/2.2420) mem 68106MB [2022-12-20 23:10:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][730/1519] eta 0:13:13 lr 0.000000 time 0.9529 (1.0063) model_time 0.9527 (1.0050) loss 0.7213 (0.8027) grad_norm 12.7005 (8.9115/2.2867) mem 68106MB [2022-12-20 23:10:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][740/1519] eta 0:13:03 lr 0.000000 time 0.9340 (1.0063) model_time 0.9338 (1.0049) loss 0.7071 (0.8024) grad_norm 7.1349 (8.9088/2.2932) mem 68106MB [2022-12-20 23:10:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][750/1519] eta 0:12:53 lr 0.000000 time 0.9292 (1.0061) model_time 0.9291 (1.0048) loss 0.7810 (0.8025) grad_norm 7.2028 (8.9259/2.2972) mem 68106MB [2022-12-20 23:11:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][760/1519] eta 0:12:43 lr 0.000000 time 0.9181 (1.0064) model_time 0.9175 (1.0051) loss 0.7170 (0.8019) grad_norm 10.4527 (8.9531/2.3355) mem 68106MB [2022-12-20 23:11:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][770/1519] eta 0:12:33 lr 0.000000 time 0.9387 (1.0065) model_time 0.9384 (1.0052) loss 0.9059 (0.8033) grad_norm 8.1964 (8.9235/2.3287) mem 68106MB [2022-12-20 23:11:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][780/1519] eta 0:12:23 lr 0.000000 time 0.9295 (1.0066) model_time 0.9293 (1.0053) loss 0.7645 (0.8031) grad_norm 6.9427 (8.9361/2.3227) mem 68106MB [2022-12-20 23:11:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][790/1519] eta 0:12:13 lr 0.000000 time 0.9303 (1.0065) model_time 0.9301 (1.0052) loss 0.8441 (0.8033) grad_norm 9.0206 (8.9314/2.3277) mem 68106MB [2022-12-20 23:11:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][800/1519] eta 0:12:03 lr 0.000000 time 0.9251 (1.0065) model_time 0.9250 (1.0053) loss 1.1279 (0.8032) grad_norm 11.2800 (8.9542/2.3241) mem 68106MB [2022-12-20 23:11:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][810/1519] eta 0:11:53 lr 0.000000 time 0.9307 (1.0064) model_time 0.9304 (1.0052) loss 0.7810 (0.8040) grad_norm 7.4670 (8.9589/2.3399) mem 68106MB [2022-12-20 23:12:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][820/1519] eta 0:11:43 lr 0.000000 time 0.9185 (1.0063) model_time 0.9184 (1.0051) loss 0.7454 (0.8041) grad_norm 9.3978 (8.9463/2.3102) mem 68106MB [2022-12-20 23:12:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][830/1519] eta 0:11:33 lr 0.000000 time 0.9272 (1.0062) model_time 0.9270 (1.0050) loss 0.7826 (0.8037) grad_norm 7.0010 (8.9149/2.3081) mem 68106MB [2022-12-20 23:12:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][840/1519] eta 0:11:23 lr 0.000000 time 0.9297 (1.0061) model_time 0.9295 (1.0049) loss 0.8180 (0.8033) grad_norm 6.5654 (8.9485/2.3314) mem 68106MB [2022-12-20 23:12:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][850/1519] eta 0:11:13 lr 0.000000 time 0.9331 (1.0062) model_time 0.9329 (1.0050) loss 0.6640 (0.8022) grad_norm 8.5072 (8.9494/2.3394) mem 68106MB [2022-12-20 23:12:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][860/1519] eta 0:11:03 lr 0.000000 time 0.9303 (1.0061) model_time 0.9301 (1.0049) loss 0.6734 (0.8022) grad_norm 9.9410 (8.9688/2.3326) mem 68106MB [2022-12-20 23:12:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][870/1519] eta 0:10:52 lr 0.000000 time 0.9280 (1.0061) model_time 0.9278 (1.0049) loss 0.7674 (0.8015) grad_norm 7.1100 (8.9516/2.3366) mem 68106MB [2022-12-20 23:13:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][880/1519] eta 0:10:42 lr 0.000000 time 0.9130 (1.0061) model_time 0.9128 (1.0049) loss 0.6803 (0.8020) grad_norm 7.8747 (8.9587/2.3314) mem 68106MB [2022-12-20 23:13:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][890/1519] eta 0:10:32 lr 0.000000 time 0.9275 (1.0060) model_time 0.9271 (1.0049) loss 0.6772 (0.8022) grad_norm 11.6603 (8.9611/2.3452) mem 68106MB [2022-12-20 23:13:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][900/1519] eta 0:10:22 lr 0.000000 time 0.9314 (1.0060) model_time 0.9312 (1.0048) loss 0.9637 (0.8024) grad_norm 6.8193 (8.9606/2.3410) mem 68106MB [2022-12-20 23:13:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][910/1519] eta 0:10:12 lr 0.000000 time 0.9344 (1.0059) model_time 0.9342 (1.0048) loss 0.7064 (0.8028) grad_norm 10.1349 (8.9875/2.3431) mem 68106MB [2022-12-20 23:13:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][920/1519] eta 0:10:02 lr 0.000000 time 0.9351 (1.0059) model_time 0.9349 (1.0047) loss 0.7316 (0.8037) grad_norm 8.4398 (8.9418/2.3117) mem 68106MB [2022-12-20 23:13:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][930/1519] eta 0:09:52 lr 0.000000 time 0.9374 (1.0059) model_time 0.9373 (1.0047) loss 1.1636 (0.8043) grad_norm 9.1599 (8.9186/2.3042) mem 68106MB [2022-12-20 23:14:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][940/1519] eta 0:09:42 lr 0.000000 time 0.9329 (1.0058) model_time 0.9327 (1.0047) loss 0.7093 (0.8044) grad_norm 8.8539 (8.9002/2.2627) mem 68106MB [2022-12-20 23:14:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][950/1519] eta 0:09:32 lr 0.000000 time 0.9297 (1.0058) model_time 0.9295 (1.0047) loss 0.8812 (0.8039) grad_norm 6.0508 (8.9111/2.2575) mem 68106MB [2022-12-20 23:14:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][960/1519] eta 0:09:22 lr 0.000000 time 0.9236 (1.0058) model_time 0.9235 (1.0047) loss 0.7223 (0.8041) grad_norm 10.6708 (8.9192/2.2855) mem 68106MB [2022-12-20 23:14:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][970/1519] eta 0:09:12 lr 0.000000 time 0.9342 (1.0057) model_time 0.9340 (1.0046) loss 0.7184 (0.8041) grad_norm 7.3644 (8.8992/2.2928) mem 68106MB [2022-12-20 23:14:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][980/1519] eta 0:09:02 lr 0.000000 time 0.9351 (1.0057) model_time 0.9350 (1.0046) loss 0.7902 (0.8040) grad_norm 6.5007 (8.8757/2.2675) mem 68106MB [2022-12-20 23:14:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][990/1519] eta 0:08:51 lr 0.000000 time 0.9345 (1.0056) model_time 0.9343 (1.0045) loss 0.8929 (0.8035) grad_norm 24.2639 (8.9416/2.4215) mem 68106MB [2022-12-20 23:15:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1000/1519] eta 0:08:41 lr 0.000000 time 0.9277 (1.0056) model_time 0.9276 (1.0045) loss 1.0433 (0.8042) grad_norm 7.5303 (8.9573/2.4250) mem 68106MB [2022-12-20 23:15:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1010/1519] eta 0:08:31 lr 0.000000 time 0.9264 (1.0055) model_time 0.9263 (1.0045) loss 0.6947 (0.8040) grad_norm 9.5372 (8.9509/2.4196) mem 68106MB [2022-12-20 23:15:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1020/1519] eta 0:08:21 