2023-10-17 14:52:54,571 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): ElectraModel( (embeddings): ElectraEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): ElectraEncoder( (layer): ModuleList( (0-11): 12 x ElectraLayer( (attention): ElectraAttention( (self): ElectraSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): ElectraSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): ElectraIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): ElectraOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 MultiCorpus: 7142 train + 698 dev + 2570 test sentences - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Train: 7142 sentences 2023-10-17 14:52:54,572 (train_with_dev=False, train_with_test=False) 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Training Params: 2023-10-17 14:52:54,572 - learning_rate: "5e-05" 2023-10-17 14:52:54,572 - mini_batch_size: "4" 2023-10-17 14:52:54,572 - max_epochs: "10" 2023-10-17 14:52:54,572 - shuffle: "True" 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Plugins: 2023-10-17 14:52:54,572 - TensorboardLogger 2023-10-17 14:52:54,572 - LinearScheduler | warmup_fraction: '0.1' 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Final evaluation on model from best epoch (best-model.pt) 2023-10-17 14:52:54,572 - metric: "('micro avg', 'f1-score')" 2023-10-17 14:52:54,572 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,572 Computation: 2023-10-17 14:52:54,572 - compute on device: cuda:0 2023-10-17 14:52:54,573 - embedding storage: none 2023-10-17 14:52:54,573 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,573 Model training base path: "hmbench-newseye/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3" 2023-10-17 14:52:54,573 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,573 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:52:54,573 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-17 14:53:03,192 epoch 1 - iter 178/1786 - loss 2.34357014 - time (sec): 8.62 - samples/sec: 2790.75 - lr: 0.000005 - momentum: 0.000000 2023-10-17 14:53:11,959 epoch 1 - iter 356/1786 - loss 1.40781693 - time (sec): 17.39 - samples/sec: 2809.96 - lr: 0.000010 - momentum: 0.000000 2023-10-17 14:53:20,638 epoch 1 - iter 534/1786 - loss 1.06730323 - time (sec): 26.06 - samples/sec: 2742.77 - lr: 0.000015 - momentum: 0.000000 2023-10-17 14:53:29,450 epoch 1 - iter 712/1786 - loss 0.84752372 - time (sec): 34.88 - samples/sec: 2786.76 - lr: 0.000020 - momentum: 0.000000 2023-10-17 14:53:38,306 epoch 1 - iter 890/1786 - loss 0.71167479 - time (sec): 43.73 - samples/sec: 2802.24 - lr: 0.000025 - momentum: 0.000000 2023-10-17 14:53:47,145 epoch 1 - iter 1068/1786 - loss 0.62544453 - time (sec): 52.57 - samples/sec: 2794.21 - lr: 0.000030 - momentum: 0.000000 2023-10-17 14:53:56,284 epoch 1 - iter 1246/1786 - loss 0.55608152 - time (sec): 61.71 - samples/sec: 2792.42 - lr: 0.000035 - momentum: 0.000000 2023-10-17 14:54:05,381 epoch 1 - iter 1424/1786 - loss 0.50198941 - time (sec): 70.81 - samples/sec: 2811.79 - lr: 0.000040 - momentum: 0.000000 2023-10-17 14:54:14,199 epoch 1 - iter 1602/1786 - loss 0.46440329 - time (sec): 79.63 - samples/sec: 2819.64 - lr: 0.000045 - momentum: 0.000000 2023-10-17 14:54:22,832 epoch 1 - iter 1780/1786 - loss 0.43392835 - time (sec): 88.26 - samples/sec: 2809.33 - lr: 0.000050 - momentum: 0.000000 