Training in progress, step 100

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "laion/CLIP-ViT-H-14-laion2B-s32B-b79K",
+  "_name_or_path": "google/vit-large-patch16-224",
   "architectures": [
     "ViTForImageClassification"
   ],
@@ -7,583 +7,68 @@
   "encoder_stride": 16,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "0",
     "1": "1",
-    "10": "107",
-    "100": "193",
-    "101": "194",
-    "102": "195",
-    "103": "196",
-    "104": "197",
-    "105": "198",
-    "106": "199",
-    "107": "2",
-    "108": "20",
-    "109": "201",
-    "11": "109",
-    "110": "202",
-    "111": "203",
-    "112": "204",
-    "113": "205",
-    "114": "206",
-    "115": "207",
-    "116": "208",
-    "117": "209",
-    "118": "21",
-    "119": "211",
-    "12": "11",
-    "120": "212",
-    "121": "213",
-    "122": "214",
-    "123": "215",
-    "124": "216",
-    "125": "217",
-    "126": "3",
-    "127": "34",
-    "128": "4",
-    "129": "44",
-    "13": "111",
-    "130": "45",
-    "131": "5",
-    "132": "56",
-    "133": "6",
-    "134": "7",
-    "135": "77",
-    "136": "78",
-    "137": "79",
-    "138": "8",
-    "139": "81",
-    "14": "112",
-    "140": "82",
-    "141": "83",
-    "142": "84",
-    "143": "85",
-    "144": "86",
-    "145": "87",
-    "146": "88",
-    "147": "89",
-    "148": "9",
-    "149": "90",
-    "15": "113",
-    "150": "91",
-    "151": "93",
-    "152": "94",
-    "153": "95",
-    "154": "96",
-    "155": "97",
-    "156": "98",
-    "157": "99",
-    "158": "Acura",
-    "159": "Amkette",
-    "16": "114",
-    "160": "Apple",
-    "161": "BACCARAT",
-    "162": "BaiDu",
-    "163": "Batman",
-    "164": "Dell",
-    "165": "Domino's Pizza",
-    "166": "Epson",
-    "167": "Hewlett-Packard",
-    "168": "IBM",
-    "169": "John Deere",
-    "17": "116",
-    "170": "KFC",
-    "171": "Kodak",
-    "172": "Lamborghini",
-    "173": "Lenovo",
-    "174": "Mazda",
-    "175": "McDonald's",
-    "176": "Mercedes-Benz",
-    "177": "Nestle",
-    "178": "Nintendo",
-    "179": "Nissan",
-    "18": "117",
-    "180": "Ovaltine",
-    "181": "Pepsi Max",
-    "182": "Pokemon",
-    "183": "Red Bull",
-    "184": "Star Wars",
-    "185": "The Pizza Company",
-    "186": "Tim Hortons",
-    "187": "b",
-    "188": "d",
-    "189": "f",
-    "19": "118",
-    "190": "h",
-    "191": "hp",
-    "192": "huawei",
-    "193": "pepsi",
-    "194": "subaru",
+    "10": "18",
+    "11": "19",
+    "12": "2",
+    "13": "20",
+    "14": "21",
+    "15": "22",
+    "16": "3",
+    "17": "4",
+    "18": "5",
+    "19": "6",
     "2": "10",
-    "20": "119",
-    "21": "12",
-    "22": "120",
-    "23": "121",
-    "24": "122",
-    "25": "123",
-    "26": "124",
-    "27": "126",
-    "28": "127",
-    "29": "128",
-    "3": "100",
-    "30": "129",
-    "31": "13",
-    "32": "130",
-    "33": "131",
-    "34": "132",
-    "35": "133",
-    "36": "134",
-    "37": "135",
-    "38": "136",
-    "39": "137",
-    "4": "101",
-    "40": "138",
-    "41": "139",
-    "42": "14",
-    "43": "140",
-    "44": "141",
-    "45": "142",
-    "46": "143",
-    "47": "144",
-    "48": "145",
-    "49": "146",
-    "5": "102",
-    "50": "147",
-    "51": "149",
-    "52": "15",
-    "53": "150",
-    "54": "151",
-    "55": "152",
-    "56": "153",
-    "57": "154",
-    "58": "155",
-    "59": "156",
-    "6": "103",
-    "60": "157",
-    "61": "158",
-    "62": "159",
-    "63": "16",
-    "64": "160",
-    "65": "161",
-    "66": "162",
-    "67": "163",
-    "68": "164",
-    "69": "165",
-    "7": "104",
-    "70": "166",
-    "71": "167",
-    "72": "168",
-    "73": "169",
-    "74": "17",
-    "75": "170",
-    "76": "171",
-    "77": "172",
-    "78": "173",
-    "79": "174",
-    "8": "105",
-    "80": "175",
-    "81": "176",
-    "82": "177",
-    "83": "178",
-    "84": "179",
-    "85": "18",
-    "86": "180",
-    "87": "181",
-    "88": "182",
-    "89": "183",
-    "9": "106",
-    "90": "184",
-    "91": "185",
-    "92": "186",
-    "93": "187",
-    "94": "188",
-    "95": "189",
-    "96": "19",
-    "97": "190",
-    "98": "191",
-    "99": "192"
+    "20": "7",
+    "21": "8",
+    "22": "9",
+    "3": "11",
+    "4": "12",
+    "5": "13",
+    "6": "14",
+    "7": "15",
+    "8": "16",
+    "9": "17"
   },
   "image_size": 224,
-  "initializer_factor": 1.0,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "0": "0",
     "1": "1",
     "10": "2",
-    "100": "3",
-    "101": "4",
-    "102": "5",
-    "103": "6",
-    "104": "7",
-    "105": "8",
-    "106": "9",