lr 0.000000 time 0.9287 (1.0055) model_time 0.9286 (1.0044) loss 0.6685 (0.8040) grad_norm 7.6246 (8.9337/2.4043) mem 68106MB [2022-12-20 23:15:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1030/1519] eta 0:08:11 lr 0.000000 time 0.9266 (1.0054) model_time 0.9264 (1.0044) loss 0.9936 (0.8039) grad_norm 9.6382 (8.9099/2.3135) mem 68106MB [2022-12-20 23:15:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1040/1519] eta 0:08:01 lr 0.000000 time 0.9770 (1.0056) model_time 0.9769 (1.0045) loss 0.6886 (0.8034) grad_norm 6.8168 (8.8967/2.3140) mem 68106MB [2022-12-20 23:15:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1050/1519] eta 0:07:51 lr 0.000000 time 0.9243 (1.0057) model_time 0.9241 (1.0046) loss 0.6820 (0.8032) grad_norm 27.1599 (8.9543/2.5425) mem 68106MB [2022-12-20 23:16:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1060/1519] eta 0:07:41 lr 0.000000 time 0.9245 (1.0056) model_time 0.9243 (1.0045) loss 0.7857 (0.8033) grad_norm 5.6885 (8.9264/2.5571) mem 68106MB [2022-12-20 23:16:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1070/1519] eta 0:07:31 lr 0.000000 time 0.9876 (1.0056) model_time 0.9874 (1.0046) loss 0.7739 (0.8030) grad_norm 6.4806 (8.9150/2.5233) mem 68106MB [2022-12-20 23:16:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1080/1519] eta 0:07:21 lr 0.000000 time 0.9351 (1.0056) model_time 0.9350 (1.0046) loss 0.9518 (0.8037) grad_norm 11.3402 (8.9424/2.5137) mem 68106MB [2022-12-20 23:16:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1090/1519] eta 0:07:11 lr 0.000000 time 1.0118 (1.0057) model_time 1.0117 (1.0047) loss 0.6831 (0.8036) grad_norm 7.6390 (8.9247/2.5209) mem 68106MB [2022-12-20 23:16:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1100/1519] eta 0:07:01 lr 0.000000 time 0.9309 (1.0056) model_time 0.9307 (1.0046) loss 0.7648 (0.8035) grad_norm 6.9291 (8.9123/2.5202) mem 68106MB [2022-12-20 23:16:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1110/1519] eta 0:06:51 lr 0.000000 time 0.9551 (1.0056) model_time 0.9549 (1.0046) loss 0.6689 (0.8027) grad_norm 10.9541 (8.9265/2.5224) mem 68106MB [2022-12-20 23:17:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1120/1519] eta 0:06:41 lr 0.000000 time 0.9279 (1.0056) model_time 0.9277 (1.0045) loss 0.7861 (0.8024) grad_norm 6.3680 (8.8942/2.5288) mem 68106MB [2022-12-20 23:17:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1130/1519] eta 0:06:31 lr 0.000000 time 0.9291 (1.0055) model_time 0.9289 (1.0045) loss 0.7196 (0.8024) grad_norm 8.6377 (8.8833/2.5254) mem 68106MB [2022-12-20 23:17:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1140/1519] eta 0:06:21 lr 0.000000 time 0.9226 (1.0055) model_time 0.9224 (1.0044) loss 0.7212 (0.8020) grad_norm 8.1190 (8.8494/2.5258) mem 68106MB [2022-12-20 23:17:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1150/1519] eta 0:06:10 lr 0.000000 time 0.9314 (1.0054) model_time 0.9312 (1.0044) loss 0.9587 (0.8023) grad_norm 7.2915 (8.8332/2.5268) mem 68106MB [2022-12-20 23:17:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1160/1519] eta 0:06:00 lr 0.000000 time 0.9298 (1.0054) model_time 0.9296 (1.0044) loss 0.6700 (0.8028) grad_norm 6.3274 (8.8363/2.5236) mem 68106MB [2022-12-20 23:17:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1170/1519] eta 