townboy commited on
Commit
7e40c86
·
verified ·
1 Parent(s): 9fe4df6

Publish final KPF-BERT Korean PII NER model

Browse files
README.md ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ko
4
+ pipeline_tag: token-classification
5
+ tags:
6
+ - korean
7
+ - ner
8
+ - pii
9
+ - kpfbert
10
+ base_model: KPF/KPF-bert-ner
11
+ ---
12
+
13
+ # KPF-BERT Korean PII NER
14
+
15
+ 한국어 문장에서 33종의 PII 라벨을 탐지하는 KPF-BERT 기반 토큰 분류 모델입니다.
16
+
17
+ ## Labels
18
+
19
+ BIO 형식의 67개 라벨(`O` 포함)을 사용합니다. 전체 라벨 매핑은 `label_map.json`과 `config.json`을 참고하세요.
20
+
21
+ ## Training data
22
+
23
+ 사용자가 제작한 합성 한국어 PII 데이터셋으로 파인튜닝했습니다. 실제 개인 정보가 포함된 데이터가 아닌 합성·검증 데이터 기반입니다.
24
+
25
+ ## Evaluation
26
+
27
+ - Evaluation: independent A/B/C holdout sets
28
+ - Average micro-F1: 0.9759
29
+ - Minimum per-label F1: 0.750
30
+
31
+ 이 수치는 holdout 구성과 평가 방식에 따라 달라질 수 있습니다.
32
+
33
+ ## Usage
34
+
35
+ ```python
36
+ from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
37
+
38
+ model_id = "townboy/kpfbert-ner"
39
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
40
+ model = AutoModelForTokenClassification.from_pretrained(model_id)
41
+ ner = pipeline("token-classification", model=model, tokenizer=tokenizer,
42
+ aggregation_strategy="simple")
43
+
44
+ print(ner("회원 이름은 홍길동이고 이메일은 hong@example.com입니다."))
45
+ ```
best_validation_metrics.json ADDED
@@ -0,0 +1,273 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1,
3
+ "train_loss": 0.005999320498434073,
4
+ "micro_precision": 0.9850155499010461,
5
+ "micro_recall": 0.987808335696059,
6
+ "micro_f1": 0.9864099660249152,
7
+ "per_label": {
8
+ "ACCOUNT_NUMBER": {
9
+ "precision": 0.918918918918919,
10
+ "recall": 1.0,
11
+ "f1": 0.9577464788732395,
12
+ "tp": 68,
13
+ "fp": 6,
14
+ "fn": 0
15
+ },
16
+ "ADDRESS": {
17
+ "precision": 0.8260869565217391,
18
+ "recall": 0.9047619047619048,
19
+ "f1": 0.8636363636363636,
20
+ "tp": 19,
21
+ "fp": 4,
22
+ "fn": 2
23
+ },
24
+ "AGE": {
25
+ "precision": 0.9428571428571428,
26
+ "recall": 0.9705882352941176,
27
+ "f1": 0.9565217391304348,
28
+ "tp": 33,
29
+ "fp": 2,
30
+ "fn": 1
31
+ },
32
+ "ALIEN_NUMBER": {
33
+ "precision": 1.0,
34
+ "recall": 0.5,
35
+ "f1": 0.6666666666666666,
36
+ "tp": 10,
37
+ "fp": 0,
38
+ "fn": 10
39
+ },
40
+ "BIRTHDATE": {
41
+ "precision": 0.9961685823754789,
42
+ "recall": 1.0,
43
+ "f1": 0.9980806142034548,
44
+ "tp": 260,
45
+ "fp": 1,
46
+ "fn": 0
47
+ },
48
+ "BLOOD_TYPE": {
49
+ "precision": 0.9819819819819819,
50
+ "recall": 1.0,
51
+ "f1": 0.9909090909090909,
52
+ "tp": 218,
53
+ "fp": 4,
54
+ "fn": 0
55
+ },
56
+ "CARD_NUMBER": {
57
+ "precision": 1.0,
58
+ "recall": 1.0,
59
+ "f1": 1.0,
60
+ "tp": 19,
61
+ "fp": 0,
62
+ "fn": 0
63
+ },
64
+ "CITY": {
65
+ "precision": 0.9988571428571429,
66
+ "recall": 1.0,
67
+ "f1": 0.9994282447112636,
68
+ "tp": 874,
69
+ "fp": 1,
70
+ "fn": 0
71
+ },
72
+ "DEPARTMENT": {
73
+ "precision": 0.9688667496886675,
74
+ "recall": 0.9676616915422885,
75
+ "f1": 0.9682638456751711,
76
+ "tp": 778,
77
+ "fp": 25,
78
+ "fn": 26
79
+ },
80
+ "DRIVER_LICENSE": {
81
+ "precision": 1.0,
82
+ "recall": 1.0,
83
+ "f1": 1.0,
84
+ "tp": 20,
85
+ "fp": 0,
86
+ "fn": 0
87
+ },
88
+ "EMAIL": {
89
+ "precision": 1.0,
90
+ "recall": 1.0,
91
+ "f1": 1.0,
92
+ "tp": 99,
93
+ "fp": 0,
94
+ "fn": 0
95
+ },
96
+ "EMPLOYEE_ID": {
97
+ "precision": 1.0,
98
+ "recall": 1.0,
99
+ "f1": 1.0,
100
+ "tp": 42,
101
+ "fp": 0,
102
+ "fn": 0
103
+ },
104
+ "GENDER": {
105
+ "precision": 1.0,
106
+ "recall": 0.9705882352941176,
107
+ "f1": 0.9850746268656716,
108
+ "tp": 33,
109
+ "fp": 0,
110
+ "fn": 1
111
+ },
112
+ "HEIGHT": {
113
+ "precision": 1.0,
114
+ "recall": 1.0,
115
+ "f1": 1.0,
116
+ "tp": 218,
117
+ "fp": 0,
118
+ "fn": 0
119
+ },
120
+ "IP_ADDRESS": {
121
+ "precision": 1.0,
122
+ "recall": 1.0,
123
+ "f1": 1.0,
124
+ "tp": 35,
125
+ "fp": 0,
126
+ "fn": 0
127
+ },
128
+ "MAJOR": {
129
+ "precision": 0.8888888888888888,
130
+ "recall": 1.0,
131
+ "f1": 0.9411764705882353,
132
+ "tp": 24,
133
+ "fp": 3,
134
+ "fn": 0
135
+ },
136
+ "MEMBER_ID": {
137
+ "precision": 0.9285714285714286,
138
+ "recall": 1.0,
139
+ "f1": 0.962962962962963,
140
+ "tp": 26,
141
+ "fp": 2,
142
+ "fn": 0
143
+ },
144
+ "NAME": {
145
+ "precision": 0.9815140845070423,
146
+ "recall": 0.9982094897045658,
147
+ "f1": 0.9897913892587662,
148
+ "tp": 1115,
149
+ "fp": 21,
150
+ "fn": 2
151
+ },
152
+ "NATIONALITY": {
153
+ "precision": 1.0,
154
+ "recall": 1.0,
155
+ "f1": 1.0,
156
+ "tp": 34,
157
+ "fp": 0,
158
+ "fn": 0
159
+ },
160
+ "NICKNAME": {
161
+ "precision": 1.0,
162
+ "recall": 0.9642857142857143,
163
+ "f1": 0.9818181818181818,
164
+ "tp": 27,
165
+ "fp": 0,
166
+ "fn": 1
167
+ },
168
+ "PARTICIPANT_ID": {
169
+ "precision": 1.0,
170
+ "recall": 0.75,
171
+ "f1": 0.8571428571428571,
172
+ "tp": 6,
173
+ "fp": 0,
174
+ "fn": 2
175
+ },
176
+ "PASSPORTNUM": {
177
+ "precision": 1.0,
178
+ "recall": 1.0,
179
+ "f1": 1.0,
180
+ "tp": 20,
181
+ "fp": 0,
182
+ "fn": 0
183
+ },
184
+ "PHONE": {
185
+ "precision": 1.0,
186
+ "recall": 1.0,
187
+ "f1": 1.0,
188
+ "tp": 801,
189
+ "fp": 0,
190
+ "fn": 0
191
+ },
192
+ "POSITION": {
193
+ "precision": 0.9848866498740554,
194
+ "recall": 0.972636815920398,
195
+ "f1": 0.9787234042553191,
196
+ "tp": 782,
197
+ "fp": 12,
198
+ "fn": 22
199
+ },
200
+ "RELIGION": {
201
+ "precision": 1.0,
202
+ "recall": 1.0,
203
+ "f1": 1.0,
204
+ "tp": 218,
205
+ "fp": 0,
206
+ "fn": 0
207
+ },
