nvbic commited on
Commit
2a00b2c
·
verified ·
1 Parent(s): 3e409bc

Upload NER model of BERT

Browse files
ner_bert/added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"[next]": 30522}
ner_bert/base_model_name.txt ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ model_type: bert
2
+ base_model: bert-base-uncased
3
+ config: bert-base-uncased
4
+ tokenizer: bert-base-uncased
ner_bert/checkpoint_23999.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0006cc142bef643b43fd9216e354dc81cbd10db6292baf8472fd5791c21cadb2
3
+ size 438282889
ner_bert/checkpoint_25999.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07f4d1577b29ad1b7106ff2068d94d13a6ef3560336a3756e2cd91db608800b4
3
+ size 438282889
ner_bert/checkpoint_34999.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45b72b5e706d6b763e0b927e25e886d37adedf35f46997998a211f0f3b234edc
3
+ size 438282889
ner_bert/config.json ADDED
@@ -0,0 +1,286 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bert-base-uncased",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84"
97
+ },
98
+ "initializer_range": 0.02,
99
+ "intermediate_size": 3072,
100
+ "label2id": {
101
+ "LABEL_0": 0,
102
+ "LABEL_1": 1,
103
+ "LABEL_10": 10,
104
+ "LABEL_11": 11,
105
+ "LABEL_12": 12,
106
+ "LABEL_13": 13,
107
+ "LABEL_14": 14,
108
+ "LABEL_15": 15,
109
+ "LABEL_16": 16,
110
+ "LABEL_17": 17,
111
+ "LABEL_18": 18,
112
+ "LABEL_19": 19,
113
+ "LABEL_2": 2,
114
+ "LABEL_20": 20,
115
+ "LABEL_21": 21,
116
+ "LABEL_22": 22,
117
+ "LABEL_23": 23,
118
+ "LABEL_24": 24,
119
+ "LABEL_25": 25,
120
+ "LABEL_26": 26,
121
+ "LABEL_27": 27,
122
+ "LABEL_28": 28,
123
+ "LABEL_29": 29,
124
+ "LABEL_3": 3,
125
+ "LABEL_30": 30,
126
+ "LABEL_31": 31,
127
+ "LABEL_32": 32,
128
+ "LABEL_33": 33,
129
+ "LABEL_34": 34,
130
+ "LABEL_35": 35,
131
+ "LABEL_36": 36,
132
+ "LABEL_37": 37,
133
+ "LABEL_38": 38,
134
+ "LABEL_39": 39,
135
+ "LABEL_4": 4,
136
+ "LABEL_40": 40,
137
+ "LABEL_41": 41,
138
+ "LABEL_42": 42,
139
+ "LABEL_43": 43,
140
+ "LABEL_44": 44,
141
+ "LABEL_45": 45,
142
+ "LABEL_46": 46,
143
+ "LABEL_47": 47,
144
+ "LABEL_48": 48,
145
+ "LABEL_49": 49,
146
+ "LABEL_5": 5,
147
+ "LABEL_50": 50,
148
+ "LABEL_51": 51,
149
+ "LABEL_52": 52,
150
+ "LABEL_53": 53,
151
+ "LABEL_54": 54,
152
+ "LABEL_55": 55,
153
+ "LABEL_56": 56,
154
+ "LABEL_57": 57,
155
+ "LABEL_58": 58,
156
+ "LABEL_59": 59,
157
+ "LABEL_6": 6,
158
+ "LABEL_60": 60,
159
+ "LABEL_61": 61,
160
+ "LABEL_62": 62,
161
+ "LABEL_63": 63,
162
+ "LABEL_64": 64,
163
+ "LABEL_65": 65,
164
+ "LABEL_66": 66,
165
+ "LABEL_67": 67,
166
+ "LABEL_68": 68,
167
+ "LABEL_69": 69,
168
+ "LABEL_7": 7,
169
+ "LABEL_70": 70,
170
+ "LABEL_71": 71,