0:05:50 lr 0.000000 time 0.9276 (1.0054) model_time 0.9275 (1.0044) loss 0.8478 (0.8025) grad_norm 9.0025 (8.8222/2.4827) mem 68106MB [2022-12-20 23:18:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1180/1519] eta 0:05:40 lr 0.000000 time 0.9287 (1.0053) model_time 0.9285 (1.0043) loss 0.6580 (0.8025) grad_norm 7.7078 (8.7936/2.4196) mem 68106MB [2022-12-20 23:18:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1190/1519] eta 0:05:30 lr 0.000000 time 0.9279 (1.0053) model_time 0.9277 (1.0043) loss 0.9573 (0.8026) grad_norm 9.8495 (8.7888/2.4087) mem 68106MB [2022-12-20 23:18:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1200/1519] eta 0:05:20 lr 0.000000 time 0.9231 (1.0053) model_time 0.9230 (1.0043) loss 0.6949 (0.8026) grad_norm 10.7528 (8.7640/2.4076) mem 68106MB [2022-12-20 23:18:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1210/1519] eta 0:05:10 lr 0.000000 time 0.9160 (1.0053) model_time 0.9158 (1.0043) loss 0.8725 (0.8021) grad_norm 8.9141 (8.7403/2.3627) mem 68106MB [2022-12-20 23:18:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1220/1519] eta 0:05:00 lr 0.000000 time 1.0076 (1.0054) model_time 1.0075 (1.0044) loss 0.6629 (0.8024) grad_norm 6.8921 (8.7282/2.3601) mem 68106MB [2022-12-20 23:18:54 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1230/1519] eta 0:04:50 lr 0.000000 time 0.9280 (1.0054) model_time 0.9278 (1.0044) loss 0.8983 (0.8027) grad_norm 7.1178 (8.7439/2.3587) mem 68106MB [2022-12-20 23:19:04 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1240/1519] eta 0:04:40 lr 0.000000 time 0.9355 (1.0054) model_time 0.9354 (1.0045) loss 0.7227 (0.8025) grad_norm 7.4375 (8.7493/2.3610) mem 68106MB [2022-12-20 23:19:14 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1250/1519] eta 0:04:30 lr 0.000000 time 0.9239 (1.0054) model_time 0.9238 (1.0044) loss 0.6854 (0.8021) grad_norm 8.5376 (8.7358/2.3632) mem 68106MB [2022-12-20 23:19:24 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1260/1519] eta 0:04:20 lr 0.000000 time 0.9191 (1.0054) model_time 0.9189 (1.0045) loss 0.6743 (0.8015) grad_norm 13.0678 (8.7648/2.3814) mem 68106MB [2022-12-20 23:19:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1270/1519] eta 0:04:10 lr 0.000000 time 1.0176 (1.0055) model_time 1.0174 (1.0045) loss 1.0208 (0.8015) grad_norm 6.9726 (8.7774/2.4094) mem 68106MB [2022-12-20 23:19:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1280/1519] eta 0:04:00 lr 0.000000 time 0.9261 (1.0054) model_time 0.9260 (1.0045) loss 1.2556 (0.8019) grad_norm 13.4408 (8.7978/2.4248) mem 68106MB [2022-12-20 23:19:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1290/1519] eta 0:03:50 lr 0.000000 time 0.9222 (1.0055) model_time 0.9221 (1.0046) loss 0.8889 (0.8015) grad_norm 7.3455 (8.7755/2.4177) mem 68106MB [2022-12-20 23:20:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1300/1519] eta 0:03:40 lr 0.000000 time 0.9325 (1.0055) model_time 0.9323 (1.0045) loss 0.6788 (0.8015) grad_norm 6.9587 (8.7464/2.4001) mem 68106MB [2022-12-20 23:20:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1310/1519] eta 0:03:30 lr 0.000000 time 0.9311 (1.0055) model_time 0.9310 (1.0046) loss 0.9007 (0.8020) grad_norm 9.1501 (8.6861/2.3780) mem 68106MB [2022-12-20 