208
+ "RRN": {
209
+ "precision": 0.8245614035087719,
210
+ "recall": 1.0,
211
+ "f1": 0.9038461538461539,
212
+ "tp": 47,
213
+ "fp": 10,
214
+ "fn": 0
215
+ },
216
+ "SCHOOL": {
217
+ "precision": 0.9230769230769231,
218
+ "recall": 1.0,
219
+ "f1": 0.9600000000000001,
220
+ "tp": 24,
221
+ "fp": 2,
222
+ "fn": 0
223
+ },
224
+ "URL": {
225
+ "precision": 1.0,
226
+ "recall": 1.0,
227
+ "f1": 1.0,
228
+ "tp": 28,
229
+ "fp": 0,
230
+ "fn": 0
231
+ },
232
+ "USER_ID": {
233
+ "precision": 1.0,
234
+ "recall": 1.0,
235
+ "f1": 1.0,
236
+ "tp": 34,
237
+ "fp": 0,
238
+ "fn": 0
239
+ },
240
+ "VEHICLE_NUMBER": {
241
+ "precision": 0.9722222222222222,
242
+ "recall": 1.0,
243
+ "f1": 0.9859154929577464,
244
+ "tp": 35,
245
+ "fp": 1,
246
+ "fn": 0
247
+ },
248
+ "WEIGHT": {
249
+ "precision": 1.0,
250
+ "recall": 1.0,
251
+ "f1": 1.0,
252
+ "tp": 218,
253
+ "fp": 0,
254
+ "fn": 0
255
+ },
256
+ "WORKPLACE": {
257
+ "precision": 0.9874213836477987,
258
+ "recall": 0.9763681592039801,
259
+ "f1": 0.9818636647904941,
260
+ "tp": 785,
261
+ "fp": 10,
262
+ "fn": 19
263
+ },
264
+ "ZIPCODE": {
265
+ "precision": 0.9,
266
+ "recall": 1.0,
267
+ "f1": 0.9473684210526316,
268
+ "tp": 18,
269
+ "fp": 2,
270
+ "fn": 0
271
+ }
272
+ }
273
+ }
class_weights.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "O": 0.1,
3
+ "B-ACCOUNT_NUMBER": 1.133483,
4
+ "I-ACCOUNT_NUMBER": 0.406703,
5
+ "B-ADDRESS": 1.02502,
6
+ "I-ADDRESS": 0.409288,
7
+ "B-AGE": 1.095445,
8
+ "I-AGE": 1.095445,
9
+ "B-ALIEN_NUMBER": 1.112518,
10
+ "I-ALIEN_NUMBER": 0.419927,
11
+ "B-BIRTHDATE": 1.430029,
12
+ "I-BIRTHDATE": 0.639418,
13
+ "B-BLOOD_TYPE": 1.51058,
14
+ "I-BLOOD_TYPE": 1.507673,
15
+ "B-CARD_NUMBER": 1.052511,
16
+ "I-CARD_NUMBER": 0.333978,
17
+ "B-CITY": 1.504782,
18
+ "I-CITY": 3.820025,
19
+ "B-DEPARTMENT": 1.137197,
20
+ "I-DEPARTMENT": 0.910754,
21
+ "B-DRIVER_LICENSE": 1.106739,
22
+ "I-DRIVER_LICENSE": 0.395149,
23
+ "B-EMAIL": 0.597217,
24
+ "I-EMAIL": 0.25,
25
+ "B-EMPLOYEE_ID": 1.071248,
26
+ "I-EMPLOYEE_ID": 0.476127,
27
+ "B-GENDER": 1.091573,
28
+ "I-GENDER": 6.0,
29
+ "B-HEIGHT": 1.5299,
30
+ "I-HEIGHT": 1.005115,
31
+ "B-IP_ADDRESS": 1.142837,
32
+ "I-IP_ADDRESS": 0.399042,
33
+ "B-MAJOR": 0.809451,
34
+ "I-MAJOR": 0.538838,
35
+ "B-MEMBER_ID": 1.085571,
36
+ "I-MEMBER_ID": 0.44862,
37
+ "B-NAME": 0.349583,
38
+ "I-NAME": 0.27994,
39
+ "B-NATIONALITY": 1.143469,
40
+ "I-NATIONALITY": 6.0,
41
+ "B-NICKNAME": 0.805222,
42
+ "I-NICKNAME": 0.49459,
43
+ "B-PARTICIPANT_ID": 1.139693,
44
+ "I-PARTICIPANT_ID": 0.473142,
45
+ "B-PASSPORTNUM": 1.041825,
46
+ "I-PASSPORTNUM": 0.47631,
47
+ "B-PHONE": 0.737529,
48
+ "I-PHONE": 0.302128,
49
+ "B-POSITION": 1.131026,
50
+ "I-POSITION": 2.111961,
51
+ "B-RELIGION": 1.51058,
52
+ "I-RELIGION": 3.99662,
53
+ "B-RRN": 1.061502,
54
+ "I-RRN": 0.408479,
55
+ "B-SCHOOL": 0.809451,
56
+ "I-SCHOOL": 0.521451,
57
+ "B-URL": 0.807662,
58
+ "I-URL": 0.25,
59
+ "B-USER_ID": 1.139693,
60
+ "I-USER_ID": 0.565513,
61
+ "B-VEHICLE_NUMBER": 1.135955,
62
+ "I-VEHICLE_NUMBER": 0.623141,
63
+ "B-WEIGHT": 1.5299,
64
+ "I-WEIGHT": 1.0,
65
+ "B-WORKPLACE": 1.123753,
66
+ "I-WORKPLACE": 0.883288,
67
+ "B-ZIPCODE": 1.020495,
68
+ "I-ZIPCODE": 0.809675
69
+ }
config.json ADDED
@@ -0,0 +1,179 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "adam_epsilon": 1e-08,
3
+ "add_cross_attention": false,
4
+ "architectures": [
5
+ "BertForTokenClassification"
6
+ ],
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "batch_size": 32,
9
+ "bos_token_id": null,
10
+ "classifier_dropout": null,
11
+ "device": "cuda",
12
+ "dtype": "float32",
13
+ "eos_token_id": null,
14
+ "epoch": 30,
15
+ "hidden_act": "gelu",
16
+ "hidden_dropout_prob": 0.1,
17
+ "hidden_size": 768,
18
+ "id2label": {
19
+ "0": "O",
20
+ "1": "B-ACCOUNT_NUMBER",
21
+ "2": "I-ACCOUNT_NUMBER",
22
+ "3": "B-ADDRESS",
23
+ "4": "I-ADDRESS",
24
+ "5": "B-AGE",
25
+ "6": "I-AGE",
26
+ "7": "B-ALIEN_NUMBER",
27
+ "8": "I-ALIEN_NUMBER",
28
+ "9": "B-BIRTHDATE",
29
+ "10": "I-BIRTHDATE",
30
+ "11": "B-BLOOD_TYPE",
31
+ "12": "I-BLOOD_TYPE",
32
+ "13": "B-CARD_NUMBER",
33
+ "14": "I-CARD_NUMBER",
34
+ "15": "B-CITY",
35
+ "16": "I-CITY",
36
+ "17": "B-DEPARTMENT",
37
+ "18": "I-DEPARTMENT",
38
+ "19": "B-DRIVER_LICENSE",
39
+ "20": "I-DRIVER_LICENSE",
40
+ "21": "B-EMAIL",
41
+ "22": "I-EMAIL",
42
+ "23": "B-EMPLOYEE_ID",
43
+ "24": "I-EMPLOYEE_ID",
44
+ "25": "B-GENDER",
45
+ "26": "I-GENDER",
46
+ "27": "B-HEIGHT",
47
+ "28": "I-HEIGHT",
48
+ "29": "B-IP_ADDRESS",
49
+ "30": "I-IP_ADDRESS",
50
+ "31": "B-MAJOR",
51
+ "32": "I-MAJOR",
52
+ "33": "B-MEMBER_ID",
53
+ "34": "I-MEMBER_ID",
54
+ "35": "B-NAME",
55
+ "36": "I-NAME",
56
+ "37": "B-NATIONALITY",
57
+ "38": "I-NATIONALITY",
58
+ "39": "B-NICKNAME",
59
+ "40": "I-NICKNAME",
60
+ "41": "B-PARTICIPANT_ID",
61
+ "42": "I-PARTICIPANT_ID",
62
+ "43": "B-PASSPORTNUM",
63
+ "44": "I-PASSPORTNUM",
64
+ "45": "B-PHONE",
65
+ "46": "I-PHONE",
66
+ "47": "B-POSITION",
67
+ "48": "I-POSITION",
68
+ "49": "B-RELIGION",
69
+ "50": "I-RELIGION",
70
+ "51": "B-RRN",
71
+ "52": "I-RRN",
72
+ "53": "B-SCHOOL",
73
+ "54": "I-SCHOOL",
74
+ "55": "B-URL",
75
+ "56": "I-URL",
76
+ "57": "B-USER_ID",
77
+ "58": "I-USER_ID",
78
+ "59": "B-VEHICLE_NUMBER",
79
+ "60": "I-VEHICLE_NUMBER",
80
+ "61": "B-WEIGHT",
81
+ "62": "I-WEIGHT",
82
+ "63": "B-WORKPLACE",
83
+ "64": "I-WORKPLACE",
84
+ "65": "B-ZIPCODE",
85
+ "66": "I-ZIPCODE"