-    "107": "10",
-    "109": "11",
-    "11": "12",
-    "111": "13",
-    "112": "14",
-    "113": "15",
-    "114": "16",
-    "116": "17",
-    "117": "18",
-    "118": "19",
-    "119": "20",
-    "12": "21",
-    "120": "22",
-    "121": "23",
-    "122": "24",
-    "123": "25",
-    "124": "26",
-    "126": "27",
-    "127": "28",
-    "128": "29",
-    "129": "30",
-    "13": "31",
-    "130": "32",
-    "131": "33",
-    "132": "34",
-    "133": "35",
-    "134": "36",
-    "135": "37",
-    "136": "38",
-    "137": "39",
-    "138": "40",
-    "139": "41",
-    "14": "42",
-    "140": "43",
-    "141": "44",
-    "142": "45",
-    "143": "46",
-    "144": "47",
-    "145": "48",
-    "146": "49",
-    "147": "50",
-    "149": "51",
-    "15": "52",
-    "150": "53",
-    "151": "54",
-    "152": "55",
-    "153": "56",
-    "154": "57",
-    "155": "58",
-    "156": "59",
-    "157": "60",
-    "158": "61",
-    "159": "62",
-    "16": "63",
-    "160": "64",
-    "161": "65",
-    "162": "66",
-    "163": "67",
-    "164": "68",
-    "165": "69",
-    "166": "70",
-    "167": "71",
-    "168": "72",
-    "169": "73",
-    "17": "74",
-    "170": "75",
-    "171": "76",
-    "172": "77",
-    "173": "78",
-    "174": "79",
-    "175": "80",
-    "176": "81",
-    "177": "82",
-    "178": "83",
-    "179": "84",
-    "18": "85",
-    "180": "86",
-    "181": "87",
-    "182": "88",
-    "183": "89",
-    "184": "90",
-    "185": "91",
-    "186": "92",
-    "187": "93",
-    "188": "94",
-    "189": "95",
-    "19": "96",
-    "190": "97",
-    "191": "98",
-    "192": "99",
-    "193": "100",
-    "194": "101",
-    "195": "102",
-    "196": "103",
-    "197": "104",
-    "198": "105",
-    "199": "106",
-    "2": "107",
-    "20": "108",
-    "201": "109",
-    "202": "110",
-    "203": "111",
-    "204": "112",
-    "205": "113",
-    "206": "114",
-    "207": "115",
-    "208": "116",
-    "209": "117",
-    "21": "118",
-    "211": "119",
-    "212": "120",
-    "213": "121",
-    "214": "122",
-    "215": "123",
-    "216": "124",
-    "217": "125",
-    "3": "126",
-    "34": "127",
-    "4": "128",
-    "44": "129",
-    "45": "130",
-    "5": "131",
-    "56": "132",
-    "6": "133",
-    "7": "134",
-    "77": "135",
-    "78": "136",
-    "79": "137",
-    "8": "138",
-    "81": "139",
-    "82": "140",
-    "83": "141",
-    "84": "142",
-    "85": "143",
-    "86": "144",
-    "87": "145",
-    "88": "146",
-    "89": "147",
-    "9": "148",
-    "90": "149",
-    "91": "150",
-    "93": "151",
-    "94": "152",
-    "95": "153",
-    "96": "154",
-    "97": "155",
-    "98": "156",
-    "99": "157",
-    "Acura": "158",
-    "Amkette": "159",
-    "Apple": "160",
-    "BACCARAT": "161",
-    "BaiDu": "162",
-    "Batman": "163",
-    "Dell": "164",
-    "Domino's Pizza": "165",
-    "Epson": "166",
-    "Hewlett-Packard": "167",
-    "IBM": "168",
-    "John Deere": "169",
-    "KFC": "170",
-    "Kodak": "171",
-    "Lamborghini": "172",
-    "Lenovo": "173",
-    "Mazda": "174",
-    "McDonald's": "175",
-    "Mercedes-Benz": "176",
-    "Nestle": "177",
-    "Nintendo": "178",
-    "Nissan": "179",
-    "Ovaltine": "180",
-    "Pepsi Max": "181",
-    "Pokemon": "182",
-    "Red Bull": "183",
-    "Star Wars": "184",
-    "The Pizza Company": "185",
-    "Tim Hortons": "186",
-    "b": "187",
-    "d": "188",
-    "f": "189",
-    "h": "190",
-    "hp": "191",
-    "huawei": "192",
-    "pepsi": "193",
-    "subaru": "194"
+    "11": "3",
+    "12": "4",
+    "13": "5",
+    "14": "6",
+    "15": "7",
+    "16": "8",
+    "17": "9",
+    "18": "10",
+    "19": "11",
+    "2": "12",
+    "20": "13",
+    "21": "14",
+    "22": "15",
+    "3": "16",
+    "4": "17",
+    "5": "18",
+    "6": "19",
+    "7": "20",
+    "8": "21",
+    "9": "22"
   },
   "layer_norm_eps": 1e-12,
-  "logit_scale_init_value": 2.6592,
   "model_type": "vit",
-  "num_attention_heads": 12,
+  "num_attention_heads": 16,
   "num_channels": 3,
-  "num_hidden_layers": 12,
+  "num_hidden_layers": 24,
   "patch_size": 16,
   "problem_type": "single_label_classification",
-  "projection_dim": 1024,
   "qkv_bias": true,
-  "text_config": {
-    "_name_or_path": "",
-    "add_cross_attention": false,
-    "architectures": null,
-    "attention_dropout": 0.0,
-    "bad_words_ids": null,
-    "bos_token_id": 0,
-    "chunk_size_feed_forward": 0,
-    "cross_attention_hidden_size": null,
-    "decoder_start_token_id": null,
-    "diversity_penalty": 0.0,
-    "do_sample": false,
-    "dropout": 0.0,
-    "early_stopping": false,