23:20:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1320/1519] eta 0:03:20 lr 0.000000 time 0.9376 (1.0054) model_time 0.9375 (1.0045) loss 0.6883 (0.8022) grad_norm 6.4074 (8.6798/2.3711) mem 68106MB [2022-12-20 23:20:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1330/1519] eta 0:03:10 lr 0.000000 time 0.9307 (1.0054) model_time 0.9305 (1.0045) loss 0.7248 (0.8020) grad_norm 7.0305 (8.6269/2.3229) mem 68106MB [2022-12-20 23:20:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1340/1519] eta 0:02:59 lr 0.000000 time 0.9188 (1.0054) model_time 0.9187 (1.0045) loss 0.7631 (0.8016) grad_norm 8.5940 (8.6322/2.3198) mem 68106MB [2022-12-20 23:20:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1350/1519] eta 0:02:49 lr 0.000000 time 0.9225 (1.0055) model_time 0.9223 (1.0046) loss 0.7644 (0.8013) grad_norm 10.3276 (8.6350/2.3274) mem 68106MB [2022-12-20 23:21:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1360/1519] eta 0:02:39 lr 0.000000 time 0.9336 (1.0056) model_time 0.9335 (1.0047) loss 0.8434 (0.8012) grad_norm 7.2057 (8.6307/2.3528) mem 68106MB [2022-12-20 23:21:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1370/1519] eta 0:02:29 lr 0.000000 time 0.9234 (1.0055) model_time 0.9230 (1.0046) loss 0.9189 (0.8013) grad_norm 7.6427 (8.6178/2.3391) mem 68106MB [2022-12-20 23:21:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1380/1519] eta 0:02:19 lr 0.000000 time 0.9680 (1.0055) model_time 0.9679 (1.0046) loss 0.9269 (0.8017) grad_norm 10.5231 (8.6280/2.3383) mem 68106MB [2022-12-20 23:21:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1390/1519] eta 0:02:09 lr 0.000000 time 0.9299 (1.0055) model_time 0.9297 (1.0046) loss 0.8857 (0.8015) grad_norm 9.0477 (8.6529/2.3744) mem 68106MB [2022-12-20 23:21:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1400/1519] eta 0:01:59 lr 0.000000 time 0.9318 (1.0055) model_time 0.9316 (1.0046) loss 0.7179 (0.8020) grad_norm 7.1611 (8.6422/2.3788) mem 68106MB [2022-12-20 23:21:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1410/1519] eta 0:01:49 lr 0.000000 time 0.9346 (1.0055) model_time 0.9345 (1.0046) loss 0.7012 (0.8019) grad_norm 10.1651 (8.6354/2.3664) mem 68106MB [2022-12-20 23:22:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1420/1519] eta 0:01:39 lr 0.000000 time 0.9352 (1.0055) model_time 0.9351 (1.0046) loss 0.7299 (0.8018) grad_norm 8.8511 (8.6087/2.3635) mem 68106MB [2022-12-20 23:22:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1430/1519] eta 0:01:29 lr 0.000000 time 0.9264 (1.0055) model_time 0.9262 (1.0046) loss 0.7131 (0.8018) grad_norm 7.1542 (8.6088/2.3608) mem 68106MB [2022-12-20 23:22:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1440/1519] eta 0:01:19 lr 0.000000 time 0.9274 (1.0055) model_time 0.9273 (1.0046) loss 0.6777 (0.8013) grad_norm 7.3510 (8.5475/2.3455) mem 68106MB [2022-12-20 23:22:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1450/1519] eta 0:01:09 lr 0.000000 time 0.9243 (1.0054) model_time 0.9241 (1.0045) loss 0.7711 (0.8010) grad_norm 7.1623 (8.5658/2.3930) mem 68106MB [2022-12-20 23:22:45 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1460/1519] eta 0:00:59 lr 0.000000 time 0.9319 (1.0054) model_time 0.9316 (1.0045) loss 0.6966 (0.8012) grad_norm 