86
+ },
87
+ "initializer_range": 0.02,
88
+ "intermediate_hidden_size": 768,
89
+ "intermediate_size": 3072,
90
+ "is_decoder": false,
91
+ "label2id": {
92
+ "B-ACCOUNT_NUMBER": 1,
93
+ "B-ADDRESS": 3,
94
+ "B-AGE": 5,
95
+ "B-ALIEN_NUMBER": 7,
96
+ "B-BIRTHDATE": 9,
97
+ "B-BLOOD_TYPE": 11,
98
+ "B-CARD_NUMBER": 13,
99
+ "B-CITY": 15,
100
+ "B-DEPARTMENT": 17,
101
+ "B-DRIVER_LICENSE": 19,
102
+ "B-EMAIL": 21,
103
+ "B-EMPLOYEE_ID": 23,
104
+ "B-GENDER": 25,
105
+ "B-HEIGHT": 27,
106
+ "B-IP_ADDRESS": 29,
107
+ "B-MAJOR": 31,
108
+ "B-MEMBER_ID": 33,
109
+ "B-NAME": 35,
110
+ "B-NATIONALITY": 37,
111
+ "B-NICKNAME": 39,
112
+ "B-PARTICIPANT_ID": 41,
113
+ "B-PASSPORTNUM": 43,
114
+ "B-PHONE": 45,
115
+ "B-POSITION": 47,
116
+ "B-RELIGION": 49,
117
+ "B-RRN": 51,
118
+ "B-SCHOOL": 53,
119
+ "B-URL": 55,
120
+ "B-USER_ID": 57,
121
+ "B-VEHICLE_NUMBER": 59,
122
+ "B-WEIGHT": 61,
123
+ "B-WORKPLACE": 63,
124
+ "B-ZIPCODE": 65,
125
+ "I-ACCOUNT_NUMBER": 2,
126
+ "I-ADDRESS": 4,
127
+ "I-AGE": 6,
128
+ "I-ALIEN_NUMBER": 8,
129
+ "I-BIRTHDATE": 10,
130
+ "I-BLOOD_TYPE": 12,
131
+ "I-CARD_NUMBER": 14,
132
+ "I-CITY": 16,
133
+ "I-DEPARTMENT": 18,
134
+ "I-DRIVER_LICENSE": 20,
135
+ "I-EMAIL": 22,
136
+ "I-EMPLOYEE_ID": 24,
137
+ "I-GENDER": 26,
138
+ "I-HEIGHT": 28,
139
+ "I-IP_ADDRESS": 30,
140
+ "I-MAJOR": 32,
141
+ "I-MEMBER_ID": 34,
142
+ "I-NAME": 36,
143
+ "I-NATIONALITY": 38,
144
+ "I-NICKNAME": 40,
145
+ "I-PARTICIPANT_ID": 42,
146
+ "I-PASSPORTNUM": 44,
147
+ "I-PHONE": 46,
148
+ "I-POSITION": 48,
149
+ "I-RELIGION": 50,
150
+ "I-RRN": 52,
151
+ "I-SCHOOL": 54,
152
+ "I-URL": 56,
153
+ "I-USER_ID": 58,
154
+ "I-VEHICLE_NUMBER": 60,
155
+ "I-WEIGHT": 62,
156
+ "I-WORKPLACE": 64,
157
+ "I-ZIPCODE": 66,
158
+ "O": 0
159
+ },
160
+ "layer_norm_eps": 1e-12,
161
+ "learning_rate": 0.005,
162
+ "max_grad_norm": 1.0,
163
+ "max_position_embeddings": 512,
164
+ "max_seq_len": 120,
165
+ "model_name": "kpfbert",
166
+ "model_type": "bert",
167
+ "num_attention_heads": 12,
168
+ "num_hidden_layers": 12,
169
+ "pad_token_id": 0,
170
+ "position_embedding_type": "absolute",
171
+ "seed": 1000,
172
+ "test_data": "21_150tags_EntityLinking",
173
+ "tie_word_embeddings": true,
174
+ "train_data": "21_150tags_EntityLinking",
175
+ "transformers_version": "5.14.1",
176
+ "type_vocab_size": 2,
177
+ "use_cache": true,
178
+ "vocab_size": 36440
179
+ }
label_map.json ADDED
@@ -0,0 +1,140 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "label2id": {
3
+ "O": 0,
4
+ "B-ACCOUNT_NUMBER": 1,
5
+ "I-ACCOUNT_NUMBER": 2,
6
+ "B-ADDRESS": 3,
7
+ "I-ADDRESS": 4,
8
+ "B-AGE": 5,
9
+ "I-AGE": 6,
10
+ "B-ALIEN_NUMBER": 7,
11
+ "I-ALIEN_NUMBER": 8,
12
+ "B-BIRTHDATE": 9,
13
+ "I-BIRTHDATE": 10,
14
+ "B-BLOOD_TYPE": 11,
15
+ "I-BLOOD_TYPE": 12,
16
+ "B-CARD_NUMBER": 13,
17
+ "I-CARD_NUMBER": 14,
18
+ "B-CITY": 15,
19
+ "I-CITY": 16,
20
+ "B-DEPARTMENT": 17,
21
+ "I-DEPARTMENT": 18,
22
+ "B-DRIVER_LICENSE": 19,
23
+ "I-DRIVER_LICENSE": 20,
24
+ "B-EMAIL": 21,
25
+ "I-EMAIL": 22,
26
+ "B-EMPLOYEE_ID": 23,
27
+ "I-EMPLOYEE_ID": 24,
28
+ "B-GENDER": 25,
29
+ "I-GENDER": 26,
30
+ "B-HEIGHT": 27,
31
+ "I-HEIGHT": 28,
32
+ "B-IP_ADDRESS": 29,
33
+ "I-IP_ADDRESS": 30,
34
+ "B-MAJOR": 31,
35
+ "I-MAJOR": 32,
36
+ "B-MEMBER_ID": 33,
37
+ "I-MEMBER_ID": 34,
38
+ "B-NAME": 35,
39
+ "I-NAME": 36,
40
+ "B-NATIONALITY": 37,
41
+ "I-NATIONALITY": 38,
42
+ "B-NICKNAME": 39,
43
+ "I-NICKNAME": 40,
44
+ "B-PARTICIPANT_ID": 41,
45
+ "I-PARTICIPANT_ID": 42,
46
+ "B-PASSPORTNUM": 43,
47
+ "I-PASSPORTNUM": 44,
48
+ "B-PHONE": 45,
49
+ "I-PHONE": 46,
50
+ "B-POSITION": 47,
51
+ "I-POSITION": 48,
52
+ "B-RELIGION": 49,
53
+ "I-RELIGION": 50,
54
+ "B-RRN": 51,
55
+ "I-RRN": 52,
56
+ "B-SCHOOL": 53,
57
+ "I-SCHOOL": 54,
58
+ "B-URL": 55,
59
+ "I-URL": 56,
60
+ "B-USER_ID": 57,
61
+ "I-USER_ID": 58,
62
+ "B-VEHICLE_NUMBER": 59,
63
+ "I-VEHICLE_NUMBER": 60,
64
+ "B-WEIGHT": 61,
65
+ "I-WEIGHT": 62,
66
+ "B-WORKPLACE": 63,
67
+ "I-WORKPLACE": 64,
68
+ "B-ZIPCODE": 65,
69
+ "I-ZIPCODE": 66
70
+ },
71
+ "id2label": {
72
+ "0": "O",
73
+ "1": "B-ACCOUNT_NUMBER",
74
+ "2": "I-ACCOUNT_NUMBER",
75
+ "3": "B-ADDRESS",
76
+ "4": "I-ADDRESS",
77
+ "5": "B-AGE",
78
+ "6": "I-AGE",
79
+ "7": "B-ALIEN_NUMBER",
80
+ "8": "I-ALIEN_NUMBER",
81
+ "9": "B-BIRTHDATE",
82
+ "10": "I-BIRTHDATE",
83
+ "11": "B-BLOOD_TYPE",
84
+ "12": "I-BLOOD_TYPE",
85
+ "13": "B-CARD_NUMBER",
86
+ "14": "I-CARD_NUMBER",
87
+ "15": "B-CITY",
88
+ "16": "I-CITY",
89
+ "17": "B-DEPARTMENT",
90
+ "18": "I-DEPARTMENT",
91
+ "19": "B-DRIVER_LICENSE",
92
+ "20": "I-DRIVER_LICENSE",
93
+ "21": "B-EMAIL",
94
+ "22": "I-EMAIL",
95
+ "23": "B-EMPLOYEE_ID",
96
+ "24": "I-EMPLOYEE_ID",
97
+ "25": "B-GENDER",
98
+ "26": "I-GENDER",
99
+ "27": "B-HEIGHT",
100
+ "28": "I-HEIGHT",
101
+ "29": "B-IP_ADDRESS",
102
+ "30": "I-IP_ADDRESS",
103
+ "31": "B-MAJOR",
104
+ "32": "I-MAJOR",
105
+ "33": "B-MEMBER_ID",
106
+ "34": "I-MEMBER_ID",
107
+ "35": "B-NAME",
108
+ "36": "I-NAME",
109
+ "37": "B-NATIONALITY",
110
+ "38": "I-NATIONALITY",
111
+ "39": "B-NICKNAME",
112
+ "40": "I-NICKNAME",
113
+ "41": "B-PARTICIPANT_ID",
114
+ "42": "I-PARTICIPANT_ID",
115
+ "43": "B-PASSPORTNUM",
116
+ "44": "I-PASSPORTNUM",
117
+ "45": "B-PHONE",
118
+ "46": "I-PHONE",
119
+ "47": "B-POSITION",
120
+ "48": "I-POSITION",
121
+ "49": "B-RELIGION",
122
+ "50": "I-RELIGION",
123
+ "51": "B-RRN",
124
+ "52": "I-RRN",
125
+ "53": "B-SCHOOL",
126
