Training in progress, epoch 0

config.json

@@ -1,13 +1,18 @@
 {
-  "_name_or_path": "WinKawaks/vit-tiny-patch16-224",
+  "_name_or_path": "microsoft/beit-base-patch16-224-pt22k-ft22k",
+  "add_fpn": false,
   "architectures": [
-    "ViTForImageClassification"
+    "BeitForImageClassification"
   ],
   "attention_probs_dropout_prob": 0.0,
-  "encoder_stride": 16,
+  "auxiliary_channels": 256,
+  "auxiliary_concat_input": false,
+  "auxiliary_loss_weight": 0.4,
+  "auxiliary_num_convs": 1,
+  "drop_path_rate": 0.1,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
-  "hidden_size": 192,
+  "hidden_size": 768,
   "id2label": {
     "0": "burn through",
     "1": "contamination",
@@ -18,7 +23,7 @@
   },
   "image_size": 224,
   "initializer_range": 0.02,
-  "intermediate_size": 768,
+  "intermediate_size": 3072,
   "label2id": {
     "burn through": "0",
     "contamination": "1",
@@ -28,13 +33,49 @@
     "misalignment": "5"
   },
   "layer_norm_eps": 1e-12,
-  "model_type": "vit",
-  "num_attention_heads": 3,
+  "layer_scale_init_value": 0.1,
+  "model_type": "beit",
+  "num_attention_heads": 12,
   "num_channels": 3,
   "num_hidden_layers": 12,
+  "out_features": [
+    "stage12"
+  ],
+  "out_indices": [
+    12
+  ],
   "patch_size": 16,
+  "pool_scales": [
+    1,
+    2,
+    3,
+    6
+  ],
   "problem_type": "single_label_classification",
-  "qkv_bias": true,
+  "reshape_hidden_states": true,
+  "semantic_loss_ignore_index": 255,
+  "stage_names": [
+    "stem",
+    "stage1",
+    "stage2",
+    "stage3",
+    "stage4",
+    "stage5",
+    "stage6",
+    "stage7",
+    "stage8",
+    "stage9",
+    "stage10",
+    "stage11",
+    "stage12"
+  ],
   "torch_dtype": "float32",
-  "transformers_version": "4.45.1"
+  "transformers_version": "4.45.1",
+  "use_absolute_position_embeddings": false,
+  "use_auxiliary_head": true,
+  "use_mask_token": false,
+  "use_mean_pooling": true,
+  "use_relative_position_bias": true,
+  "use_shared_relative_position_bias": false,
+  "vocab_size": 8192
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76bbb7cf68b217d04dead23289e75056c7880b4471b67861d430afa16263321b
-size 22125008
+oid sha256:1ea923c624c757dfc857d56449e4970d499e785ed119e2869c2f0be43340d1aa
+size 343092632

preprocessor_config.json

@@ -1,5 +1,11 @@
 {
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": false,
   "do_normalize": true,
+  "do_reduce_labels": false,
   "do_rescale": true,
   "do_resize": true,
   "image_mean": [
@@ -7,7 +13,7 @@
     0.5,
     0.5
   ],
-  "image_processor_type": "ViTImageProcessor",
+  "image_processor_type": "BeitImageProcessor",
   "image_std": [
     0.5,
     0.5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23a9e5252b9ea82a68e99ab0b2ddc2fcbfbbeda2691d7c705853e741f624b392
-size 5240
+oid sha256:d6966e26d0010efbb6d1d47ae70d7608df5e22b51e4a7629d7763af27f1f3624
+size 5176