Tim Miller
commited on
Commit
•
26adb3a
1
Parent(s):
e8154cc
First version of model.
Browse files- added_tokens.json +10 -0
- config.json +108 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +25 -0
- tokenizer.json +0 -0
- tokenizer_config.json +26 -0
- vocab.json +0 -0
added_tokens.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"</a1>": 50268,
|
3 |
+
"</a2>": 50270,
|
4 |
+
"</e>": 50266,
|
5 |
+
"<a1>": 50267,
|
6 |
+
"<a2>": 50269,
|
7 |
+
"<cr>": 50271,
|
8 |
+
"<e>": 50265,
|
9 |
+
"<neg>": 50272
|
10 |
+
}
|
config.json
ADDED
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"CnlpModelForClassification"
|
4 |
+
],
|
5 |
+
"encoder_config": {
|
6 |
+
"_name_or_path": "roberta-base",
|
7 |
+
"add_cross_attention": false,
|
8 |
+
"architectures": [
|
9 |
+
"RobertaForMaskedLM"
|
10 |
+
],
|
11 |
+
"attention_probs_dropout_prob": 0.1,
|
12 |
+
"bad_words_ids": null,
|
13 |
+
"bos_token_id": 0,
|
14 |
+
"chunk_size_feed_forward": 0,
|
15 |
+
"classifier_dropout": null,
|
16 |
+
"cross_attention_hidden_size": null,
|
17 |
+
"decoder_start_token_id": null,
|
18 |
+
"diversity_penalty": 0.0,
|
19 |
+
"do_sample": false,
|
20 |
+
"early_stopping": false,
|
21 |
+
"encoder_no_repeat_ngram_size": 0,
|
22 |
+
"eos_token_id": 2,
|
23 |
+
"exponential_decay_length_penalty": null,
|
24 |
+
"finetuning_task": null,
|
25 |
+
"forced_bos_token_id": null,
|
26 |
+
"forced_eos_token_id": null,
|
27 |
+
"hidden_act": "gelu",
|
28 |
+
"hidden_dropout_prob": 0.1,
|
29 |
+
"hidden_size": 768,
|
30 |
+
"id2label": {
|
31 |
+
"0": "LABEL_0",
|
32 |
+
"1": "LABEL_1"
|
33 |
+
},
|
34 |
+
"initializer_range": 0.02,
|
35 |
+
"intermediate_size": 3072,
|
36 |
+
"is_decoder": false,
|
37 |
+
"is_encoder_decoder": false,
|
38 |
+
"label2id": {
|
39 |
+
"LABEL_0": 0,
|
40 |
+
"LABEL_1": 1
|
41 |
+
},
|
42 |
+
"layer_norm_eps": 1e-05,
|
43 |
+
"length_penalty": 1.0,
|
44 |
+
"max_length": 20,
|
45 |
+
"max_position_embeddings": 514,
|
46 |
+
"min_length": 0,
|
47 |
+
"model_type": "roberta",
|
48 |
+
"no_repeat_ngram_size": 0,
|
49 |
+
"num_attention_heads": 12,
|
50 |
+
"num_beam_groups": 1,
|
51 |
+
"num_beams": 1,
|
52 |
+
"num_hidden_layers": 12,
|
53 |
+
"num_return_sequences": 1,
|
54 |
+
"output_attentions": false,
|
55 |
+
"output_hidden_states": false,
|
56 |
+
"output_scores": false,
|
57 |
+
"pad_token_id": 1,
|
58 |
+
"position_embedding_type": "absolute",
|
59 |
+
"prefix": null,
|
60 |
+
"problem_type": null,
|
61 |
+
"pruned_heads": {},
|
62 |
+
"remove_invalid_values": false,
|
63 |
+
"repetition_penalty": 1.0,
|
64 |
+
"return_dict": true,
|
65 |
+
"return_dict_in_generate": false,
|
66 |
+
