small-model-all-categories

config.json

@@ -1,57 +1,55 @@
 {
+  "_name_or_path": "../Models/dual_encoder_pretrained",
   "architectures": [
     "VisionTextDualEncoderModel"
   ],
   "logit_scale_init_value": 2.6592,
   "model_type": "vision-text-dual-encoder",
-  "projection_dim": 512,
+  "projection_dim": 256,
   "text_config": {
-    "_name_or_path": "",
+    "_name_or_path": "lassl/roberta-ko-small",
     "add_cross_attention": false,
     "architectures": [
-      "BertForMaskedLM"
+      "RobertaForMaskedLM"
     ],
     "attention_probs_dropout_prob": 0.1,
     "bad_words_ids": null,
-    "bos_token_id": null,
+    "bos_token_id": 0,
     "chunk_size_feed_forward": 0,
     "classifier_dropout": null,
     "cross_attention_hidden_size": null,
     "decoder_start_token_id": null,
-    "directionality": "bidi",
     "diversity_penalty": 0.0,
     "do_sample": false,
     "early_stopping": false,
-    "embedding_size": 768,
     "encoder_no_repeat_ngram_size": 0,
-    "eos_token_id": null,
+    "eos_token_id": 2,
     "finetuning_task": null,
     "forced_bos_token_id": null,
     "forced_eos_token_id": null,
-    "gradient_checkpointing": false,
     "hidden_act": "gelu",
     "hidden_dropout_prob": 0.1,
-    "hidden_size": 768,
+    "hidden_size": 256,
     "id2label": {
       "0": "LABEL_0",
       "1": "LABEL_1"
     },
     "initializer_range": 0.02,
-    "intermediate_size": 3072,
+    "intermediate_size": 1024,
     "is_decoder": false,
     "is_encoder_decoder": false,
     "label2id": {
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layer_norm_eps": 1e-12,
+    "layer_norm_eps": 1e-05,
     "length_penalty": 1.0,
     "max_length": 20,
     "max_position_embeddings": 512,
     "min_length": 0,
-    "model_type": "bert",
+    "model_type": "roberta",
     "no_repeat_ngram_size": 0,
-    "num_attention_heads": 12,
+    "num_attention_heads": 4,
     "num_beam_groups": 1,
     "num_beams": 1,
     "num_hidden_layers": 12,
@@ -59,12 +57,7 @@
     "output_attentions": false,
     "output_hidden_states": false,
     "output_scores": false,
-    "pad_token_id": 0,
-    "pooler_fc_size": 768,
-    "pooler_num_attention_heads": 12,
-    "pooler_num_fc_layers": 3,
-    "pooler_size_per_head": 128,
-    "pooler_type": "first_token_transform",
+    "pad_token_id": 1,
     "position_embedding_type": "absolute",
     "prefix": null,
     "problem_type": null,
@@ -81,22 +74,24 @@
     "tokenizer_class": null,
     "top_k": 50,
     "top_p": 1.0,
-    "torch_dtype": null,
+    "torch_dtype": "float32",
     "torchscript": false,
     "transformers_version": "4.17.0",
-    "type_vocab_size": 2,
+    "type_vocab_size": 1,
     "typical_p": 1.0,
     "use_bfloat16": false,
     "use_cache": true,
-    "vocab_size": 42000
+    "vocab_size": 51200
   },
   "torch_dtype": "float32",
   "transformers_version": null,
   "vision_config": {
-    "_name_or_path": "",
+    "_name_or_path": "facebook/dino-vits8",
     "add_cross_attention": false,
-    "architectures": null,
-    "attention_dropout": 0.0,
+    "architectures": [
+      "ViTModel"
+    ],
+    "attention_probs_dropout_prob": 0.0,
     "bad_words_ids": null,
     "bos_token_id": null,
     "chunk_size_feed_forward": 0,
@@ -104,49 +99,50 @@
     "decoder_start_token_id": null,
     "diversity_penalty": 0.0,
     "do_sample": false,
-    "dropout": 0.0,
     "early_stopping": false,
     "encoder_no_repeat_ngram_size": 0,
+    "encoder_stride": 16,
     "eos_token_id": null,
     "finetuning_task": null,
     "forced_bos_token_id": null,
     "forced_eos_token_id": null,
-    "gradient_checkpointing": false,
-    "hidden_act": "quick_gelu",
-    "hidden_size": 768,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 384,
     "id2label": {
       "0": "LABEL_0",
       "1": "LABEL_1"
     },
     "image_size": 224,
-    "initializer_factor": 1.0,
     "initializer_range": 0.02,
-    "intermediate_size": 3072,
+    "intermediate_size": 1536,
     "is_decoder": false,
     "is_encoder_decoder": false,
     "label2id": {
       "LABEL_0": 0,
       "LABEL_1": 1
     },
-    "layer_norm_eps": 1e-05,
+    "layer_norm_eps": 1e-12,
     "length_penalty": 1.0,
     "max_length": 20,
     "min_length": 0,
-    "model_type": "clip_vision_model",
+    "model_type": "vit",
     "no_repeat_ngram_size": 0,
-    "num_attention_heads": 12,
+    "num_attention_heads": 6,
     "num_beam_groups": 1,
     "num_beams": 1,
+    "num_channels": 3,
     "num_hidden_layers": 12,
     "num_return_sequences": 1,
     "output_attentions": false,
     "output_hidden_states": false,
     "output_scores": false,
     "pad_token_id": null,
-    "patch_size": 32,
+    "patch_size": 8,
     "prefix": null,
     "problem_type": null,
     "pruned_heads": {},
+    "qkv_bias": true,
     "remove_invalid_values": false,
     "repetition_penalty": 1.0,
     "return_dict": true,
@@ -159,7 +155,7 @@
     "tokenizer_class": null,
     "top_k": 50,
     "top_p": 1.0,
-    "torch_dtype": null,
+    "torch_dtype": "float32",
     "torchscript": false,
     "transformers_version": "4.17.0",
     "typical_p": 1.0,

preprocessor_config.json

@@ -1,20 +1,18 @@
 {
-  "crop_size": 224,
-  "do_center_crop": true,
   "do_normalize": true,
-  "do_resize": "/workspace/Models/koclip_v_0_0_1",
-  "feature_extractor_type": "CLIPFeatureExtractor",
+  "do_resize": true,
+  "feature_extractor_type": "ViTFeatureExtractor",
   "image_mean": [
-    0.48145466,
-    0.4578275,
-    0.40821073
+    0.485,
+    0.456,
+    0.406
   ],
   "image_std": [
-    0.26862954,
-    0.26130258,
-    0.27577711
+    0.229,
+    0.224,
+    0.225
   ],
   "processor_class": "VisionTextDualEncoderProcessor",
-  "resample": 3,
+  "resample": 2,
   "size": 224
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:962741e82185d4f00307ba2dfbc87c3daae89604235580dba10036107d1d4181
-size 826324123
+oid sha256:ef1fe083f3fdf763d528e0872ce828b6ccc4be872598a9575a38413572095587
+size 179228443

special_tokens_map.json

@@ -1 +1 @@
-{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "/workspace/Models/koclip_v_0_0_1", "tokenizer_class": "BertTokenizer", "processor_class": "VisionTextDualEncoderProcessor"}
+{"errors": "replace", "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false, "__type": "AddedToken"}, "add_prefix_space": false, "trim_offsets": true, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "../Models/dual_encoder_pretrained", "processor_class": "VisionTextDualEncoderProcessor", "tokenizer_class": "RobertaTokenizer"}