2023-10-17 14:54:23,104 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:54:23,104 EPOCH 1 done: loss 0.4330 - lr: 0.000050 2023-10-17 14:54:27,153 DEV : loss 0.11458766460418701 - f1-score (micro avg) 0.7403 2023-10-17 14:54:27,169 saving best model 2023-10-17 14:54:27,519 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:54:36,852 epoch 2 - iter 178/1786 - loss 0.14020957 - time (sec): 9.33 - samples/sec: 2564.60 - lr: 0.000049 - momentum: 0.000000 2023-10-17 14:54:46,095 epoch 2 - iter 356/1786 - loss 0.13246321 - time (sec): 18.57 - samples/sec: 2692.71 - lr: 0.000049 - momentum: 0.000000 2023-10-17 14:54:55,132 epoch 2 - iter 534/1786 - loss 0.12996578 - time (sec): 27.61 - samples/sec: 2710.66 - lr: 0.000048 - momentum: 0.000000 2023-10-17 14:55:04,131 epoch 2 - iter 712/1786 - loss 0.13085505 - time (sec): 36.61 - samples/sec: 2724.67 - lr: 0.000048 - momentum: 0.000000 2023-10-17 14:55:12,791 epoch 2 - iter 890/1786 - loss 0.13070280 - time (sec): 45.27 - samples/sec: 2727.42 - lr: 0.000047 - momentum: 0.000000 2023-10-17 14:55:21,707 epoch 2 - iter 1068/1786 - loss 0.12652829 - time (sec): 54.19 - samples/sec: 2744.97 - lr: 0.000047 - momentum: 0.000000 2023-10-17 14:55:30,604 epoch 2 - iter 1246/1786 - loss 0.12732731 - time (sec): 63.08 - samples/sec: 2736.02 - lr: 0.000046 - momentum: 0.000000 2023-10-17 14:55:39,573 epoch 2 - iter 1424/1786 - loss 0.12632951 - time (sec): 72.05 - samples/sec: 2752.70 - lr: 0.000046 - momentum: 0.000000 2023-10-17 14:55:48,453 epoch 2 - iter 1602/1786 - loss 0.12684493 - time (sec): 80.93 - samples/sec: 2764.66 - lr: 0.000045 - momentum: 0.000000 2023-10-17 14:55:57,231 epoch 2 - iter 1780/1786 - loss 0.12675945 - time (sec): 89.71 - samples/sec: 2764.76 - lr: 0.000044 - momentum: 0.000000 2023-10-17 14:55:57,511 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:55:57,512 EPOCH 2 done: loss 0.1265 - lr: 0.000044 2023-10-17 14:56:02,464 DEV : loss 0.1421259194612503 - f1-score (micro avg) 0.7695 2023-10-17 14:56:02,482 saving best model 2023-10-17 14:56:02,942 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:56:11,753 epoch 3 - iter 178/1786 - loss 0.09648196 - time (sec): 8.81 - samples/sec: 2560.72 - lr: 0.000044 - momentum: 0.000000 2023-10-17 14:56:20,746 epoch 3 - iter 356/1786 - loss 0.08669305 - time (sec): 17.80 - samples/sec: 2701.77 - lr: 0.000043 - momentum: 0.000000 2023-10-17 14:56:29,588 epoch 3 - iter 534/1786 - loss 0.08198309 - time (sec): 26.64 - samples/sec: 2747.94 - lr: 0.000043 - momentum: 0.000000 2023-10-17 14:56:38,450 epoch 3 - iter 712/1786 - loss 0.08374774 - time (sec): 35.51 - samples/sec: 2754.30 - lr: 0.000042 - momentum: 0.000000 2023-10-17 14:56:47,471 epoch 3 - iter 890/1786 - loss 0.08441770 - time (sec): 44.53 - samples/sec: 2744.81 - lr: 0.000042 - momentum: 0.000000 2023-10-17 14:56:56,230 epoch 3 - iter 1068/1786 - loss 0.08501205 - time (sec): 53.29 - samples/sec: 2735.07 - lr: 0.000041 - momentum: 0.000000 2023-10-17 14:57:05,041 epoch 3 - iter 1246/1786 - loss 0.08606738 - time (sec): 62.10 - samples/sec: 2772.35 - lr: 0.000041 - momentum: 0.000000 2023-10-17 14:57:13,860 epoch 3 - iter 1424/1786 - loss 0.08594360 - time (sec): 70.92 - samples/sec: 2790.52 - lr: 0.000040 - momentum: 0.000000 2023-10-17 14:57:22,820 epoch 3 - iter 1602/1786 - loss 0.08524957 - time (sec): 79.88 - samples/sec: 2805.33 - lr: 0.000039 - momentum: 0.000000 2023-10-17 14:57:31,499 epoch 3 - iter 1780/1786 - loss 0.08632208 - time (sec): 88.55 - samples/sec: 2799.84 - lr: 0.000039 - momentum: 0.000000 2023-10-17 14:57:31,759 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:57:31,760 EPOCH 3 done: loss 0.0866 - lr: 0.000039 2023-10-17 14:57:36,017 DEV : loss 0.16046550869941711 - f1-score (micro avg) 0.7866 2023-10-17 14:57:36,034 saving best model 2023-10-17 14:57:36,496 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:57:44,948 epoch 4 - iter 178/1786 - loss 0.05362232 - time (sec): 8.45 - samples/sec: 2870.63 - lr: 0.000038 - momentum: 0.000000 2023-10-17 14:57:54,046 epoch 4 - iter 356/1786 - loss 0.06672816 - time (sec): 17.55 - samples/sec: 2907.75 - lr: 0.000038 - momentum: 0.000000 2023-10-17 14:58:02,956 epoch 4 - iter 534/1786 - loss 0.06403011 - time (sec): 26.46 - samples/sec: 2887.17 - lr: 0.000037 - momentum: 0.000000 2023-10-17 14:58:11,815 epoch 4 - iter 712/1786 - loss 0.06474490 - time (sec): 35.32 - samples/sec: 2857.05 - lr: 0.000037 - momentum: 0.000000 2023-10-17 14:58:20,533 epoch 4 - iter 890/1786 - loss 0.06436458 - time (sec): 44.03 - samples/sec: 2846.34 - lr: 0.000036 - momentum: 0.000000 2023-10-17 14:58:29,239 epoch 4 - iter 1068/1786 - loss 0.06292209 - time (sec): 52.74 - samples/sec: 2841.50 - lr: 0.000036 - momentum: 0.000000 2023-10-17 14:58:37,846 epoch 4 - iter 1246/1786 - loss 0.06337824 - time (sec): 61.35 - samples/sec: 2817.24 - lr: 0.000035 - momentum: 0.000000 2023-10-17 14:58:46,836 epoch 4 - iter 1424/1786 - loss 0.06433756 - time (sec): 70.34 - samples/sec: 2813.22 - lr: 0.000034 - momentum: 0.000000 2023-10-17 14:58:55,515 epoch 4 - iter 1602/1786 - loss 0.06494053 - time (sec): 79.02 - samples/sec: 2820.58 - lr: 0.000034 - momentum: 0.000000 2023-10-17 14:59:04,297 epoch 4 - iter 1780/1786 - loss 0.06441300 - time (sec): 87.80 - samples/sec: 2825.51 - lr: 0.000033 - momentum: 0.000000 2023-10-17 14:59:04,567 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:59:04,567 EPOCH 4 done: loss 0.0646 - lr: 0.000033 2023-10-17 14:59:09,315 DEV : loss 0.18103061616420746 - f1-score (micro avg) 0.7779 2023-10-17 14:59:09,334 ---------------------------------------------------------------------------------------------------- 2023-10-17 14:59:18,302 epoch 5 - iter 178/1786 - loss 0.03339553 - time (sec): 8.97 - samples/sec: 2750.16 - lr: 0.000033 - momentum: 0.000000 2023-10-17 14:59:27,368 epoch 5 - iter 356/1786 - loss 0.03967613 - time (sec): 18.03 - samples/sec: 2817.11 - lr: 0.000032 - momentum: 0.000000 2023-10-17 14:59:36,320 epoch 5 - iter 534/1786 - loss 0.04092048 - time (sec): 26.98 - samples/sec: 2843.97 - lr: 0.000032 - momentum: 0.000000 2023-10-17 14:59:45,069 epoch 5 - iter 712/1786 - loss 0.04374407 - time (sec): 35.73 - samples/sec: 2841.71 - lr: 0.000031 - momentum: 0.000000 2023-10-17 14:59:53,905 epoch 5 - iter 890/1786 - loss 0.04683876 - time (sec): 44.57 - samples/sec: 2813.94 - lr: 0.000031 - momentum: 0.000000 2023-10-17 15:00:03,115 epoch 5 - iter 1068/1786 - loss 0.04672502 - time (sec): 53.78 - samples/sec: 2817.89 - lr: 0.000030 - momentum: 0.000000 2023-10-17 15:00:12,273 epoch 5 - iter 1246/1786 - loss 0.04806529 - time (sec): 62.94 - samples/sec: 2794.30 - lr: 0.000029 - momentum: 0.000000 2023-10-17 15:00:21,497 epoch 5 - iter 1424/1786 - loss 0.04855191 - time (sec): 72.16 - samples/sec: 2774.59 - lr: 0.000029 - momentum: 0.000000 2023-10-17 15:00:30,474 epoch 5 - iter 1602/1786 - loss 0.04738804 - time (sec): 81.14 - samples/sec: 2767.55 - lr: 0.000028 - momentum: 0.000000 2023-10-17 15:00:39,349 epoch 5 - iter 1780/1786 - loss 0.04826172 - time (sec): 90.01 - samples/sec: 2756.61 - lr: 0.000028 - momentum: 0.000000 2023-10-17 15:00:39,661 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:00:39,661 EPOCH 5 done: loss 0.0482 - lr: 0.000028 2023-10-17 15:00:44,200 DEV : loss 0.20736941695213318 - f1-score (micro avg) 0.7612 2023-10-17 15:00:44,219 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:00:53,164 epoch 6 - iter 178/1786 - loss 0.02216489 - time (sec): 8.94 - samples/sec: 2750.82 - lr: 0.000027 - momentum: 0.000000 2023-10-17 15:01:02,289 epoch 6 - iter 356/1786 - loss 0.03505341 - time (sec): 18.07 - samples/sec: 2819.10 - lr: 0.000027 - momentum: 0.000000 2023-10-17 15:01:11,960 epoch 6 - iter 534/1786 - loss 0.03233599 - time (sec): 27.74 - samples/sec: 2719.90 - lr: 0.000026 - momentum: 0.000000 2023-10-17 15:01:21,146 epoch 6 - iter 712/1786 - loss 0.03206542 - time (sec): 36.93 - samples/sec: 2693.98 - lr: 0.000026 - momentum: 0.000000 2023-10-17 15:01:30,650 epoch 6 - iter 890/1786 - loss 0.03265728 - time (sec): 46.43 - samples/sec: 2662.60 - lr: 0.000025 - momentum: 0.000000 2023-10-17 15:01:40,031 epoch 6 - iter 1068/1786 - loss 0.03291601 - time (sec): 55.81 - samples/sec: 2653.82 - lr: 0.000024 - momentum: 0.000000 2023-10-17 15:01:48,833 epoch 6 - iter 1246/1786 - loss 0.03440927 - time (sec): 64.61 - samples/sec: 2679.28 - lr: 0.000024 - momentum: 0.000000 2023-10-17 15:01:57,756 epoch 6 - iter 1424/1786 - loss 0.03461216 - time (sec): 73.54 - samples/sec: 2693.26 - lr: 0.000023 - momentum: 0.000000 2023-10-17 15:02:06,771 epoch 6 - iter 1602/1786 - loss 0.03499687 - time (sec): 82.55 - samples/sec: 2705.31 - lr: 0.000023 - momentum: 0.000000 2023-10-17 15:02:15,500 epoch 6 - iter 1780/1786 - loss 0.03462841 - time (sec): 91.28 - samples/sec: 2717.57 - lr: 0.000022 - momentum: 0.000000 2023-10-17 15:02:15,798 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:02:15,798 EPOCH 6 done: loss 0.0345 - lr: 0.000022 2023-10-17 15:02:20,235 DEV : loss 0.2402777373790741 - f1-score (micro avg) 0.8112 2023-10-17 15:02:20,253 saving best model 2023-10-17 15:02:20,692 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:02:29,984 epoch 7 - iter 178/1786 - loss 0.02884356 - time (sec): 9.29 - samples/sec: 2773.61 - lr: 0.000022 - momentum: 0.000000 2023-10-17 15:02:39,181 epoch 7 - iter 356/1786 - loss 0.02553577 - time (sec): 18.48 - samples/sec: 2715.73 - lr: 0.000021 - momentum: 0.000000 2023-10-17 15:02:48,182 epoch 7 - iter 534/1786 - loss 0.02422475 - time (sec): 27.48 - samples/sec: 2715.65 - lr: 0.000021 - momentum: 0.000000 2023-10-17 15:02:57,380 epoch 7 - iter 712/1786 - loss 0.02794156 - time (sec): 36.68 - samples/sec: 2752.78 - lr: 0.000020 - momentum: 0.000000 2023-10-17 15:03:06,635 epoch 7 - iter 890/1786 - loss 0.02949847 - time (sec): 45.94 - samples/sec: 2733.04 - lr: 0.000019 - momentum: 0.000000 2023-10-17 15:03:16,160 epoch 7 - iter 1068/1786 - loss 0.02887539 - time (sec): 55.46 - samples/sec: 2720.71 - lr: 0.000019 - momentum: 0.000000 2023-10-17 15:03:25,064 epoch 7 - iter 