+ "54": "I-SCHOOL",
127
+ "55": "B-URL",
128
+ "56": "I-URL",
129
+ "57": "B-USER_ID",
130
+ "58": "I-USER_ID",
131
+ "59": "B-VEHICLE_NUMBER",
132
+ "60": "I-VEHICLE_NUMBER",
133
+ "61": "B-WEIGHT",
134
+ "62": "I-WEIGHT",
135
+ "63": "B-WORKPLACE",
136
+ "64": "I-WORKPLACE",
137
+ "65": "B-ZIPCODE",
138
+ "66": "I-ZIPCODE"
139
+ }
140
+ }
label_thresholds_calibration_v1.json ADDED
@@ -0,0 +1,403 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "split": "threshold_calibration",
3
+ "model": "kpf_bert_pii_ner_model_final",
4
+ "method": "entity confidence; low=FP p95, high=TP p05",
5
+ "labels": {
6
+ "ACCOUNT_NUMBER": {
7
+ "gold": 20,
8
+ "pred": 19,
9
+ "missed": 5,
10
+ "tp_count": 15,
11
+ "fp_count": 4,
12
+ "tp_p05": 0.33447131514549255,
13
+ "tp_median": 0.45647820830345154,
14
+ "fp_p95": 0.793274462223053,
15
+ "low_candidate": 0.793274462223053,
16
+ "high_candidate": 0.33447131514549255
17
+ },
18
+ "ADDRESS": {
19
+ "gold": 20,
20
+ "pred": 20,
21
+ "missed": 0,
22
+ "tp_count": 20,
23
+ "fp_count": 0,
24
+ "tp_p05": 0.9934954643249512,
25
+ "tp_median": 0.9978099465370178,
26
+ "fp_p95": null,
27
+ "low_candidate": 0.35,
28
+ "high_candidate": 0.9934954643249512
29
+ },
30
+ "AGE": {
31
+ "gold": 20,
32
+ "pred": 21,
33
+ "missed": 20,
34
+ "tp_count": 0,
35
+ "fp_count": 21,
36
+ "tp_p05": null,
37
+ "tp_median": null,
38
+ "fp_p95": 0.824558675289154,
39
+ "low_candidate": 0.824558675289154,
40
+ "high_candidate": 0.85
41
+ },
42
+ "ALIEN_NUMBER": {
43
+ "gold": 20,
44
+ "pred": 3,
45
+ "missed": 20,
46
+ "tp_count": 0,
47
+ "fp_count": 3,
48
+ "tp_p05": null,
49
+ "tp_median": null,
50
+ "fp_p95": 0.332609087228775,
51
+ "low_candidate": 0.332609087228775,
52
+ "high_candidate": 0.85
53
+ },
54
+ "BIRTHDATE": {
55
+ "gold": 20,
56
+ "pred": 10,
57
+ "missed": 18,
58
+ "tp_count": 2,
59
+ "fp_count": 8,
60
+ "tp_p05": 0.18990333378314972,
61
+ "tp_median": 0.18990333378314972,
62
+ "fp_p95": 0.512396514415741,
63
+ "low_candidate": 0.512396514415741,
64
+ "high_candidate": 0.18990333378314972
65
+ },
66
+ "BLOOD_TYPE": {
67
+ "gold": 20,
68
+ "pred": 22,
69
+ "missed": 4,
70
+ "tp_count": 16,
71
+ "fp_count": 6,
72
+ "tp_p05": 0.8868210315704346,
73
+ "tp_median": 0.9296256303787231,
74
+ "fp_p95": 0.9648946523666382,
75
+ "low_candidate": 0.9648946523666382,
76
+ "high_candidate": 0.8868210315704346
77
+ },
78
+ "CARD_NUMBER": {
79
+ "gold": 20,
80
+ "pred": 20,
81
+ "missed": 0,
82
+ "tp_count": 20,
83
+ "fp_count": 0,
84
+ "tp_p05": 0.9542779326438904,
85
+ "tp_median": 0.9900541305541992,
86
+ "fp_p95": null,
87
+ "low_candidate": 0.35,
88
+ "high_candidate": 0.9542779326438904
89
+ },
90
+ "CITY": {
91
+ "gold": 20,
92
+ "pred": 20,
93
+ "missed": 0,
94
+ "tp_count": 20,
95
+ "fp_count": 0,
96
+ "tp_p05": 0.6697056293487549,
97
+ "tp_median": 0.9951282739639282,
98
+ "fp_p95": null,
99
+ "low_candidate": 0.35,
100
+ "high_candidate": 0.6697056293487549
101
+ },
102
+ "DEPARTMENT": {
103
+ "gold": 20,
104
+ "pred": 20,
105
+ "missed": 0,
106
+ "tp_count": 20,
107
+ "fp_count": 0,
108
+ "tp_p05": 0.9642444849014282,
109
+ "tp_median": 0.9694795608520508,
110
+ "fp_p95": null,
111
+ "low_candidate": 0.35,
112
+ "high_candidate": 0.9642444849014282
113
+ },
114
+ "DRIVER_LICENSE": {
115
+ "gold": 20,
116
+ "pred": 20,
117
+ "missed": 0,
118
+ "tp_count": 20,
119
+ "fp_count": 0,
120
+ "tp_p05": 0.936443567276001,
121
+ "tp_median": 0.9880808591842651,
122
+ "fp_p95": null,
123
+ "low_candidate": 0.35,
124
+ "high_candidate": 0.936443567276001
125
+ },
126
+ "EMAIL": {
127
+ "gold": 20,
128
+ "pred": 32,
129
+ "missed": 0,
130
+ "tp_count": 20,
131
+ "fp_count": 12,
132
+ "tp_p05": 0.9263460040092468,
133
+ "tp_median": 0.995662271976471,
134
+ "fp_p95": 0.898489236831665,
135
+ "low_candidate": 0.898489236831665,
136
+ "high_candidate": 0.9263460040092468
137
+ },
138
+ "EMPLOYEE_ID": {
139
+ "gold": 20,
140
+ "pred": 41,
141
+ "missed": 17,
142
+ "tp_count": 3,
143
+ "fp_count": 38,
144
+ "tp_p05": 0.57096266746521,
145
+ "tp_median": 0.8768562078475952,
146
+ "fp_p95": 0.9567230343818665,
147
+ "low_candidate": 0.9567230343818665,
148
+ "high_candidate": 0.57096266746521
149
+ },
150
+ "GENDER": {
151
+ "gold": 20,
152
+ "pred": 20,
153
+ "missed": 0,
154
+ "tp_count": 20,
155
+ "fp_count": 0,
156
+ "tp_p05": 0.9979225993156433,
157
+ "tp_median": 0.9984620809555054,
158
+ "fp_p95": null,
159
+ "low_candidate": 0.35,
160
+ "high_candidate": 0.9979225993156433
161
+ },
162
+ "HEIGHT": {
163
+ "gold": 20,
164
+ "pred": 0,
165
+ "missed": 20,
166
+ "tp_count": 0,
167
+ "fp_count": 0,
168
+ "tp_p05": null,
169
+ "tp_median": null,
170
+ "fp_p95": null,
171
+ "low_candidate": 0.35,
172
+ "high_candidate": 0.85
173
+ },
174
+ "IP_ADDRESS": {
175
+ "gold": 20,
176
+ "pred": 5,
177
+ "missed": 20,
178
+ "tp_count": 0,
179
+ "fp_count": 5,
180
+ "tp_p05": null,
181
+ "tp_median": null,
182
+ "fp_p95": 0.4374067187309265,
183
+ "low_candidate": 0.4374067187309265,
184
+ "high_candidate": 0.85
185
+ },
186
+ "MAJOR": {
187
+ "gold": 20,
188
+ "pred": 20,
189
+ "missed": 0,
190
+ "tp_count": 20,
191
+ "fp_count": 0,
192
+ "tp_p05": 0.9973071813583374,
193
+ "tp_median": 0.9990133047103882,
194
+ "fp_p95": null,
195
+ "low_candidate": 0.35,
196
+ "high_candidate": 0.9973071813583374
197
+ },
198
+ "MEMBER_ID": {
199
+ "gold": 20,
200
+ "pred": 35,
201
+ "missed": 17,
202
+ "tp_count": 3,
203
+ "fp_count": 32,
204
