MixTex commited on
Commit
37da049
1 Parent(s): 83c498d

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +193 -193
config.json CHANGED
@@ -1,193 +1,193 @@
1
- {
2
- "_name_or_path": "MixTex_ZhEn",
3
- "architectures": [
4
- "VisionEncoderDecoderModel"
5
- ],
6
- "decoder": {
7
- "_name_or_path": "roberta",
8
- "add_cross_attention": true,
9
- "architectures": [
10
- "RobertaForMaskedLM"
11
- ],
12
- "attention_probs_dropout_prob": 0.1,
13
- "bad_words_ids": null,
14
- "begin_suppress_tokens": null,
15
- "bos_token_id": 0,
16
- "chunk_size_feed_forward": 0,
17
- "classifier_dropout": null,
18
- "cross_attention_hidden_size": null,
19
- "decoder_start_token_id": null,
20
- "diversity_penalty": 0.0,
21
- "do_sample": false,
22
- "early_stopping": false,
23
- "encoder_no_repeat_ngram_size": 0,
24
- "eos_token_id": 2,
25
- "exponential_decay_length_penalty": null,
26
- "finetuning_task": null,
27
- "forced_bos_token_id": null,
28
- "forced_eos_token_id": null,
29
- "hidden_act": "gelu",
30
- "hidden_dropout_prob": 0.1,
31
- "hidden_size": 768,
32
- "id2label": {
33
- "0": "LABEL_0",
34
- "1": "LABEL_1"
35
- },
36
- "initializer_range": 0.02,
37
- "intermediate_size": 3072,
38
- "is_decoder": true,
39
- "is_encoder_decoder": false,
40
- "label2id": {
41
- "LABEL_0": 0,
42
- "LABEL_1": 1
43
- },
44
- "layer_norm_eps": 1e-12,
45
- "length_penalty": 1.0,
46
- "max_length": 20,
47
- "max_position_embeddings": 300,
48
- "min_length": 0,
49
- "model_type": "roberta",
50
- "no_repeat_ngram_size": 0,
51
- "num_attention_heads": 12,
52
- "num_beam_groups": 1,
53
- "num_beams": 1,
54
- "num_hidden_layers": 4,
55
- "num_return_sequences": 1,
56
- "output_attentions": false,
57
- "output_hidden_states": false,
58
- "output_scores": false,
59
- "pad_token_id": 1,
60
- "position_embedding_type": "absolute",
61
- "prefix": null,
62
- "problem_type": null,
63
- "pruned_heads": {},
64
- "remove_invalid_values": false,
65
- "repetition_penalty": 1.0,
66
- "return_dict": true,
67
- "return_dict_in_generate": false,
68
- "sep_token_id": null,
69
- "suppress_tokens": null,
70
- "task_specific_params": null,
71
- "temperature": 1.0,
72
- "tf_legacy_loss": false,
73
- "tie_encoder_decoder": false,
74
- "tie_word_embeddings": true,
75
- "tokenizer_class": null,
76
- "top_k": 50,
77
- "top_p": 1.0,
78
- "torch_dtype": "float32",
79
- "torchscript": false,
80
- "type_vocab_size": 1,
81
- "typical_p": 1.0,
82
- "use_bfloat16": false,
83
- "use_cache": false,
84
- "vocab_size": 25681
85
- },
86
- "decoder_start_token_id": 0,
87
- "early_stopping": true,
88
- "encoder": {
89
- "_name_or_path": "microsoft/swin-tiny-patch4-window7-224",
90
- "add_cross_attention": false,
91
- "architectures": [
92
- "SwinForImageClassification"
93
- ],
94
- "attention_probs_dropout_prob": 0.0,
95
- "bad_words_ids": null,
96
- "begin_suppress_tokens": null,
97
- "bos_token_id": null,
98
- "chunk_size_feed_forward": 0,
99
- "cross_attention_hidden_size": null,
100
- "decoder_start_token_id": null,
101
- "depths": [
102
- 2,
103
- 2,
104
- 6,
105
- 2
106
- ],
107
- "diversity_penalty": 0.0,
108
- "do_sample": false,
109
- "drop_path_rate": 0.1,
110
- "early_stopping": false,
111
- "embed_dim": 96,
112
- "encoder_no_repeat_ngram_size": 0,
113
- "encoder_stride": 32,
114
- "eos_token_id": null,
115
- "exponential_decay_length_penalty": null,
116
- "finetuning_task": null,
117
- "forced_bos_token_id": null,
118
- "forced_eos_token_id": null,
119
- "hidden_act": "gelu",
120