8.2414 (8.5747/2.4018) mem 68106MB [2022-12-20 23:22:55 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1470/1519] eta 0:00:49 lr 0.000000 time 0.9283 (1.0054) model_time 0.9282 (1.0046) loss 1.1173 (0.8012) grad_norm 7.1541 (8.5654/2.3996) mem 68106MB [2022-12-20 23:23:05 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1480/1519] eta 0:00:39 lr 0.000000 time 0.9196 (1.0054) model_time 0.9194 (1.0045) loss 0.9718 (0.8017) grad_norm 9.6645 (8.5617/2.4021) mem 68106MB [2022-12-20 23:23:15 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1490/1519] eta 0:00:29 lr 0.000000 time 0.9194 (1.0054) model_time 0.9193 (1.0045) loss 0.7105 (0.8013) grad_norm 7.9423 (8.5376/2.3947) mem 68106MB [2022-12-20 23:23:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1500/1519] eta 0:00:19 lr 0.000000 time 0.9294 (1.0053) model_time 0.9293 (1.0045) loss 0.6633 (0.8012) grad_norm 9.7338 (8.5787/2.4544) mem 68106MB [2022-12-20 23:23:36 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [97/100][1510/1519] eta 0:00:09 lr 0.000000 time 0.9368 (1.0054) model_time 0.9367 (1.0046) loss 0.6727 (0.8010) grad_norm 8.1186 (8.5468/2.4491) mem 68106MB [2022-12-20 23:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 450): INFO EPOCH 97 training takes 0:25:27 [2022-12-20 23:23:44 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_97.pth saving...... [2022-12-20 23:24:09 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_97.pth saved !!! [2022-12-20 23:24:10 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [0/85] Time 0.635 (0.635) Loss 0.5388 (0.5388) Acc@1 92.708 (92.708) Acc@5 98.611 (98.611) Mem 68106MB [2022-12-20 23:24:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [10/85] Time 0.297 (0.329) Loss 0.5339 (0.5080) Acc@1 92.014 (92.803) Acc@5 98.611 (98.485) Mem 68106MB [2022-12-20 23:24:16 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [20/85] Time 0.297 (0.314) Loss 0.4886 (0.5036) Acc@1 91.319 (92.791) Acc@5 99.306 (98.495) Mem 68106MB [2022-12-20 23:24:19 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [30/85] Time 0.298 (0.309) Loss 0.6354 (0.5107) Acc@1 90.972 (92.585) Acc@5 98.264 (98.466) Mem 68106MB [2022-12-20 23:24:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [40/85] Time 0.300 (0.307) Loss 0.4606 (0.5015) Acc@1 93.750 (92.674) Acc@5 99.306 (98.569) Mem 68106MB [2022-12-20 23:24:25 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [50/85] Time 0.300 (0.306) Loss 0.4936 (0.4990) Acc@1 92.361 (92.749) Acc@5 99.653 (98.625) Mem 68106MB [2022-12-20 23:24:28 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [60/85] Time 0.297 (0.305) Loss 0.5104 (0.4988) Acc@1 90.972 (92.674) Acc@5 98.264 (98.600) Mem 68106MB [2022-12-20 23:24:31 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [70/85] Time 0.295 (0.304) Loss 0.5464 (0.5000) Acc@1 93.056 (92.640) Acc@5 97.917 (98.577) Mem 68106MB [2022-12-20 23:24:34 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 522): INFO Test: [80/85] Time 0.294 (0.303) Loss 0.4312 (0.4987) Acc@1 92.708 (92.661) Acc@5 98.264 (98.603) Mem 68106MB [2022-12-20 23:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 530): INFO [Epoch:97] * Acc@1 92.633 Acc@5 98.604 [2022-12-20 23:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 283): INFO