+ "tp_p05": 0.7067432403564453,
205
+ "tp_median": 0.7406010031700134,
206
+ "fp_p95": 0.7461152672767639,
207
+ "low_candidate": 0.7461152672767639,
208
+ "high_candidate": 0.7067432403564453
209
+ },
210
+ "NAME": {
211
+ "gold": 25,
212
+ "pred": 25,
213
+ "missed": 0,
214
+ "tp_count": 25,
215
+ "fp_count": 0,
216
+ "tp_p05": 0.9995038509368896,
217
+ "tp_median": 0.9998273253440857,
218
+ "fp_p95": null,
219
+ "low_candidate": 0.35,
220
+ "high_candidate": 0.9995038509368896
221
+ },
222
+ "NATIONALITY": {
223
+ "gold": 20,
224
+ "pred": 16,
225
+ "missed": 4,
226
+ "tp_count": 16,
227
+ "fp_count": 0,
228
+ "tp_p05": 0.911325991153717,
229
+ "tp_median": 0.9334752559661865,
230
+ "fp_p95": null,
231
+ "low_candidate": 0.35,
232
+ "high_candidate": 0.911325991153717
233
+ },
234
+ "NICKNAME": {
235
+ "gold": 20,
236
+ "pred": 40,
237
+ "missed": 0,
238
+ "tp_count": 20,
239
+ "fp_count": 20,
240
+ "tp_p05": 0.9965251088142395,
241
+ "tp_median": 0.9993723034858704,
242
+ "fp_p95": 0.9586448073387146,
243
+ "low_candidate": 0.9586448073387146,
244
+ "high_candidate": 0.9965251088142395
245
+ },
246
+ "PARTICIPANT_ID": {
247
+ "gold": 20,
248
+ "pred": 31,
249
+ "missed": 11,
250
+ "tp_count": 9,
251
+ "fp_count": 22,
252
+ "tp_p05": 0.49748900532722473,
253
+ "tp_median": 0.6439980268478394,
254
+ "fp_p95": 0.9855802655220032,
255
+ "low_candidate": 0.9855802655220032,
256
+ "high_candidate": 0.49748900532722473
257
+ },
258
+ "PASSPORTNUM": {
259
+ "gold": 20,
260
+ "pred": 20,
261
+ "missed": 2,
262
+ "tp_count": 18,
263
+ "fp_count": 2,
264
+ "tp_p05": 0.8496622443199158,
265
+ "tp_median": 0.9994340538978577,
266
+ "fp_p95": 0.4733753502368927,
267
+ "low_candidate": 0.4733753502368927,
268
+ "high_candidate": 0.8496622443199158
269
+ },
270
+ "PHONE": {
271
+ "gold": 20,
272
+ "pred": 22,
273
+ "missed": 0,
274
+ "tp_count": 20,
275
+ "fp_count": 2,
276
+ "tp_p05": 0.9992443323135376,
277
+ "tp_median": 0.9996128678321838,
278
+ "fp_p95": 0.8399693965911865,
279
+ "low_candidate": 0.8399693965911865,
280
+ "high_candidate": 0.9992443323135376
281
+ },
282
+ "POSITION": {
283
+ "gold": 20,
284
+ "pred": 16,
285
+ "missed": 4,
286
+ "tp_count": 16,
287
+ "fp_count": 0,
288
+ "tp_p05": 0.9981757402420044,
289
+ "tp_median": 0.9984045624732971,
290
+ "fp_p95": null,
291
+ "low_candidate": 0.35,
292
+ "high_candidate": 0.9981757402420044
293
+ },
294
+ "RELIGION": {
295
+ "gold": 20,
296
+ "pred": 20,
297
+ "missed": 0,
298
+ "tp_count": 20,
299
+ "fp_count": 0,
300
+ "tp_p05": 0.98134845495224,
301
+ "tp_median": 0.9906643629074097,
302
+ "fp_p95": null,
303
+ "low_candidate": 0.35,
304
+ "high_candidate": 0.98134845495224
305
+ },
306
+ "RRN": {
307
+ "gold": 20,
308
+ "pred": 19,
309
+ "missed": 6,
310
+ "tp_count": 14,
311
+ "fp_count": 5,
312
+ "tp_p05": 0.7186772227287292,
313
+ "tp_median": 0.7683890461921692,
314
+ "fp_p95": 0.6961906552314758,
315
+ "low_candidate": 0.6961906552314758,
316
+ "high_candidate": 0.7186772227287292
317
+ },
318
+ "SCHOOL": {
319
+ "gold": 20,
320
+ "pred": 20,
321
+ "missed": 0,
322
+ "tp_count": 20,
323
+ "fp_count": 0,
324
+ "tp_p05": 0.999293327331543,
325
+ "tp_median": 0.9995866417884827,
326
+ "fp_p95": null,
327
+ "low_candidate": 0.35,
328
+ "high_candidate": 0.999293327331543
329
+ },
330
+ "URL": {
331
+ "gold": 20,
332
+ "pred": 0,
333
+ "missed": 20,
334
+ "tp_count": 0,
335
+ "fp_count": 0,
336
+ "tp_p05": null,
337
+ "tp_median": null,
338
+ "fp_p95": null,
339
+ "low_candidate": 0.35,
340
+ "high_candidate": 0.85
341
+ },
342
+ "USER_ID": {
343
+ "gold": 20,
344
+ "pred": 5,
345
+ "missed": 20,
346
+ "tp_count": 0,
347
+ "fp_count": 5,
348
+ "tp_p05": null,
349
+ "tp_median": null,
350
+ "fp_p95": 0.6109357476234436,
351
+ "low_candidate": 0.6109357476234436,
352
+ "high_candidate": 0.85
353
+ },
354
+ "VEHICLE_NUMBER": {
355
+ "gold": 20,
356
+ "pred": 1,
357
+ "missed": 20,
358
+ "tp_count": 0,
359
+ "fp_count": 1,
360
+ "tp_p05": null,
361
+ "tp_median": null,
362
+ "fp_p95": 0.473734050989151,
363
+ "low_candidate": 0.473734050989151,
364
+ "high_candidate": 0.85
365
+ },
366
+ "WEIGHT": {
367
+ "gold": 20,
368
+ "pred": 0,
369
+ "missed": 20,
370
+ "tp_count": 0,
371
+ "fp_count": 0,
372
+ "tp_p05": null,
373
+ "tp_median": null,
374
+ "fp_p95": null,
375
+ "low_candidate": 0.35,
376
+ "high_candidate": 0.85
377
+ },
378
+ "WORKPLACE": {
379
+ "gold": 20,
380
+ "pred": 20,
381
+ "missed": 0,
382
+ "tp_count": 20,
383
+ "fp_count": 0,
384
+ "tp_p05": 0.9837009310722351,
385
+ "tp_median": 0.991530179977417,
386
+ "fp_p95": null,
387
+ "low_candidate": 0.35,
388
+ "high_candidate": 0.9837009310722351
389
+ },
390
+ "ZIPCODE": {
391
+ "gold": 20,
392
+ "pred": 3,
393
+ "missed": 20,
394
+ "tp_count": 0,
395
+ "fp_count": 3,
396
+ "tp_p05": null,
397
+ "tp_median": null,
398
+ "fp_p95": 0.6629543304443359,
399
+ "low_candidate": 0.6629543304443359,
400
+ "high_candidate": 0.85
401
+ }
402
+ }
403
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b222d4635f26888cd57596ccd005104e48712f0b460149fafb499d6434ff3cc5
3
+ size 453976140
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "cls_token": "[CLS]",
4
+ "do_basic_tokenize": true,
5
+ "do_lower_case": false,
6
+ "is_local": true,
7
+ "local_files_only": false,
8
+ "mask_token": "[MASK]",
9
+ "model_max_length": 1000000000000000019884624838656,
10
+ "never_split": null,
11
+ "pad_token": "[PAD]",
12
