Training in progress, step 200

config.json

@@ -0,0 +1,123 @@
+{
+  "_name_or_path": "dandelin/vilt-b32-finetuned-vqa",
+  "architectures": [
+    "ViltForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Women",
+    "1": "Trousers",
+    "2": "Men",
+    "3": "Grey",
+    "4": "Casual Shoes",
+    "5": "Sunglasses",
+    "6": "Black",
+    "7": "Dresses",
+    "8": "Tshirts",
+    "9": "Multi",
+    "10": "Shorts",
+    "11": "Jeans",
+    "12": "Yellow",
+    "13": "Orange",
+    "14": "Sports",
+    "15": "Sandals",
+    "16": "Blue",
+    "17": "Formal Shoes",
+    "18": "Sports Shoes",
+    "19": "Winter",
+    "20": "Socks",
+    "21": "Beige",
+    "22": "Summer",
+    "23": "Navy Blue",
+    "24": "Formal",
+    "25": "Silver",
+    "26": "Green",
+    "27": "Pink",
+    "28": "Shirts",
+    "29": "Casual",
+    "30": "Unisex",
+    "31": "Cream",
+    "32": "Boys",
+    "33": "Tops",
+    "34": "Handbags",
+    "35": "Peach",
+    "36": "Flip Flops",
+    "37": "Track Pants",
+    "38": "Brown",
+    "39": "Skirts",
+    "40": "Purple",
+    "41": "Fall",
+    "42": "White",
+    "43": "Waistcoat",
+    "44": "Caps"
+  },
+  "image_size": 384,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Beige": 21,
+    "Black": 6,
+    "Blue": 16,
+    "Boys": 32,
+    "Brown": 38,
+    "Caps": 44,
+    "Casual": 29,
+    "Casual Shoes": 4,
+    "Cream": 31,
+    "Dresses": 7,
+    "Fall": 41,
+    "Flip Flops": 36,
+    "Formal": 24,
+    "Formal Shoes": 17,
+    "Green": 26,
+    "Grey": 3,
+    "Handbags": 34,
+    "Jeans": 11,
+    "Men": 2,
+    "Multi": 9,
+    "Navy Blue": 23,
+    "Orange": 13,
+    "Peach": 35,
+    "Pink": 27,
+    "Purple": 40,
+    "Sandals": 15,
+    "Shirts": 28,
+    "Shorts": 10,
+    "Silver": 25,
+    "Skirts": 39,
+    "Socks": 20,
+    "Sports": 14,
+    "Sports Shoes": 18,
+    "Summer": 22,
+    "Sunglasses": 5,
+    "Tops": 33,
+    "Track Pants": 37,
+    "Trousers": 1,
+    "Tshirts": 8,
+    "Unisex": 30,
+    "Waistcoat": 43,
+    "White": 42,
+    "Winter": 19,
+    "Women": 0,
+    "Yellow": 12
+  },
+  "layer_norm_eps": 1e-12,
+  "max_image_length": -1,
+  "max_position_embeddings": 40,
+  "modality_type_vocab_size": 2,
+  "model_type": "vilt",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "num_images": -1,
+  "patch_size": 32,
+  "qkv_bias": true,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.4",
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55a4e56649bb1900d1f8349fa7af82458e3f0db7e2c74afc7f5f31057133f8f9
+size 451418532

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViltImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "ViltProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 384
+  },
+  "size_divisor": 32
+}

runs/Jul25_12-34-34_49ff2a879370/events.out.tfevents.1721910878.49ff2a879370.1320.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60fb1ef9af94dfa3dc042d3b1f4c95f82d4d1c56ae6945ce905e01046bef3f95
+size 7579

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 40,
+  "pad_token": "[PAD]",
+  "processor_class": "ViltProcessor",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3375688ae7eaa05adb8b70c8f0b025d7b19b5366fc7d2d3d0f0cf80a373e71b7
+size 5176