Accuracy of the network on the 24426 test images: 92.6% [2022-12-20 23:24:35 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 348): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saving...... [2022-12-20 23:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (utils.py 350): INFO work_dirs/dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384/ckpt_epoch_best.pth saved !!! [2022-12-20 23:25:00 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 291): INFO Max accuracy: 92.63% [2022-12-20 23:25:01 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][0/1519] eta 0:36:00 lr 0.000000 time 1.4222 (1.4222) model_time 0.9781 (0.9781) loss 0.7061 (0.7061) grad_norm 0.0000 (0.0000/0.0000) mem 68106MB [2022-12-20 23:25:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][10/1519] eta 0:26:32 lr 0.000000 time 0.9274 (1.0553) model_time 0.9272 (1.0146) loss 0.7187 (0.7568) grad_norm 9.4506 (7.7054/1.7815) mem 68106MB [2022-12-20 23:25:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][20/1519] eta 0:25:49 lr 0.000000 time 0.9334 (1.0338) model_time 0.9333 (1.0123) loss 0.7916 (0.7644) grad_norm 13.6787 (8.4998/2.4817) mem 68106MB [2022-12-20 23:25:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][30/1519] eta 0:25:25 lr 0.000000 time 0.9830 (1.0247) model_time 0.9828 (1.0101) loss 0.6854 (0.7743) grad_norm 15.0320 (9.3603/2.7222) mem 68106MB [2022-12-20 23:25:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][40/1519] eta 0:25:06 lr 0.000000 time 0.9214 (1.0183) model_time 0.9213 (1.0071) loss 0.7066 (0.7863) grad_norm 8.6073 (9.0356/2.4424) mem 68106MB [2022-12-20 23:25:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][50/1519] eta 0:24:56 lr 0.000000 time 0.9782 (1.0187) model_time 0.9780 (1.0096) loss 1.0650 (0.7977) grad_norm 9.7086 (8.9648/2.2643) mem 68106MB [2022-12-20 23:26:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][60/1519] eta 0:24:41 lr 0.000000 time 0.9278 (1.0151) model_time 0.9276 (1.0075) loss 0.8176 (0.8011) grad_norm 9.4625 (9.0100/2.2837) mem 68106MB [2022-12-20 23:26:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][70/1519] eta 0:24:29 lr 0.000000 time 0.9319 (1.0141) model_time 0.9318 (1.0074) loss 0.6815 (0.8054) grad_norm 9.6528 (8.8987/2.1850) mem 68106MB [2022-12-20 23:26:22 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][80/1519] eta 0:24:18 lr 0.000000 time 0.9177 (1.0133) model_time 0.9175 (1.0075) loss 0.7064 (0.8040) grad_norm 8.6899 (8.8039/2.0907) mem 68106MB [2022-12-20 23:26:32 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][90/1519] eta 0:24:07 lr 0.000000 time 1.0043 (1.0128) model_time 1.0041 (1.0075) loss 0.7061 (0.7968) grad_norm 6.6858 (8.9873/2.5381) mem 68106MB [2022-12-20 23:26:42 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][100/1519] eta 0:23:56 lr 0.000000 time 0.9272 (1.0121) model_time 0.9270 (1.0073) loss 0.8924 (0.7966) grad_norm 6.8439 (9.0155/2.5730) mem 68106MB [2022-12-20 23:26:52 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][110/1519] eta 0:23:45 lr 0.000000 time 0.9245 (1.0120) model_time 0.9243 (1.0076) loss 0.8979 (0.7946) grad_norm 9.3514 (8.9974/2.5147) mem 68106MB [2022-12-20 23:27:02 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][120/1519] eta 0:23:35 lr 0.000000 