+ "sep_token": "[SEP]",
13
+ "strip_accents": null,
14
+ "tokenize_chinese_chars": true,
15
+ "tokenizer_class": "BertTokenizer",
16
+ "unk_token": "[UNK]"
17
+ }
training_history.json ADDED
@@ -0,0 +1,821 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "epoch": 1,
4
+ "train_loss": 0.005999320498434073,
5
+ "micro_precision": 0.9850155499010461,
6
+ "micro_recall": 0.987808335696059,
7
+ "micro_f1": 0.9864099660249152,
8
+ "per_label": {
9
+ "ACCOUNT_NUMBER": {
10
+ "precision": 0.918918918918919,
11
+ "recall": 1.0,
12
+ "f1": 0.9577464788732395,
13
+ "tp": 68,
14
+ "fp": 6,
15
+ "fn": 0
16
+ },
17
+ "ADDRESS": {
18
+ "precision": 0.8260869565217391,
19
+ "recall": 0.9047619047619048,
20
+ "f1": 0.8636363636363636,
21
+ "tp": 19,
22
+ "fp": 4,
23
+ "fn": 2
24
+ },
25
+ "AGE": {
26
+ "precision": 0.9428571428571428,
27
+ "recall": 0.9705882352941176,
28
+ "f1": 0.9565217391304348,
29
+ "tp": 33,
30
+ "fp": 2,
31
+ "fn": 1
32
+ },
33
+ "ALIEN_NUMBER": {
34
+ "precision": 1.0,
35
+ "recall": 0.5,
36
+ "f1": 0.6666666666666666,
37
+ "tp": 10,
38
+ "fp": 0,
39
+ "fn": 10
40
+ },
41
+ "BIRTHDATE": {
42
+ "precision": 0.9961685823754789,
43
+ "recall": 1.0,
44
+ "f1": 0.9980806142034548,
45
+ "tp": 260,
46
+ "fp": 1,
47
+ "fn": 0
48
+ },
49
+ "BLOOD_TYPE": {
50
+ "precision": 0.9819819819819819,
51
+ "recall": 1.0,
52
+ "f1": 0.9909090909090909,
53
+ "tp": 218,
54
+ "fp": 4,
55
+ "fn": 0
56
+ },
57
+ "CARD_NUMBER": {
58
+ "precision": 1.0,
59
+ "recall": 1.0,
60
+ "f1": 1.0,
61
+ "tp": 19,
62
+ "fp": 0,
63
+ "fn": 0
64
+ },
65
+ "CITY": {
66
+ "precision": 0.9988571428571429,
67
+ "recall": 1.0,
68
+ "f1": 0.9994282447112636,
69
+ "tp": 874,
70
+ "fp": 1,
71
+ "fn": 0
72
+ },
73
+ "DEPARTMENT": {
74
+ "precision": 0.9688667496886675,
75
+ "recall": 0.9676616915422885,
76
+ "f1": 0.9682638456751711,
77
+ "tp": 778,
78
+ "fp": 25,
79
+ "fn": 26
80
+ },
81
+ "DRIVER_LICENSE": {
82
+ "precision": 1.0,
83
+ "recall": 1.0,
84
+ "f1": 1.0,
85
+ "tp": 20,
86
+ "fp": 0,
87
+ "fn": 0
88
+ },
89
+ "EMAIL": {
90
+ "precision": 1.0,
91
+ "recall": 1.0,
92
+ "f1": 1.0,
93
+ "tp": 99,
94
+ "fp": 0,
95
+ "fn": 0
96
+ },
97
+ "EMPLOYEE_ID": {
98
+ "precision": 1.0,
99
+ "recall": 1.0,
100
+ "f1": 1.0,
101
+ "tp": 42,
102
+ "fp": 0,
103
+ "fn": 0
104
+ },
105
+ "GENDER": {
106
+ "precision": 1.0,
107
+ "recall": 0.9705882352941176,
108
+ "f1": 0.9850746268656716,
109
+ "tp": 33,
110
+ "fp": 0,
111
+ "fn": 1
112
+ },
113
+ "HEIGHT": {
114
+ "precision": 1.0,
115
+ "recall": 1.0,
116
+ "f1": 1.0,
117
+ "tp": 218,
118
+ "fp": 0,
119
+ "fn": 0
120
+ },
121
+ "IP_ADDRESS": {
122
+ "precision": 1.0,
123
+ "recall": 1.0,
124
+ "f1": 1.0,
125
+ "tp": 35,
126
+ "fp": 0,
127
+ "fn": 0
128
+ },
129
+ "MAJOR": {
130
+ "precision": 0.8888888888888888,
131
+ "recall": 1.0,
132
+ "f1": 0.9411764705882353,
133
+ "tp": 24,
134
+ "fp": 3,
135
+ "fn": 0
136
+ },
137
+ "MEMBER_ID": {
138
+ "precision": 0.9285714285714286,
139
+ "recall": 1.0,
140
+ "f1": 0.962962962962963,
141
+ "tp": 26,
142
+ "fp": 2,
143
+ "fn": 0
144
+ },
145
+ "NAME": {
146
+ "precision": 0.9815140845070423,
147
+ "recall": 0.9982094897045658,
148
+ "f1": 0.9897913892587662,
149
+ "tp": 1115,
150
+ "fp": 21,
151
+ "fn": 2
152
+ },
153
+ "NATIONALITY": {
154
+ "precision": 1.0,
155
+ "recall": 1.0,
156
+ "f1": 1.0,
157
+ "tp": 34,
158
+ "fp": 0,
159
+ "fn": 0
160
+ },
161
+ "NICKNAME": {
162
+ "precision": 1.0,
163
+ "recall": 0.9642857142857143,
164
+ "f1": 0.9818181818181818,
165
+ "tp": 27,
166
+ "fp": 0,
167
+ "fn": 1
168
+ },
169
+ "PARTICIPANT_ID": {
170
+ "precision": 1.0,
171
+ "recall": 0.75,
172
+ "f1": 0.8571428571428571,
173
+ "tp": 6,
174
+ "fp": 0,
175
+ "fn": 2
176
+ },
177
+ "PASSPORTNUM": {
178
+ "precision": 1.0,
179
+ "recall": 1.0,
180
+ "f1": 1.0,
181
+ "tp": 20,
182
+ "fp": 0,
183
+ "fn": 0
184
+ },
185
+ "PHONE": {
186
+ "precision": 1.0,
187
+ "recall": 1.0,
188
+ "f1": 1.0,
189
+ "tp": 801,
190
+ "fp": 0,
191
+ "fn": 0
192
+ },
193
+ "POSITION": {
194
+ "precision": 0.9848866498740554,
195
+ "recall": 0.972636815920398,
196
+ "f1": 0.9787234042553191,
197
+ "tp": 782,
198
+ "fp": 12,
199
+ "fn": 22
200
+ },
201
+ "RELIGION": {
202
+ "precision": 1.0,
203
+ "recall": 1.0,
204
+ "f1": 1.0,
205
+ "tp": 218,
206
+ "fp": 0,
207
+ "fn": 0
208
+ },
209
+ "RRN": {
210
+ "precision": 0.8245614035087719,
211
+ "recall": 1.0,
212
+ "f1": 0.9038461538461539,
213
+ "tp": 47,
214
+ "fp": 10,
215
+ "fn": 0
216
+ },
217
+ "SCHOOL": {
218
+ "precision": 0.9230769230769231,
219
+ "recall": 1.0,
220
+ "f1": 0.9600000000000001,
221
+ "tp": 24,
222
+ "fp": 2,
223
+ "fn": 0
224
+ },
225
+ "URL": {
226
+ "precision": 1.0,
227
+ "recall": 1.0,
228
+ "f1": 1.0,
229
+ "tp": 28,
230
+ "fp": 0,
231
+ "fn": 0
232
+ },
233
+ "USER_ID": {
234
+ "precision": 1.0,
235
+ "recall": 1.0,
236
+ "f1": 1.0,
237
+ "tp": 34,
238
+ "fp": 0,
239
+ "fn": 0
240
+ },
241
+ "VEHICLE_NUMBER": {
242
+ "precision": 0.9722222222222222,
243
+ "recall": 1.0,
244
+ "f1": 0.9859154929577464,
245
+ "tp": 35,
246
+ "fp": 1,
247
+ "fn": 0
248
+ },
249
+ "WEIGHT": {
250
+ "precision": 1.0,
251
+ "recall": 1.0,
252
+ "f1": 1.0,
253
+ "tp": 218,
254
+ "fp": 0,
255
+ "fn": 0
256
+ },
257
+ "WORKPLACE": {
258
+ "precision": 0.9874213836477987,
259
+ "recall": 0.9763681592039801,
260
+ "f1": 0.9818636647904941,
261
+ "tp": 785,
262
+ "fp": 10,
263
+ "fn": 19
264
+ },
265
+ "ZIPCODE": {
266
+ "precision": 0.9,
267
+ "recall": 1.0,
268