-    "encoder_no_repeat_ngram_size": 0,
-    "eos_token_id": 2,
-    "exponential_decay_length_penalty": null,
-    "finetuning_task": null,
-    "forced_bos_token_id": null,
-    "forced_eos_token_id": null,
-    "hidden_act": "gelu",
-    "hidden_size": 1024,
-    "id2label": {
-      "0": "LABEL_0",
-      "1": "LABEL_1"
-    },
-    "initializer_factor": 1.0,
-    "initializer_range": 0.02,
-    "intermediate_size": 4096,
-    "is_decoder": false,
-    "is_encoder_decoder": false,
-    "label2id": {
-      "LABEL_0": 0,
-      "LABEL_1": 1
-    },
-    "layer_norm_eps": 1e-05,
-    "length_penalty": 1.0,
-    "max_length": 20,
-    "max_position_embeddings": 77,
-    "min_length": 0,
-    "model_type": "clip_text_model",
-    "no_repeat_ngram_size": 0,
-    "num_attention_heads": 16,
-    "num_beam_groups": 1,
-    "num_beams": 1,
-    "num_hidden_layers": 24,
-    "num_return_sequences": 1,
-    "output_attentions": false,
-    "output_hidden_states": false,
-    "output_scores": false,
-    "pad_token_id": 1,
-    "prefix": null,
-    "problem_type": null,
-    "projection_dim": 1024,
-    "pruned_heads": {},
-    "remove_invalid_values": false,
-    "repetition_penalty": 1.0,
-    "return_dict": true,
-    "return_dict_in_generate": false,
-    "sep_token_id": null,
-    "task_specific_params": null,
-    "temperature": 1.0,
-    "tf_legacy_loss": false,
-    "tie_encoder_decoder": false,
-    "tie_word_embeddings": true,
-    "tokenizer_class": null,
-    "top_k": 50,
-    "top_p": 1.0,
-    "torch_dtype": null,
-    "torchscript": false,
-    "transformers_version": "4.21.3",
-    "typical_p": 1.0,
-    "use_bfloat16": false,
-    "vocab_size": 49408
-  },
-  "text_config_dict": {
-    "hidden_act": "gelu",
-    "hidden_size": 1024,
-    "intermediate_size": 4096,
-    "num_attention_heads": 16,
-    "num_hidden_layers": 24,
-    "projection_dim": 1024
-  },
   "torch_dtype": "float32",
-  "transformers_version": "4.38.0.dev0",
-  "vision_config": {
-    "_name_or_path": "",
-    "add_cross_attention": false,
-    "architectures": null,
-    "attention_dropout": 0.0,
-    "bad_words_ids": null,
-    "bos_token_id": null,
-    "chunk_size_feed_forward": 0,
-    "cross_attention_hidden_size": null,
-    "decoder_start_token_id": null,
-    "diversity_penalty": 0.0,
-    "do_sample": false,
-    "dropout": 0.0,
-    "early_stopping": false,
-    "encoder_no_repeat_ngram_size": 0,
-    "eos_token_id": null,
-    "exponential_decay_length_penalty": null,
-    "finetuning_task": null,
-    "forced_bos_token_id": null,
-    "forced_eos_token_id": null,
-    "hidden_act": "gelu",
-    "hidden_size": 1280,
-    "id2label": {
-      "0": "LABEL_0",
-      "1": "LABEL_1"
-    },
-    "image_size": 224,
-    "initializer_factor": 1.0,
-    "initializer_range": 0.02,
-    "intermediate_size": 5120,
-    "is_decoder": false,
-    "is_encoder_decoder": false,
-    "label2id": {
-      "LABEL_0": 0,
-      "LABEL_1": 1
-    },
-    "layer_norm_eps": 1e-05,
-    "length_penalty": 1.0,
-    "max_length": 20,
-    "min_length": 0,
-    "model_type": "clip_vision_model",
-    "no_repeat_ngram_size": 0,
-    "num_attention_heads": 16,
-    "num_beam_groups": 1,
-    "num_beams": 1,
-    "num_channels": 3,
-    "num_hidden_layers": 32,
-    "num_return_sequences": 1,
-    "output_attentions": false,
-    "output_hidden_states": false,
-    "output_scores": false,
-    "pad_token_id": null,
-    "patch_size": 14,
-    "prefix": null,
-    "problem_type": null,
-    "projection_dim": 1024,
-    "pruned_heads": {},
-    "remove_invalid_values": false,
-    "repetition_penalty": 1.0,
-    "return_dict": true,
-    "return_dict_in_generate": false,
-    "sep_token_id": null,
-    "task_specific_params": null,
-    "temperature": 1.0,
-    "tf_legacy_loss": false,
-    "tie_encoder_decoder": false,
-    "tie_word_embeddings": true,
-    "tokenizer_class": null,
-    "top_k": 50,
-    "top_p": 1.0,
-    "torch_dtype": null,
-    "torchscript": false,
-    "transformers_version": "4.21.3",
-    "typical_p": 1.0,
-    "use_bfloat16": false
-  },
-  "vision_config_dict": {
-    "hidden_act": "gelu",
-    "hidden_size": 1280,
-    "intermediate_size": 5120,
-    "num_attention_heads": 16,
-    "num_hidden_layers": 32,
-    "patch_size": 14,
-    "projection_dim": 1024
-  }
+  "transformers_version": "4.38.0.dev0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26ad21aab0ab20c9546c7ebcf585b4deceb7113ef35f75bd8aa22d597aa84a67
-size 343817652
+oid sha256:6112c0035d58af0328bd538cf65f89109a4f6de848c4d09ead372a13a54b12e5
+size 1213347372

