laituan245 commited on
Commit
7381fbd
1 Parent(s): 06bf522

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +218 -0
config.json ADDED
@@ -0,0 +1,218 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "RagSequenceForGeneration"
4
+ ],
5
+ "dataset": "wiki_dpr",
6
+ "dataset_split": "train",
7
+ "do_deduplication": true,
8
+ "do_marginalize": false,
9
+ "doc_sep": " // ",
10
+ "exclude_bos_score": false,
11
+ "forced_eos_token_id": 2,
12
+ "generator": {
13
+ "_name_or_path": "facebook/bart-base",
14
+ "activation_dropout": 0.1,
15
+ "activation_function": "gelu",
16
+ "add_bias_logits": false,
17
+ "add_cross_attention": false,
18
+ "add_final_layer_norm": false,
19
+ "architectures": [
20
+ "BartModel"
21
+ ],
22
+ "attention_dropout": 0.1,
23
+ "bad_words_ids": null,
24
+ "bos_token_id": 0,
25
+ "chunk_size_feed_forward": 0,
26
+ "classif_dropout": 0.1,
27
+ "classifier_dropout": 0.0,
28
+ "cross_attention_hidden_size": null,
29
+ "d_model": 768,
30
+ "decoder_attention_heads": 12,
31
+ "decoder_ffn_dim": 3072,
32
+ "decoder_layerdrop": 0.0,
33
+ "decoder_layers": 6,
34
+ "decoder_start_token_id": 2,
35
+ "diversity_penalty": 0.0,
36
+ "do_sample": false,
37
+ "dropout": 0.1,
38
+ "early_stopping": true,
39
+ "encoder_attention_heads": 12,
40
+ "encoder_ffn_dim": 3072,
41
+ "encoder_layerdrop": 0.0,
42
+ "encoder_layers": 6,
43
+ "encoder_no_repeat_ngram_size": 0,
44
+ "eos_token_id": 2,
45
+ "exponential_decay_length_penalty": null,
46
+ "finetuning_task": null,
47
+ "forced_bos_token_id": 0,
48
+ "forced_eos_token_id": 2,
49
+ "gradient_checkpointing": false,
50
+ "id2label": {
51
+ "0": "LABEL_0",
52
+ "1": "LABEL_1",
53
+ "2": "LABEL_2"
54
+ },
55
+ "init_std": 0.02,
56
+ "is_decoder": false,
57
+ "is_encoder_decoder": true,
58
+ "label2id": {
59
+ "LABEL_0": 0,
60
+ "LABEL_1": 1,
61
+ "LABEL_2": 2
62
+ },
63
+ "length_penalty": 1.0,
64
+ "max_length": 20,
65
+ "max_position_embeddings": 1024,
66
+ "min_length": 0,
67
+ "model_type": "bart",
68
+ "no_repeat_ngram_size": 3,
69
+ "normalize_before": false,
70
+ "normalize_embedding": true,
71
+ "num_beam_groups": 1,
72
+ "num_beams": 4,
73
+ "num_hidden_layers": 6,
74
+ "num_return_sequences": 1,
75
+ "output_attentions": false,
76
+ "output_hidden_states": false,
77
+ "output_scores": false,
78
+ "pad_token_id": 1,
79
+ "prefix": null,
80
+ "problem_type": null,
81
+ "pruned_heads": {},
82
+ "remove_invalid_values": false,
83
+ "repetition_penalty": 1.0,
84
+ "return_dict": true,
85
+ "return_dict_in_generate": false,
86
+ "scale_embedding": false,
87
+ "sep_token_id": null,
88
+ "task_specific_params": {
89
+ "summarization": {
90
+ "length_penalty": 1.0,
91
+ "max_length": 128,
92
+ "min_length": 12,
93
+ "num_beams": 4
94
+ },
95
+ "summarization_cnn": {
96
+ "length_penalty": 2.0,
97
+ "max_length": 142,
98
+ "min_length": 56,
99
+ "num_beams": 4
100
+ },
101
+ "summarization_xsum": {
102
+ "length_penalty": 1.0,
103
+ "max_length": 62,
104
+ "min_length": 11,
105
+ "num_beams": 6
106
+ }
107
+ },
108
+ "temperature": 1.0,
109
+ "tie_encoder_decoder": false,
110
+ "tie_word_embeddings": true,
111