+ "f1": 0.9473684210526316,
269
+ "tp": 18,
270
+ "fp": 2,
271
+ "fn": 0
272
+ }
273
+ }
274
+ },
275
+ {
276
+ "epoch": 2,
277
+ "train_loss": 0.0012487326017776835,
278
+ "micro_precision": 0.9828878517889973,
279
+ "micro_recall": 0.9852565920045364,
280
+ "micro_f1": 0.984070796460177,
281
+ "per_label": {
282
+ "ACCOUNT_NUMBER": {
283
+ "precision": 0.918918918918919,
284
+ "recall": 1.0,
285
+ "f1": 0.9577464788732395,
286
+ "tp": 68,
287
+ "fp": 6,
288
+ "fn": 0
289
+ },
290
+ "ADDRESS": {
291
+ "precision": 0.8260869565217391,
292
+ "recall": 0.9047619047619048,
293
+ "f1": 0.8636363636363636,
294
+ "tp": 19,
295
+ "fp": 4,
296
+ "fn": 2
297
+ },
298
+ "AGE": {
299
+ "precision": 0.9428571428571428,
300
+ "recall": 0.9705882352941176,
301
+ "f1": 0.9565217391304348,
302
+ "tp": 33,
303
+ "fp": 2,
304
+ "fn": 1
305
+ },
306
+ "ALIEN_NUMBER": {
307
+ "precision": 1.0,
308
+ "recall": 0.5,
309
+ "f1": 0.6666666666666666,
310
+ "tp": 10,
311
+ "fp": 0,
312
+ "fn": 10
313
+ },
314
+ "BIRTHDATE": {
315
+ "precision": 0.9961685823754789,
316
+ "recall": 1.0,
317
+ "f1": 0.9980806142034548,
318
+ "tp": 260,
319
+ "fp": 1,
320
+ "fn": 0
321
+ },
322
+ "BLOOD_TYPE": {
323
+ "precision": 0.9819819819819819,
324
+ "recall": 1.0,
325
+ "f1": 0.9909090909090909,
326
+ "tp": 218,
327
+ "fp": 4,
328
+ "fn": 0
329
+ },
330
+ "CARD_NUMBER": {
331
+ "precision": 1.0,
332
+ "recall": 1.0,
333
+ "f1": 1.0,
334
+ "tp": 19,
335
+ "fp": 0,
336
+ "fn": 0
337
+ },
338
+ "CITY": {
339
+ "precision": 0.9988571428571429,
340
+ "recall": 1.0,
341
+ "f1": 0.9994282447112636,
342
+ "tp": 874,
343
+ "fp": 1,
344
+ "fn": 0
345
+ },
346
+ "DEPARTMENT": {
347
+ "precision": 0.9488139825218477,
348
+ "recall": 0.945273631840796,
349
+ "f1": 0.9470404984423676,
350
+ "tp": 760,
351
+ "fp": 41,
352
+ "fn": 44
353
+ },
354
+ "DRIVER_LICENSE": {
355
+ "precision": 1.0,
356
+ "recall": 1.0,
357
+ "f1": 1.0,
358
+ "tp": 20,
359
+ "fp": 0,
360
+ "fn": 0
361
+ },
362
+ "EMAIL": {
363
+ "precision": 1.0,
364
+ "recall": 1.0,
365
+ "f1": 1.0,
366
+ "tp": 99,
367
+ "fp": 0,
368
+ "fn": 0
369
+ },
370
+ "EMPLOYEE_ID": {
371
+ "precision": 1.0,
372
+ "recall": 1.0,
373
+ "f1": 1.0,
374
+ "tp": 42,
375
+ "fp": 0,
376
+ "fn": 0
377
+ },
378
+ "GENDER": {
379
+ "precision": 1.0,
380
+ "recall": 0.9705882352941176,
381
+ "f1": 0.9850746268656716,
382
+ "tp": 33,
383
+ "fp": 0,
384
+ "fn": 1
385
+ },
386
+ "HEIGHT": {
387
+ "precision": 1.0,
388
+ "recall": 1.0,
389
+ "f1": 1.0,
390
+ "tp": 218,
391
+ "fp": 0,
392
+ "fn": 0
393
+ },
394
+ "IP_ADDRESS": {
395
+ "precision": 1.0,
396
+ "recall": 1.0,
397
+ "f1": 1.0,
398
+ "tp": 35,
399
+ "fp": 0,
400
+ "fn": 0
401
+ },
402
+ "MAJOR": {
403
+ "precision": 0.8571428571428571,
404
+ "recall": 1.0,
405
+ "f1": 0.923076923076923,
406
+ "tp": 24,
407
+ "fp": 4,
408
+ "fn": 0
409
+ },
410
+ "MEMBER_ID": {
411
+ "precision": 0.9285714285714286,
412
+ "recall": 1.0,
413
+ "f1": 0.962962962962963,
414
+ "tp": 26,
415
+ "fp": 2,
416
+ "fn": 0
417
+ },
418
+ "NAME": {
419
+ "precision": 0.986737400530504,
420
+ "recall": 0.999104744852283,
421
+ "f1": 0.99288256227758,
422
+ "tp": 1116,
423
+ "fp": 15,
424
+ "fn": 1
425
+ },
426
+ "NATIONALITY": {
427
+ "precision": 1.0,
428
+ "recall": 1.0,
429
+ "f1": 1.0,
430
+ "tp": 34,
431
+ "fp": 0,
432
+ "fn": 0
433
+ },
434
+ "NICKNAME": {
435
+ "precision": 1.0,
436
+ "recall": 0.9642857142857143,
437
+ "f1": 0.9818181818181818,
438
+ "tp": 27,
439
+ "fp": 0,
440
+ "fn": 1
441
+ },
442
+ "PARTICIPANT_ID": {
443
+ "precision": 1.0,
444
+ "recall": 0.75,
445
+ "f1": 0.8571428571428571,
446
+ "tp": 6,
447
+ "fp": 0,
448
+ "fn": 2
449
+ },
450
+ "PASSPORTNUM": {
451
+ "precision": 1.0,
452
+ "recall": 1.0,
453
+ "f1": 1.0,
454
+ "tp": 20,
455
+ "fp": 0,
456
+ "fn": 0
457
+ },
458
+ "PHONE": {
459
+ "precision": 1.0,
460
+ "recall": 1.0,
461
+ "f1": 1.0,
462
+ "tp": 801,
463
+ "fp": 0,
464
+ "fn": 0
465
+ },
466
+ "POSITION": {
467
+ "precision": 0.9848484848484849,
468
+ "recall": 0.9701492537313433,
469
+ "f1": 0.9774436090225564,
470
+ "tp": 780,
471
+ "fp": 12,
472
+ "fn": 24
473
+ },
474
+ "RELIGION": {
475
+ "precision": 1.0,
476
+ "recall": 1.0,
477
+ "f1": 1.0,
478
+ "tp": 218,
479
+ "fp": 0,
480
+ "fn": 0
481
+ },
482
+ "RRN": {
483
+ "precision": 0.8245614035087719,
484
+ "recall": 1.0,
485
+ "f1": 0.9038461538461539,
486
+ "tp": 47,
487
+ "fp": 10,
488
+ "fn": 0
489
+ },
490
+ "SCHOOL": {
491
+ "precision": 0.8888888888888888,
492
+ "recall": 1.0,
493
+ "f1": 0.9411764705882353,
494
+ "tp": 24,
495
+ "fp": 3,
496
+ "fn": 0
497
+ },
498
+ "URL": {
499
+ "precision": 1.0,
500
+ "recall": 1.0,
501
+ "f1": 1.0,
502
+ "tp": 28,
503
+ "fp": 0,
504
+ "fn": 0
505
+ },
506
+ "USER_ID": {
507
+ "precision": 1.0,
508
+ "recall": 1.0,
509
+ "f1": 1.0,
510
+ "tp": 34,
511
+ "fp": 0,
512
+ "fn": 0
513
+ },
514
+ "VEHICLE_NUMBER": {
515
+ "precision": 0.9722222222222222,
516
+ "recall": 1.0,
517
+ "f1": 0.9859154929577464,
518
+ "tp": 35,
519
+ "fp": 1,
520
+ "fn": 0
521
+ },
522
+ "WEIGHT": {
523
+ "precision": 1.0,
524
+ "recall": 1.0,
525
+ "f1": 1.0,
526
+ "tp": 218,
527
+ "fp": 0,
528
+ "fn": 0
529
+ },
530
+ "WORKPLACE": {
531
+ "precision": 0.983729662077597,
532
+ "recall": 0.9776119402985075,
533
+ "f1": 0.9806612601372428,
534
+ "tp": 786,
535
+ "fp": 13,
536
+ "fn": 18
537
+ },
538
+ "ZIPCODE": {
539
+ "precision": 0.9,
540
+ "recall": 1.0,
541
+ "f1": 0.9473684210526316,
542