preprocessor_config.json

@@ -1,21 +1,19 @@
 {
-  "crop_size": 224,
-  "do_center_crop": true,
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,
   "image_mean": [
-    0.48145466,
-    0.4578275,
-    0.40821073
+    0.5,
+    0.5,
+    0.5
   ],
   "image_processor_type": "ViTImageProcessor",
   "image_std": [
-    0.26862954,
-    0.26130258,
-    0.27577711
+    0.5,
+    0.5,
+    0.5
   ],
-  "resample": 3,
+  "resample": 2,
   "rescale_factor": 0.00392156862745098,
   "size": {
     "height": 224,

runs/Feb03_03-26-44_c46e18eeb47a/events.out.tfevents.1706930809.c46e18eeb47a.1017.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b83bcb96c4f1bbefcdde33d45eb74bf1bcb3d8b9b7a4efbde6b9157c5889682
+size 4184

runs/Feb03_03-29-55_c46e18eeb47a/events.out.tfevents.1706930999.c46e18eeb47a.3380.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f185c76a4ee858120b4f1328822c648935e8df8c9a38598c942d4b717ef22690
+size 4184

runs/Feb03_03-32-21_c46e18eeb47a/events.out.tfevents.1706931146.c46e18eeb47a.4146.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31566154a3bea117157d0424814227548c9460797126e4ee647ff63db9a28065
+size 7169

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49b3d23b71225000d40756516e1457f0faf320b3a3f244955779dffbe88c104b
+oid sha256:a0df834a064202344629829fe948dbefa8462af544011692fa35ffcc67a80fb5
 size 4728