Training in progress, epoch 0

config.json

@@ -1,13 +1,26 @@
 {
-  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "_name_or_path": "apple/mobilevit-small",
   "architectures": [
-    "ViTForImageClassification"
+    "MobileViTForImageClassification"
+  ],
+  "aspp_dropout_prob": 0.1,
+  "aspp_out_channels": 256,
+  "atrous_rates": [
+    6,
+    12,
+    18
   ],
   "attention_probs_dropout_prob": 0.0,
-  "encoder_stride": 16,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.0,
-  "hidden_size": 768,
+  "classifier_dropout_prob": 0.1,
+  "conv_kernel_size": 3,
+  "expand_ratio": 4.0,
+  "hidden_act": "silu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_sizes": [
+    144,
+    192,
+    240
+  ],
   "id2label": {
     "0": "apple_pie",
     "1": "baby_back_ribs",
@@ -111,9 +124,8 @@
     "98": "tiramisu",
     "99": "tuna_tartare"
   },
-  "image_size": 224,
+  "image_size": 256,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
   "label2id": {
     "apple_pie": "0",
     "baby_back_ribs": "1",
@@ -217,14 +229,25 @@
     "tuna_tartare": "99",
     "waffles": "100"
   },
-  "layer_norm_eps": 1e-12,
-  "model_type": "vit",
-  "num_attention_heads": 12,
+  "layer_norm_eps": 1e-05,
+  "mlp_ratio": 2.0,
+  "model_type": "mobilevit",
+  "neck_hidden_sizes": [
+    16,
+    32,
+    64,
+    96,
+    128,
+    160,
+    640
+  ],
+  "num_attention_heads": 4,
   "num_channels": 3,
-  "num_hidden_layers": 12,
-  "patch_size": 16,
+  "output_stride": 32,
+  "patch_size": 2,
   "problem_type": "single_label_classification",
   "qkv_bias": true,
+  "semantic_loss_ignore_index": 255,
   "torch_dtype": "float32",
-  "transformers_version": "4.44.2"
+  "transformers_version": "4.45.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24239f9dfba75ec0d9728b80f904d848d4cd8808a915cb0ee5390e2debf21418
-size 343528508
+oid sha256:74b8804a2fa528e5d0a7ea1beb063130a5827cea203bf65b973fe4921d65e84e
+size 20105428

preprocessor_config.json

@@ -1,22 +1,17 @@
 {
-  "do_normalize": true,
+  "crop_size": {
+    "height": 256,
+    "width": 256
+  },
+  "do_center_crop": true,
+  "do_flip_channel_order": true,
+  "do_flip_channels": true,
   "do_rescale": true,
   "do_resize": true,
-  "image_mean": [
-    0.5,
-    0.5,
-    0.5
-  ],
-  "image_processor_type": "ViTImageProcessor",
-  "image_std": [
-    0.5,
-    0.5,
-    0.5
-  ],
+  "image_processor_type": "MobileViTImageProcessor",
   "resample": 2,
   "rescale_factor": 0.00392156862745098,
   "size": {
-    "height": 224,
-    "width": 224
+    "shortest_edge": 288
   }
 }

runs/Oct08_14-43-23_a6c7f189a935/events.out.tfevents.1728398606.a6c7f189a935.30.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e9e60144dbfb48067025d5ea8f2db53e9ff8ca5cbe996cac28cf361ee44ceb0
+size 11142

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c80955f678989fb3db81a51e8af2c201ca5c9f730b9d2032db97b45326c0463
-size 5176
+oid sha256:157af608c340abd797b8b02e18799e6e74ce1487a22455a7882bf00fe3b07c62
+size 5240