171
+ "LABEL_72": 72,
172
+ "LABEL_73": 73,
173
+ "LABEL_74": 74,
174
+ "LABEL_75": 75,
175
+ "LABEL_76": 76,
176
+ "LABEL_77": 77,
177
+ "LABEL_78": 78,
178
+ "LABEL_79": 79,
179
+ "LABEL_8": 8,
180
+ "LABEL_80": 80,
181
+ "LABEL_81": 81,
182
+ "LABEL_82": 82,
183
+ "LABEL_83": 83,
184
+ "LABEL_84": 84,
185
+ "LABEL_9": 9
186
+ },
187
+ "label2idx": {
188
+ "B-Abuse": 5,
189
+ "B-Alcohol_freq": 7,
190
+ "B-Alcohol_other": 8,
191
+ "B-Alcohol_type": 10,
192
+ "B-Alcohol_use": 12,
193
+ "B-Drug_freq": 15,
194
+ "B-Drug_other": 16,
195
+ "B-Drug_type": 19,
196
+ "B-Drug_use": 21,
197
+ "B-Education": 23,
198
+ "B-Employment_location": 25,
199
+ "B-Employment_status": 27,
200
+ "B-Ethnicity": 29,
201
+ "B-Financial_constrain": 30,
202
+ "B-Gender": 33,
203
+ "B-Gender_status": 34,
204
+ "B-Language": 36,
205
+ "B-Living_Condition": 37,
206
+ "B-Living_supply": 40,
207
+ "B-Marital_status": 42,
208
+ "B-Occupation": 43,
209
+ "B-Partner": 46,
210
+ "B-Physical_act": 48,
211
+ "B-Protection": 50,
212
+ "B-Race": 52,
213
+ "B-SDoH_ICD": 53,
214
+ "B-Sdoh_freq": 56,
215
+ "B-Sdoh_status": 58,
216
+ "B-Sex_act": 59,
217
+ "B-Sex_act_other": 62,
218
+ "B-Smoking_freq_other": 64,
219
+ "B-Smoking_freq_ppd": 65,
220
+ "B-Smoking_freq_py": 68,
221
+ "B-Smoking_freq_qy": 70,
222
+ "B-Smoking_freq_sy": 72,
223
+ "B-Smoking_other": 74,
224
+ "B-Smoking_type": 76,
225
+ "B-Social_cohesion": 78,
226
+ "B-Substance_use_status": 80,
227
+ "B-Tobacco_use": 82,
228
+ "B-Transportation": 83,
229
+ "CLS": 1,
230
+ "I-Abuse": 4,
231
+ "I-Alcohol_freq": 6,
232
+ "I-Alcohol_other": 9,
233
+ "I-Alcohol_type": 11,
234
+ "I-Alcohol_use": 13,
235
+ "I-Drug_freq": 14,
236
+ "I-Drug_other": 17,
237
+ "I-Drug_type": 18,
238
+ "I-Drug_use": 20,
239
+ "I-Education": 22,
240
+ "I-Employment_location": 24,
241
+ "I-Employment_status": 26,
242
+ "I-Ethnicity": 28,
243
+ "I-Financial_constrain": 31,
244
+ "I-Gender": 32,
245
+ "I-Gender_status": 35,
246
+ "I-Living_Condition": 38,
247
+ "I-Living_supply": 39,
248
+ "I-Marital_status": 41,
249
+ "I-Occupation": 44,
250
+ "I-Partner": 45,
251
+ "I-Physical_act": 47,
252
+ "I-Protection": 49,
253
+ "I-Race": 51,
254
+ "I-SDoH_ICD": 54,
255
+ "I-Sdoh_freq": 55,
256
+ "I-Sdoh_status": 57,
257
+ "I-Sex_act": 60,
258
+ "I-Sex_act_other": 61,
259
+ "I-Smoking_freq_other": 63,
260
+ "I-Smoking_freq_ppd": 66,
261
+ "I-Smoking_freq_py": 67,
262
+ "I-Smoking_freq_qy": 69,
263
+ "I-Smoking_freq_sy": 71,
264
+ "I-Smoking_other": 73,
265
+ "I-Smoking_type": 75,
266
+ "I-Social_cohesion": 77,
267
+ "I-Substance_use_status": 79,
268
+ "I-Tobacco_use": 81,
269
+ "I-Transportation": 84,
270
+ "O": 3,
271
+ "PAD": 0,