"sep_token_id": null,
|
67 |
+
"task_specific_params": null,
|
68 |
+
"temperature": 1.0,
|
69 |
+
"tf_legacy_loss": false,
|
70 |
+
"tie_encoder_decoder": false,
|
71 |
+
"tie_word_embeddings": true,
|
72 |
+
"tokenizer_class": null,
|
73 |
+
"top_k": 50,
|
74 |
+
"top_p": 1.0,
|
75 |
+
"torch_dtype": null,
|
76 |
+
"torchscript": false,
|
77 |
+
"transformers_version": "4.22.0.dev0",
|
78 |
+
"type_vocab_size": 1,
|
79 |
+
"typical_p": 1.0,
|
80 |
+
"use_bfloat16": false,
|
81 |
+
"use_cache": true,
|
82 |
+
"vocab_size": 50273
|
83 |
+
},
|
84 |
+
"encoder_name": "roberta-base",
|
85 |
+
"finetuning_task": [
|
86 |
+
"TermExists"
|
87 |
+
],
|
88 |
+
"hidden_dropout_prob": 0.1,
|
89 |
+
"hidden_size": 768,
|
90 |
+
"layer": -1,
|
91 |
+
"model_type": "cnlpt",
|
92 |
+
"num_labels_list": [
|
93 |
+
2
|
94 |
+
],
|
95 |
+
"num_rel_attention_heads": 12,
|
96 |
+
"rel_attention_head_dims": 64,
|
97 |
+
"relations": [
|
98 |
+
false
|
99 |
+
],
|
100 |
+
"tagger": [
|
101 |
+
false
|
102 |
+
],
|
103 |
+
"tokens": false,
|
104 |
+
"torch_dtype": "float32",
|
105 |
+
"transformers_version": "4.22.0.dev0",
|
106 |
+
"use_prior_tasks": false,
|
107 |
+
"vocab_size": 50273
|
108 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bf8917bb83e56202f1b1817b0d3511ecb02ce38b055b7d090a17777a5f522665
|
3 |
+
size 501047977
|
special_tokens_map.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<e>",
|
4 |
+
"</e>",
|
5 |
+
"<a1>",
|
6 |
+
"</a1>",
|
7 |
+
"<a2>",
|
8 |
+
"</a2>",
|
9 |
+
"<cr>",
|
10 |
+
"<neg>"
|
11 |
+
],
|
12 |
+
"bos_token": "<s>",
|
13 |
+
"cls_token": "<s>",
|
14 |
+
"eos_token": "</s>",
|
15 |
+
"mask_token": {
|
16 |
+
"content": "<mask>",
|
17 |
+
"lstrip": true,
|
18 |
+
"normalized": false,
|
19 |
+
"rstrip": false,
|
20 |
+
"single_word": false
|
21 |
+
},
|
22 |
+
"pad_token": "<pad>",
|
23 |
+
"sep_token": "</s>",
|
24 |
+
"unk_token": "<unk>"
|
25 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": true,
|
3 |
+
"additional_special_tokens": [
|
4 |
+
"<e>",
|
5 |
+
"</e>",
|
6 |
+
"<a1>",
|
7 |
+
"</a1>",
|
8 |
+
"<a2>",
|
9 |
+
"</a2>",
|
10 |
+
"<cr>",
|
11 |
+
"<neg>"
|
12 |
+
],
|
13 |
+
"bos_token": "<s>",
|
14 |
+
"cls_token": "<s>",
|
15 |
+
"eos_token": "</s>",
|
16 |
+
"errors": "replace",
|
17 |
+
"mask_token": "<mask>",
|
18 |
+
"model_max_length": 512,
|
19 |
+
"name_or_path": "roberta-base",
|
20 |
+
"pad_token": "<pad>",
|
21 |
+
"sep_token": "</s>",
|
22 |
+
"special_tokens_map_file": null,
|
23 |
+
"tokenizer_class": "RobertaTokenizer",
|
24 |
+
"trim_offsets": true,
|
25 |
+
"unk_token": "<unk>"
|
26 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|