1246/1786 - loss 0.02971168 - time (sec): 64.37 - samples/sec: 2725.25 - lr: 0.000018 - momentum: 0.000000 2023-10-17 15:03:33,838 epoch 7 - iter 1424/1786 - loss 0.02979510 - time (sec): 73.14 - samples/sec: 2715.19 - lr: 0.000018 - momentum: 0.000000 2023-10-17 15:03:42,556 epoch 7 - iter 1602/1786 - loss 0.02897065 - time (sec): 81.86 - samples/sec: 2725.86 - lr: 0.000017 - momentum: 0.000000 2023-10-17 15:03:51,419 epoch 7 - iter 1780/1786 - loss 0.02808217 - time (sec): 90.72 - samples/sec: 2736.21 - lr: 0.000017 - momentum: 0.000000 2023-10-17 15:03:51,685 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:03:51,686 EPOCH 7 done: loss 0.0281 - lr: 0.000017 2023-10-17 15:03:56,572 DEV : loss 0.2226112335920334 - f1-score (micro avg) 0.7949 2023-10-17 15:03:56,590 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:04:05,424 epoch 8 - iter 178/1786 - loss 0.01609297 - time (sec): 8.83 - samples/sec: 2697.83 - lr: 0.000016 - momentum: 0.000000 2023-10-17 15:04:14,488 epoch 8 - iter 356/1786 - loss 0.01460834 - time (sec): 17.90 - samples/sec: 2783.84 - lr: 0.000016 - momentum: 0.000000 2023-10-17 15:04:23,260 epoch 8 - iter 534/1786 - loss 0.01586187 - time (sec): 26.67 - samples/sec: 2714.18 - lr: 0.000015 - momentum: 0.000000 2023-10-17 15:04:32,238 epoch 8 - iter 712/1786 - loss 0.01654564 - time (sec): 35.65 - samples/sec: 2737.96 - lr: 0.000014 - momentum: 0.000000 2023-10-17 15:04:41,330 epoch 8 - iter 890/1786 - loss 0.01727812 - time (sec): 44.74 - samples/sec: 2757.92 - lr: 0.000014 - momentum: 0.000000 2023-10-17 15:04:50,003 epoch 8 - iter 1068/1786 - loss 0.01761990 - time (sec): 53.41 - samples/sec: 2777.98 - lr: 0.000013 - momentum: 0.000000 2023-10-17 15:04:58,640 epoch 8 - iter 1246/1786 - loss 0.01726152 - time (sec): 62.05 - samples/sec: 2782.42 - lr: 0.000013 - momentum: 0.000000 2023-10-17 15:05:07,456 epoch 8 - iter 1424/1786 - loss 0.01689803 - time (sec): 70.86 - samples/sec: 2771.40 - lr: 0.000012 - momentum: 0.000000 2023-10-17 15:05:16,193 epoch 8 - iter 1602/1786 - loss 0.01640200 - time (sec): 79.60 - samples/sec: 2769.89 - lr: 0.000012 - momentum: 0.000000 2023-10-17 15:05:25,538 epoch 8 - iter 1780/1786 - loss 0.01650120 - time (sec): 88.95 - samples/sec: 2786.95 - lr: 0.000011 - momentum: 0.000000 2023-10-17 15:05:25,849 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:05:25,850 EPOCH 8 done: loss 0.0165 - lr: 0.000011 2023-10-17 15:05:30,212 DEV : loss 0.23785527050495148 - f1-score (micro avg) 0.796 2023-10-17 15:05:30,230 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:05:39,268 epoch 9 - iter 178/1786 - loss 0.01549311 - time (sec): 9.04 - samples/sec: 2757.67 - lr: 0.000011 - momentum: 0.000000 2023-10-17 15:05:48,446 epoch 9 - iter 356/1786 - loss 0.01455785 - time (sec): 18.21 - samples/sec: 2698.56 - lr: 0.000010 - momentum: 0.000000 2023-10-17 15:05:57,751 epoch 9 - iter 534/1786 - loss 0.01158104 - time (sec): 27.52 - samples/sec: 2723.46 - lr: 0.000009 - momentum: 0.000000 2023-10-17 15:06:06,429 epoch 9 - iter 712/1786 - loss 0.01064053 - time (sec): 36.20 - samples/sec: 2738.43 - lr: 0.000009 - momentum: 0.000000 2023-10-17 15:06:15,245 epoch 9 - iter 890/1786 - loss 0.01068597 - time (sec): 45.01 - samples/sec: 2731.24 - lr: 0.000008 - momentum: 0.000000 2023-10-17 15:06:24,206 epoch 9 - iter 1068/1786 - loss 0.01073553 - time (sec): 53.97 - samples/sec: 2750.68 - lr: 0.000008 - momentum: 0.000000 2023-10-17 15:06:33,658 epoch 9 - iter 1246/1786 - loss 0.01060456 - time (sec): 63.43 - samples/sec: 2747.29 - lr: 0.000007 - momentum: 0.000000 2023-10-17 15:06:42,635 epoch 9 - iter 1424/1786 - loss 0.01063471 - time (sec): 72.40 - samples/sec: 2754.79 - lr: 0.000007 - momentum: 0.000000 2023-10-17 15:06:51,504 epoch 9 - iter 1602/1786 - loss 0.01083306 - time (sec): 81.27 - samples/sec: 2743.16 - lr: 0.000006 - momentum: 0.000000 2023-10-17 15:07:00,815 epoch 9 - iter 1780/1786 - loss 0.01063372 - time (sec): 90.58 - samples/sec: 2737.25 - lr: 0.000006 - momentum: 0.000000 2023-10-17 15:07:01,104 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:07:01,104 EPOCH 9 done: loss 0.0107 - lr: 0.000006 2023-10-17 15:07:06,033 DEV : loss 0.25563544034957886 - f1-score (micro avg) 0.7975 2023-10-17 15:07:06,052 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:07:14,958 epoch 10 - iter 178/1786 - loss 0.00707783 - time (sec): 8.91 - samples/sec: 2888.08 - lr: 0.000005 - momentum: 0.000000 2023-10-17 15:07:23,959 epoch 10 - iter 356/1786 - loss 0.00883515 - time (sec): 17.91 - samples/sec: 2733.41 - lr: 0.000004 - momentum: 0.000000 2023-10-17 15:07:32,883 epoch 10 - iter 534/1786 - loss 0.00797908 - time (sec): 26.83 - samples/sec: 2774.80 - lr: 0.000004 - momentum: 0.000000 2023-10-17 15:07:41,548 epoch 10 - iter 712/1786 - loss 0.00720083 - time (sec): 35.50 - samples/sec: 2745.13 - lr: 0.000003 - momentum: 0.000000 2023-10-17 15:07:50,574 epoch 10 - iter 890/1786 - loss 0.00714750 - time (sec): 44.52 - samples/sec: 2750.51 - lr: 0.000003 - momentum: 0.000000 2023-10-17 15:07:59,491 epoch 10 - iter 1068/1786 - loss 0.00664834 - time (sec): 53.44 - samples/sec: 2731.24 - lr: 0.000002 - momentum: 0.000000 2023-10-17 15:08:08,680 epoch 10 - iter 1246/1786 - loss 0.00708464 - time (sec): 62.63 - samples/sec: 2764.16 - lr: 0.000002 - momentum: 0.000000 2023-10-17 15:08:17,473 epoch 10 - iter 1424/1786 - loss 0.00733589 - time (sec): 71.42 - samples/sec: 2762.64 - lr: 0.000001 - momentum: 0.000000 2023-10-17 15:08:26,511 epoch 10 - iter 1602/1786 - loss 0.00779833 - time (sec): 80.46 - samples/sec: 2780.02 - lr: 0.000001 - momentum: 0.000000 2023-10-17 15:08:35,379 epoch 10 - iter 1780/1786 - loss 0.00728986 - time (sec): 89.33 - samples/sec: 2776.90 - lr: 0.000000 - momentum: 0.000000 2023-10-17 15:08:35,656 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:08:35,657 EPOCH 10 done: loss 0.0073 - lr: 0.000000 2023-10-17 15:08:39,959 DEV : loss 0.2582792043685913 - f1-score (micro avg) 0.7978 2023-10-17 15:08:40,317 ---------------------------------------------------------------------------------------------------- 2023-10-17 15:08:40,319 Loading model from best epoch ... 2023-10-17 15:08:41,660 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-17 15:08:51,322 Results: - F-score (micro) 0.6992 - F-score (macro) 0.6347 - Accuracy 0.5516 By class: precision recall f1-score support LOC 0.6819 0.7087 0.6950 1095 PER 0.7733 0.7549 0.7640 1012 ORG 0.5791 0.5126 0.5438 357 HumanProd 0.4062 0.7879 0.5361 33 micro avg 0.6979 0.7004 0.6992 2497 macro avg 0.6101 0.6910 0.6347 2497 weighted avg 0.7006 0.7004 0.6993 2497 2023-10-17 15:08:51,323 ----------------------------------------------------------------------------------------------------