time 0.9537 (1.0116) model_time 0.9536 (1.0076) loss 0.7015 (0.7884) grad_norm 14.4917 (8.9663/2.5704) mem 68106MB [2022-12-20 23:27:12 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][130/1519] eta 0:23:23 lr 0.000000 time 0.9325 (1.0106) model_time 0.9324 (1.0069) loss 0.7799 (0.7870) grad_norm 10.4338 (8.8819/2.5164) mem 68106MB [2022-12-20 23:27:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][140/1519] eta 0:23:13 lr 0.000000 time 0.9210 (1.0103) model_time 0.9208 (1.0068) loss 0.6556 (0.7891) grad_norm 8.9098 (8.8420/2.4536) mem 68106MB [2022-12-20 23:27:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][150/1519] eta 0:23:03 lr 0.000000 time 0.9227 (1.0106) model_time 0.9226 (1.0073) loss 0.8660 (0.7876) grad_norm 8.0629 (8.7468/2.4109) mem 68106MB [2022-12-20 23:27:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][160/1519] eta 0:22:52 lr 0.000000 time 0.9183 (1.0100) model_time 0.9181 (1.0068) loss 0.8129 (0.7892) grad_norm 12.5407 (8.8005/2.4183) mem 68106MB [2022-12-20 23:27:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][170/1519] eta 0:22:41 lr 0.000000 time 0.9289 (1.0093) model_time 0.9287 (1.0064) loss 0.9134 (0.7892) grad_norm 8.6055 (8.7241/2.3826) mem 68106MB [2022-12-20 23:28:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][180/1519] eta 0:22:30 lr 0.000000 time 0.9261 (1.0088) model_time 0.9259 (1.0060) loss 0.6610 (0.7874) grad_norm 8.8993 (8.7944/2.3565) mem 68106MB [2022-12-20 23:28:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][190/1519] eta 0:22:20 lr 0.000000 time 0.9302 (1.0088) model_time 0.9301 (1.0061) loss 0.7202 (0.7891) grad_norm 7.2260 (9.0340/3.2321) mem 68106MB [2022-12-20 23:28:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][200/1519] eta 0:22:10 lr 0.000000 time 0.9215 (1.0087) model_time 0.9213 (1.0061) loss 0.7958 (0.7926) grad_norm 7.1436 (8.9738/3.1664) mem 68106MB [2022-12-20 23:28:33 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][210/1519] eta 0:21:59 lr 0.000000 time 0.9238 (1.0083) model_time 0.9237 (1.0058) loss 0.8021 (0.7948) grad_norm 10.5265 (9.0249/3.1167) mem 68106MB [2022-12-20 23:28:43 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][220/1519] eta 0:21:49 lr 0.000000 time 0.9389 (1.0079) model_time 0.9387 (1.0055) loss 0.8525 (0.7950) grad_norm 10.1152 (9.0517/3.0647) mem 68106MB [2022-12-20 23:28:53 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][230/1519] eta 0:21:39 lr 0.000000 time 0.9715 (1.0084) model_time 0.9714 (1.0061) loss 0.7854 (0.7928) grad_norm 12.6112 (9.0296/3.0357) mem 68106MB [2022-12-20 23:29:03 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][240/1519] eta 0:21:29 lr 0.000000 time 0.9276 (1.0083) model_time 0.9274 (1.0061) loss 0.7612 (0.7941) grad_norm 7.9901 (9.0859/3.0181) mem 68106MB [2022-12-20 23:29:13 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][250/1519] eta 0:21:19 lr 0.000000 time 0.9370 (1.0083) model_time 0.9368 (1.0062) loss 0.8782 (0.7938) grad_norm 7.6780 (9.0452/2.9706) mem 68106MB [2022-12-20 23:29:23 dcnv3_5_h_new3_1k_224_ft_try28_1_inat_384] (main_inat18.py 440): INFO Train: [98/100][260/1519] eta 0:21:09 lr 0.000000 time 0.9364 (1.0082) model_time 0.9363 (1.0062) loss 0.6819 (0.7934) grad_norm 8.8173 (9.0399/2.9297) mem 68106MB