- "hidden_dropout_prob": 0.0,
121
- "hidden_size": 768,
122
- "layer_norm_eps": 1e-05,
123
- "length_penalty": 1.0,
124
- "max_length": 20,
125
- "min_length": 0,
126
- "mlp_ratio": 4.0,
127
- "model_type": "swin",
128
- "no_repeat_ngram_size": 0,
129
- "num_beam_groups": 1,
130
- "num_beams": 1,
131
- "num_channels": 3,
132
- "num_heads": [
133
- 3,
134
- 6,
135
- 12,
136
- 24
137
- ],
138
- "num_layers": 4,
139
- "num_return_sequences": 1,
140
- "out_features": [
141
- "stage4"
142
- ],
143
- "out_indices": [
144
- 4
145
- ],
146
- "output_attentions": false,
147
- "output_hidden_states": false,
148
- "output_scores": false,
149
- "pad_token_id": null,
150
- "patch_size": 4,
151
- "path_norm": true,
152
- "prefix": null,
153
- "problem_type": null,
154
- "pruned_heads": {},
155
- "qkv_bias": true,
156
- "remove_invalid_values": false,
157
- "repetition_penalty": 1.0,
158
- "return_dict": true,
159
- "return_dict_in_generate": false,
160
- "sep_token_id": null,
161
- "stage_names": [
162
- "stem",
163
- "stage1",
164
- "stage2",
165
- "stage3",
166
- "stage4"
167
- ],
168
- "suppress_tokens": null,
169
- "task_specific_params": null,
170
- "temperature": 1.0,
171
- "tf_legacy_loss": false,
172
- "tie_encoder_decoder": false,
173
- "tie_word_embeddings": true,
174
- "tokenizer_class": null,
175
- "top_k": 50,
176
- "top_p": 1.0,
177
- "torch_dtype": "float32",
178
- "torchscript": false,
179
- "typical_p": 1.0,
180
- "use_absolute_embeddings": false,
181
- "use_bfloat16": false,
182
- "window_size": 7
183
- },
184
- "eos_token_id": 25678,
185
- "is_encoder_decoder": true,
186
- "max_length": 296,
187
- "model_type": "vision-encoder-decoder",
188
- "pad_token_id": 1,
189
- "tie_word_embeddings": false,
190
- "torch_dtype": "float16",
191
- "transformers_version": "4.42.4",
192
- "vocab_size": 25681
193
- }
 
1
+ {
2
+ "_name_or_path": "MixTex_ZhEn",
3
+ "architectures": [
4
+ "VisionEncoderDecoderModel"
5
+ ],
6
+ "decoder": {
7
+ "_name_or_path": "roberta",
8
+ "add_cross_attention": true,
9
+ "architectures": [
10
+ "RobertaForMaskedLM"
11
+ ],
12
+ "attention_probs_dropout_prob": 0.1,
13
+ "bad_words_ids": null,
14
+ "begin_suppress_tokens": null,
15
+ "bos_token_id": 0,
16
+ "chunk_size_feed_forward": 0,
17
+ "classifier_dropout": null,
18
+ "cross_attention_hidden_size": null,
19
+ "decoder_start_token_id": null,
20
+ "diversity_penalty": 0.0,
21
+ "do_sample": false,
22
+ "early_stopping": false,
23
+ "encoder_no_repeat_ngram_size": 0,
24
+ "eos_token_id": 2,
25
+ "exponential_decay_length_penalty": null,
26
+ "finetuning_task": null,
27
+ "forced_bos_token_id": null,
28
+ "forced_eos_token_id": null,
29
+ "hidden_act": "gelu",
30
+ "hidden_dropout_prob": 0.1,
31
+ "hidden_size": 768,
32
+ "id2label": {
33
+ "0": "LABEL_0",
34
+ "1": "LABEL_1"
35
+ },
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "is_decoder": true,
39
+ "is_encoder_decoder": false,
40
+ "label2id": {
41
+ "LABEL_0": 0,
42
+ "LABEL_1": 1
43
+ },
44
+ "layer_norm_eps": 1e-12,
45
+ "length_penalty": 1.0,
46
+ "max_length": 20,
47
+ "max_position_embeddings": 300,
48
+ "min_length": 0,
49
+ "model_type": "roberta",
50
+ "no_repeat_ngram_size": 0,
51
+ "num_attention_heads": 12,
52
+ "num_beam_groups": 1,
53
+ "num_beams": 1,
54
+ "num_hidden_layers": 4,
55
+ "num_return_sequences": 1,
56
+ "output_attentions": false,
57
+ "output_hidden_states": false,
58
+ "output_scores": false,
59
+ "pad_token_id": 1,
60
+ "position_embedding_type": "absolute",
61