+ "tp": 18,
543
+ "fp": 2,
544
+ "fn": 0
545
+ }
546
+ }
547
+ },
548
+ {
549
+ "epoch": 3,
550
+ "train_loss": 0.001141177041269236,
551
+ "micro_precision": 0.9837271826800622,
552
+ "micro_recall": 0.9855401190813723,
553
+ "micro_f1": 0.9846328163727782,
554
+ "per_label": {
555
+ "ACCOUNT_NUMBER": {
556
+ "precision": 0.918918918918919,
557
+ "recall": 1.0,
558
+ "f1": 0.9577464788732395,
559
+ "tp": 68,
560
+ "fp": 6,
561
+ "fn": 0
562
+ },
563
+ "ADDRESS": {
564
+ "precision": 0.8260869565217391,
565
+ "recall": 0.9047619047619048,
566
+ "f1": 0.8636363636363636,
567
+ "tp": 19,
568
+ "fp": 4,
569
+ "fn": 2
570
+ },
571
+ "AGE": {
572
+ "precision": 0.9428571428571428,
573
+ "recall": 0.9705882352941176,
574
+ "f1": 0.9565217391304348,
575
+ "tp": 33,
576
+ "fp": 2,
577
+ "fn": 1
578
+ },
579
+ "ALIEN_NUMBER": {
580
+ "precision": 1.0,
581
+ "recall": 0.5,
582
+ "f1": 0.6666666666666666,
583
+ "tp": 10,
584
+ "fp": 0,
585
+ "fn": 10
586
+ },
587
+ "BIRTHDATE": {
588
+ "precision": 0.9961685823754789,
589
+ "recall": 1.0,
590
+ "f1": 0.9980806142034548,
591
+ "tp": 260,
592
+ "fp": 1,
593
+ "fn": 0
594
+ },
595
+ "BLOOD_TYPE": {
596
+ "precision": 0.9864253393665159,
597
+ "recall": 1.0,
598
+ "f1": 0.9931662870159453,
599
+ "tp": 218,
600
+ "fp": 3,
601
+ "fn": 0
602
+ },
603
+ "CARD_NUMBER": {
604
+ "precision": 1.0,
605
+ "recall": 1.0,
606
+ "f1": 1.0,
607
+ "tp": 19,
608
+ "fp": 0,
609
+ "fn": 0
610
+ },
611
+ "CITY": {
612
+ "precision": 0.9988571428571429,
613
+ "recall": 1.0,
614
+ "f1": 0.9994282447112636,
615
+ "tp": 874,
616
+ "fp": 1,
617
+ "fn": 0
618
+ },
619
+ "DEPARTMENT": {
620
+ "precision": 0.951310861423221,
621
+ "recall": 0.9477611940298507,
622
+ "f1": 0.9495327102803738,
623
+ "tp": 762,
624
+ "fp": 39,
625
+ "fn": 42
626
+ },
627
+ "DRIVER_LICENSE": {
628
+ "precision": 1.0,
629
+ "recall": 1.0,
630
+ "f1": 1.0,
631
+ "tp": 20,
632
+ "fp": 0,
633
+ "fn": 0
634
+ },
635
+ "EMAIL": {
636
+ "precision": 1.0,
637
+ "recall": 1.0,
638
+ "f1": 1.0,
639
+ "tp": 99,
640
+ "fp": 0,
641
+ "fn": 0
642
+ },
643
+ "EMPLOYEE_ID": {
644
+ "precision": 1.0,
645
+ "recall": 1.0,
646
+ "f1": 1.0,
647
+ "tp": 42,
648
+ "fp": 0,
649
+ "fn": 0
650
+ },
651
+ "GENDER": {
652
+ "precision": 1.0,
653
+ "recall": 0.9705882352941176,
654
+ "f1": 0.9850746268656716,
655
+ "tp": 33,
656
+ "fp": 0,
657
+ "fn": 1
658
+ },
659
+ "HEIGHT": {
660
+ "precision": 1.0,
661
+ "recall": 1.0,
662
+ "f1": 1.0,
663
+ "tp": 218,
664
+ "fp": 0,
665
+ "fn": 0
666
+ },
667
+ "IP_ADDRESS": {
668
+ "precision": 1.0,
669
+ "recall": 1.0,
670
+ "f1": 1.0,
671
+ "tp": 35,
672
+ "fp": 0,
673
+ "fn": 0
674
+ },
675
+ "MAJOR": {
676
+ "precision": 0.8571428571428571,
677
+ "recall": 1.0,
678
+ "f1": 0.923076923076923,
679
+ "tp": 24,
680
+ "fp": 4,
681
+ "fn": 0
682
+ },
683
+ "MEMBER_ID": {
684
+ "precision": 0.9285714285714286,
685
+ "recall": 1.0,
686
+ "f1": 0.962962962962963,
687
+ "tp": 26,
688
+ "fp": 2,
689
+ "fn": 0
690
+ },
691
+ "NAME": {
692
+ "precision": 0.9884853852967228,
693
+ "recall": 0.999104744852283,
694
+ "f1": 0.993766696349065,
695
+ "tp": 1116,
696
+ "fp": 13,
697
+ "fn": 1
698
+ },
699
+ "NATIONALITY": {
700
+ "precision": 1.0,
701
+ "recall": 1.0,
702
+ "f1": 1.0,
703
+ "tp": 34,
704
+ "fp": 0,
705
+ "fn": 0
706
+ },
707
+ "NICKNAME": {
708
+ "precision": 1.0,
709
+ "recall": 0.9642857142857143,
710
+ "f1": 0.9818181818181818,
711
+ "tp": 27,
712
+ "fp": 0,
713
+ "fn": 1
714
+ },
715
+ "PARTICIPANT_ID": {
716
+ "precision": 1.0,
717
+ "recall": 0.75,
718
+ "f1": 0.8571428571428571,
719
+ "tp": 6,
720
+ "fp": 0,
721
+ "fn": 2
722
+ },
723
+ "PASSPORTNUM": {
724
+ "precision": 1.0,
725
+ "recall": 1.0,
726
+ "f1": 1.0,
727
+ "tp": 20,
728
+ "fp": 0,
729
+ "fn": 0
730
+ },
731
+ "PHONE": {
732
+ "precision": 1.0,
733
+ "recall": 1.0,
734
+ "f1": 1.0,
735
+ "tp": 801,
736
+ "fp": 0,
737
+ "fn": 0
738
+ },
739
+ "POSITION": {
740
+ "precision": 0.9848484848484849,
741
+ "recall": 0.9701492537313433,
742
+ "f1": 0.9774436090225564,
743
+ "tp": 780,
744
+ "fp": 12,
745
+ "fn": 24
746
+ },
747
+ "RELIGION": {
748
+ "precision": 1.0,
749
+ "recall": 1.0,
750
+ "f1": 1.0,
751
+ "tp": 218,
752
+ "fp": 0,
753
+ "fn": 0
754
+ },
755
+ "RRN": {
756
+ "precision": 0.8245614035087719,
757
+ "recall": 1.0,
758
+ "f1": 0.9038461538461539,
759
+ "tp": 47,
760
+ "fp": 10,
761
+ "fn": 0
762
+ },
763
+ "SCHOOL": {
764
+ "precision": 0.8888888888888888,
765
+ "recall": 1.0,
766
+ "f1": 0.9411764705882353,
767
+ "tp": 24,
768
+ "fp": 3,
769
+ "fn": 0
770
+ },
771
+ "URL": {
772
+ "precision": 1.0,
773
+ "recall": 1.0,
774
+ "f1": 1.0,
775
+ "tp": 28,
776
+ "fp": 0,
777
+ "fn": 0
778
+ },
779
+ "USER_ID": {
780
+ "precision": 1.0,
781
+ "recall": 1.0,
782
+ "f1": 1.0,
783
+ "tp": 34,
784
+ "fp": 0,
785
+ "fn": 0
786
+ },
787
+ "VEHICLE_NUMBER": {
788
+ "precision": 1.0,
789
+ "recall": 1.0,
790
+ "f1": 1.0,
791
+ "tp": 35,
792
+ "fp": 0,
793
+ "fn": 0
794
+ },
795
+ "WEIGHT": {
796
+ "precision": 1.0,
797
+ "recall": 1.0,
798
+ "f1": 1.0,
799
+ "tp": 218,
800
+ "fp": 0,
801
+ "fn": 0
802
+ },
803
+ "WORKPLACE": {
804
+ "precision": 0.983729662077597,
805
+ "recall": 0.9776119402985075,
806
+ "f1": 0.9806612601372428,
807
+ "tp": 786,
808
+ "fp": 13,
809
+ "fn": 18
810
+ },
811
+ "ZIPCODE": {
812
+ "precision": 0.9,
813
+ "recall": 1.0,
814
+ "f1": 0.9473684210526316,
815
+ "tp": 18,
816
+ "fp": 2,
817
+ "fn": 0
818
+ }
819
+ }
820
+ }
821
+ ]