Training in progress, epoch 5

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "_name_or_path": "google/vit-large-patch32-224-in21k",
   "architectures": [
     "ViTForImageClassification"
   ],
@@ -7,7 +7,7 @@
   "encoder_stride": 16,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "anger",
     "1": "contempt",
@@ -20,7 +20,7 @@
   },
   "image_size": 224,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "anger": "0",
     "contempt": "1",
@@ -33,10 +33,10 @@
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",
-  "num_attention_heads": 12,
+  "num_attention_heads": 16,
   "num_channels": 3,
-  "num_hidden_layers": 12,
-  "patch_size": 16,
+  "num_hidden_layers": 24,
+  "patch_size": 32,
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8f47c673d15374af09801f99430b0357016b5e070c072cfab68096f8e6046b5
-size 343284397
+oid sha256:cca924c73dd41fe79ce4beb6bef89b97729991dc73fea2e543580b35f5acb4a7
+size 1222208301

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ed4c4fa39c95401442ce26084427ddbbfefe7ff9951d2ae5d87abf7291eed10
+oid sha256:90278cd54c7b7d640213b12c867e36bae82f168938accac4348a7f947b5fef48
 size 4027