+ "prefix": null,
62
+ "problem_type": null,
63
+ "pruned_heads": {},
64
+ "remove_invalid_values": false,
65
+ "repetition_penalty": 1.0,
66
+ "return_dict": true,
67
+ "return_dict_in_generate": false,
68
+ "sep_token_id": null,
69
+ "suppress_tokens": null,
70
+ "task_specific_params": null,
71
+ "temperature": 1.0,
72
+ "tf_legacy_loss": false,
73
+ "tie_encoder_decoder": false,
74
+ "tie_word_embeddings": true,
75
+ "tokenizer_class": null,
76
+ "top_k": 50,
77
+ "top_p": 1.0,
78
+ "torch_dtype": "float32",
79
+ "torchscript": false,
80
+ "type_vocab_size": 1,
81
+ "typical_p": 1.0,
82
+ "use_bfloat16": false,
83
+ "use_cache": true,
84
+ "vocab_size": 25681
85
+ },
86
+ "decoder_start_token_id": 0,
87
+ "early_stopping": false,
88
+ "encoder": {
89
+ "_name_or_path": "microsoft/swin-tiny-patch4-window7-224",
90
+ "add_cross_attention": false,
91
+ "architectures": [
92
+ "SwinForImageClassification"
93
+ ],
94
+ "attention_probs_dropout_prob": 0.0,
95
+ "bad_words_ids": null,
96
+ "begin_suppress_tokens": null,
97
+ "bos_token_id": null,
98
+ "chunk_size_feed_forward": 0,
99
+ "cross_attention_hidden_size": null,
100
+ "decoder_start_token_id": null,
101
+ "depths": [
102
+ 2,
103
+ 2,
104
+ 6,
105
+ 2
106
+ ],
107
+ "diversity_penalty": 0.0,
108
+ "do_sample": false,
109
+ "drop_path_rate": 0.1,
110
+ "early_stopping": false,
111
+ "embed_dim": 96,
112
+ "encoder_no_repeat_ngram_size": 0,
113
+ "encoder_stride": 32,
114
+ "eos_token_id": null,
115
+ "exponential_decay_length_penalty": null,
116
+ "finetuning_task": null,
117
+ "forced_bos_token_id": null,
118
+ "forced_eos_token_id": null,
119
+ "hidden_act": "gelu",
120
+ "hidden_dropout_prob": 0.0,
121
+ "hidden_size": 768,
122
+ "layer_norm_eps": 1e-05,
123
+ "length_penalty": 1.0,
124
+ "max_length": 20,
125
+ "min_length": 0,
126
+ "mlp_ratio": 4.0,
127
+ "model_type": "swin",
128
+ "no_repeat_ngram_size": 0,
129
+ "num_beam_groups": 1,
130
+ "num_beams": 1,
131
+ "num_channels": 3,
132
+ "num_heads": [
133
+ 3,
134
+ 6,
135
+ 12,
136
+ 24
137
+ ],
138
+ "num_layers": 4,
139
+ "num_return_sequences": 1,
140
+ "out_features": [
141
+ "stage4"
142
+ ],
143
+ "out_indices": [
144
+ 4
145
+ ],
146
+ "output_attentions": false,
147
+ "output_hidden_states": false,
148
+ "output_scores": false,
149
+ "pad_token_id": null,
150
+ "patch_size": 4,
151
+ "path_norm": true,
152
+ "prefix": null,
153
+ "problem_type": null,
154
+ "pruned_heads": {},
155
+ "qkv_bias": true,
156
+ "remove_invalid_values": false,
157
+ "repetition_penalty": 1.0,
158
+ "return_dict": true,
159
+ "return_dict_in_generate": false,
160
+ "sep_token_id": null,
161
+ "stage_names": [
162
+ "stem",
163
+ "stage1",
164
+ "stage2",
165
+ "stage3",
166
+ "stage4"
167
+ ],
168
+ "suppress_tokens": null,
169
+ "task_specific_params": null,
170
+ "temperature": 1.0,
171
+ "tf_legacy_loss": false,
172
+ "tie_encoder_decoder": false,
173
+ "tie_word_embeddings": true,
174
+ "tokenizer_class": null,
175
+ "top_k": 50,
176
+ "top_p": 1.0,
177
+ "torch_dtype": "float32",
178
+ "torchscript": false,
179
+ "typical_p": 1.0,
180
+ "use_absolute_embeddings": false,
181
+ "use_bfloat16": false,
182
+ "window_size": 7
183
+ },
184
+ "eos_token_id": 25678,
185
+ "is_encoder_decoder": true,
186
+ "max_length": 296,
187
+ "model_type": "vision-encoder-decoder",
188
+ "pad_token_id": 1,
189
+ "tie_word_embeddings": false,
190
+ "torch_dtype": "float16",
191
+ "transformers_version": "4.42.4",
192
+ "vocab_size": 25681
193
+ }