272
+ "X": 2
273
+ },
274
+ "layer_norm_eps": 1e-12,
275
+ "max_position_embeddings": 512,
276
+ "model_type": "bert",
277
+ "num_attention_heads": 12,
278
+ "num_hidden_layers": 12,
279
+ "pad_token_id": 0,
280
+ "position_embedding_type": "absolute",
281
+ "transformers_version": "4.3.1",
282
+ "type_vocab_size": 2,
283
+ "use_cache": true,
284
+ "use_crf": false,
285
+ "vocab_size": 30523
286
+ }
ner_bert/label2idx.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"O": 3, "X": 2, "PAD": 0, "CLS": 1, "I-Abuse": 4, "B-Abuse": 5, "I-Alcohol_freq": 6, "B-Alcohol_freq": 7, "B-Alcohol_other": 8, "I-Alcohol_other": 9, "B-Alcohol_type": 10, "I-Alcohol_type": 11, "B-Alcohol_use": 12, "I-Alcohol_use": 13, "I-Drug_freq": 14, "B-Drug_freq": 15, "B-Drug_other": 16, "I-Drug_other": 17, "I-Drug_type": 18, "B-Drug_type": 19, "I-Drug_use": 20, "B-Drug_use": 21, "I-Education": 22, "B-Education": 23, "I-Employment_location": 24, "B-Employment_location": 25, "I-Employment_status": 26, "B-Employment_status": 27, "I-Ethnicity": 28, "B-Ethnicity": 29, "B-Financial_constrain": 30, "I-Financial_constrain": 31, "I-Gender": 32, "B-Gender": 33, "B-Gender_status": 34, "I-Gender_status": 35, "B-Language": 36, "B-Living_Condition": 37, "I-Living_Condition": 38, "I-Living_supply": 39, "B-Living_supply": 40, "I-Marital_status": 41, "B-Marital_status": 42, "B-Occupation": 43, "I-Occupation": 44, "I-Partner": 45, "B-Partner": 46, "I-Physical_act": 47, "B-Physical_act": 48, "I-Protection": 49, "B-Protection": 50, "I-Race": 51, "B-Race": 52, "B-SDoH_ICD": 53, "I-SDoH_ICD": 54, "I-Sdoh_freq": 55, "B-Sdoh_freq": 56, "I-Sdoh_status": 57, "B-Sdoh_status": 58, "B-Sex_act": 59, "I-Sex_act": 60, "I-Sex_act_other": 61, "B-Sex_act_other": 62, "I-Smoking_freq_other": 63, "B-Smoking_freq_other": 64, "B-Smoking_freq_ppd": 65, "I-Smoking_freq_ppd": 66, "I-Smoking_freq_py": 67, "B-Smoking_freq_py": 68, "I-Smoking_freq_qy": 69, "B-Smoking_freq_qy": 70, "I-Smoking_freq_sy": 71, "B-Smoking_freq_sy": 72, "I-Smoking_other": 73, "B-Smoking_other": 74, "I-Smoking_type": 75, "B-Smoking_type": 76, "I-Social_cohesion": 77, "B-Social_cohesion": 78, "I-Substance_use_status": 79, "B-Substance_use_status": 80, "I-Tobacco_use": 81, "B-Tobacco_use": 82, "B-Transportation": 83, "I-Transportation": 84}
ner_bert/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2abc8014f90746d58a168aee4f5ee338c804d132a236e9703e6d118ebb721456
3
+ size 438018551
ner_bert/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
ner_bert/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "bert-base-uncased"}
ner_bert/vocab.txt ADDED
The diff for this file is too large to render. See raw diff