clip-italian-final / config.json
1
{
2
  "architectures": [
3
    "HybridCLIP"
4
  ],
5
  "freeze_backbones": true,
6
  "initializer_factor": 1.0,
7
  "model_type": "hybrid-clip",
8
  "projection_dim": 512,
9
  "seed": 42,
10
  "text_config": {
11
    "_name_or_path": "",
12
    "add_cross_attention": false,
13
    "architectures": [
14
      "BertForMaskedLM"
15
    ],
16
    "attention_probs_dropout_prob": 0.1,
17
    "bad_words_ids": null,
18
    "bos_token_id": null,
19
    "chunk_size_feed_forward": 0,
20
    "decoder_start_token_id": null,
21
    "diversity_penalty": 0.0,
22
    "do_sample": false,
23
    "early_stopping": false,
24
    "encoder_no_repeat_ngram_size": 0,
25
    "eos_token_id": null,
26
    "finetuning_task": null,
27
    "forced_bos_token_id": null,
28
    "forced_eos_token_id": null,
29
    "gradient_checkpointing": false,
30
    "hidden_act": "gelu",
31
    "hidden_dropout_prob": 0.1,
32
    "hidden_size": 768,
33
    "id2label": {
34
      "0": "LABEL_0",
35
      "1": "LABEL_1"
36
    },
37
    "initializer_range": 0.02,
38
    "intermediate_size": 3072,
39
    "is_decoder": false,
40
    "is_encoder_decoder": false,
41
    "label2id": {
42
      "LABEL_0": 0,
43
      "LABEL_1": 1
44
    },
45
    "layer_norm_eps": 1e-12,
46
    "length_penalty": 1.0,
47
    "max_length": 20,
48
    "max_position_embeddings": 512,
49
    "min_length": 0,
50
    "model_type": "bert",
51
    "no_repeat_ngram_size": 0,
52
    "num_attention_heads": 12,
53
    "num_beam_groups": 1,
54
    "num_beams": 1,
55
    "num_hidden_layers": 12,
56
    "num_return_sequences": 1,
57
    "output_attentions": false,
58
    "output_hidden_states": false,
59
    "output_scores": false,
60
    "pad_token_id": 0,
61
    "position_embedding_type": "absolute",
62
    "prefix": null,
63
    "problem_type": null,
64
    "pruned_heads": {},
65
    "remove_invalid_values": false,
66
    "repetition_penalty": 1.0,
67
    "return_dict": true,
68
    "return_dict_in_generate": false,
69
    "sep_token_id": null,
70
    "task_specific_params": null,
71
    "temperature": 1.0,
72
    "tie_encoder_decoder": false,
73
    "tie_word_embeddings": true,
74
    "tokenizer_class": null,
75
    "top_k": 50,
76
    "top_p": 1.0,
77
    "torch_dtype": null,
78
    "torchscript": false,
79
    "transformers_version": "4.9.0.dev0",
80
    "type_vocab_size": 2,
81
    "use_bfloat16": false,
82
    "use_cache": true,
83
    "vocab_size": 32102
84
  },
85
  "transformers_version": null,
86
  "vision_config": {
87
    "_name_or_path": "",
88
    "add_cross_attention": false,
89
    "architectures": null,
90
    "attention_dropout": 0.0,
91
    "bad_words_ids": null,
92
    "bos_token_id": null,
93
    "chunk_size_feed_forward": 0,
94
    "decoder_start_token_id": null,
95
    "diversity_penalty": 0.0,
96
    "do_sample": false,
97
    "dropout": 0.0,
98
    "early_stopping": false,
99
    "encoder_no_repeat_ngram_size": 0,
100
    "eos_token_id": null,
101
    "finetuning_task": null,
102
    "forced_bos_token_id": null,
103
    "forced_eos_token_id": null,
104
    "gradient_checkpointing": false,
105
    "hidden_act": "quick_gelu",
106
    "hidden_size": 768,
107
    "id2label": {
108
      "0": "LABEL_0",
109
      "1": "LABEL_1"
110
    },
111
    "image_size": 224,
112
    "initializer_factor": 1.0,
113
    "initializer_range": 0.02,
114
    "intermediate_size": 3072,
115
    "is_decoder": false,
116
    "is_encoder_decoder": false,
117
    "label2id": {
118
      "LABEL_0": 0,
119
      "LABEL_1": 1
120
    },
121
    "layer_norm_eps": 1e-05,
122
    "length_penalty": 1.0,
123
    "max_length": 20,
124
    "min_length": 0,
125
    "model_type": "clip_vision_model",
126
    "no_repeat_ngram_size": 0,
127
    "num_attention_heads": 12,
128
    "num_beam_groups": 1,
129
    "num_beams": 1,
130
    "num_hidden_layers": 12,
131
    "num_return_sequences": 1,
132
    "output_attentions": false,
133
    "output_hidden_states": false,
134
    "output_scores": false,
135
    "pad_token_id": null,
136
    "patch_size": 32,
137
    "prefix": null,
138
    "problem_type": null,
139
    "pruned_heads": {},
140
    "remove_invalid_values": false,
141
    "repetition_penalty": 1.0,
142
    "return_dict": true,
143
    "return_dict_in_generate": false,
144
    "sep_token_id": null,
145
    "task_specific_params": null,
146
    "temperature": 1.0,
147
    "tie_encoder_decoder": false,
148
    "tie_word_embeddings": true,
149
    "tokenizer_class": null,
150
    "top_k": 50,
151
    "top_p": 1.0,
152
    "torch_dtype": null,
153
    "torchscript": false,
154
    "transformers_version": "4.9.0.dev0",
155
    "use_bfloat16": false
156
  }
157
}
158