+ "tokenizer_class": null,
112
+ "top_k": 50,
113
+ "top_p": 1.0,
114
+ "torch_dtype": "float32",
115
+ "torchscript": false,
116
+ "transformers_version": "4.19.2",
117
+ "typical_p": 1.0,
118
+ "use_bfloat16": false,
119
+ "use_cache": true,
120
+ "vocab_size": 50265
121
+ },
122
+ "index_name": "exact",
123
+ "index_path": null,
124
+ "is_encoder_decoder": true,
125
+ "label_smoothing": 0.0,
126
+ "max_combined_length": 300,
127
+ "model_type": "rag",
128
+ "n_docs": 5,
129
+ "output_retrieved": false,
130
+ "passages_path": null,
131
+ "question_encoder": {
132
+ "_name_or_path": "facebook/dpr-question_encoder-single-nq-base",
133
+ "add_cross_attention": false,
134
+ "architectures": [
135
+ "DPRQuestionEncoder"
136
+ ],
137
+ "attention_probs_dropout_prob": 0.1,
138
+ "bad_words_ids": null,
139
+ "bos_token_id": null,
140
+ "chunk_size_feed_forward": 0,
141
+ "cross_attention_hidden_size": null,
142
+ "decoder_start_token_id": null,
143
+ "diversity_penalty": 0.0,
144
+ "do_sample": false,
145
+ "early_stopping": false,
146
+ "encoder_no_repeat_ngram_size": 0,
147
+ "eos_token_id": null,
148
+ "exponential_decay_length_penalty": null,
149
+ "finetuning_task": null,
150
+ "forced_bos_token_id": null,
151
+ "forced_eos_token_id": null,
152
+ "gradient_checkpointing": false,
153
+ "hidden_act": "gelu",
154
+ "hidden_dropout_prob": 0.1,
155
+ "hidden_size": 768,
156
+ "id2label": {
157
+ "0": "LABEL_0",
158
+ "1": "LABEL_1"
159
+ },
160
+ "initializer_range": 0.02,
161
+ "intermediate_size": 3072,
162
+ "is_decoder": false,
163
+ "is_encoder_decoder": false,
164
+ "label2id": {
165
+ "LABEL_0": 0,
166
+ "LABEL_1": 1
167
+ },
168
+ "layer_norm_eps": 1e-12,
169
+ "length_penalty": 1.0,
170
+ "max_length": 20,
171
+ "max_position_embeddings": 512,
172
+ "min_length": 0,
173
+ "model_type": "dpr",
174
+ "no_repeat_ngram_size": 0,
175
+ "num_attention_heads": 12,
176
+ "num_beam_groups": 1,
177
+ "num_beams": 1,
178
+ "num_hidden_layers": 12,
179
+ "num_return_sequences": 1,
180
+ "output_attentions": false,
181
+ "output_hidden_states": false,
182
+ "output_scores": false,
183
+ "pad_token_id": 0,
184
+ "position_embedding_type": "absolute",
185
+ "prefix": null,
186
+ "problem_type": null,
187
+ "projection_dim": 0,
188
+ "pruned_heads": {},
189
+ "remove_invalid_values": false,
190
+ "repetition_penalty": 1.0,
191
+ "return_dict": true,
192
+ "return_dict_in_generate": false,
193
+ "sep_token_id": null,
194
+ "task_specific_params": null,
195
+ "temperature": 1.0,
196
+ "tie_encoder_decoder": false,
197
+ "tie_word_embeddings": true,
198
+ "tokenizer_class": null,
199
+ "top_k": 50,
200
+ "top_p": 1.0,
201
+ "torch_dtype": null,
202
+ "torchscript": false,
203
+ "transformers_version": "4.19.2",
204
+ "type_vocab_size": 2,
205
+ "typical_p": 1.0,
206
+ "use_bfloat16": false,
207
+ "vocab_size": 30522
208
+ },
209
+ "reduce_loss": false,
210
+ "retrieval_batch_size": 8,
211
+ "retrieval_vector_size": 768,
212
+ "title_sep": " / ",
213
+ "torch_dtype": "float32",
214
+ "transformers_version": null,
215
+ "use_cache": true,
216
+ "use_dummy_dataset": true,
217
+ "vocab_size": null
218
+ }