Isotonic commited on
Commit
c4830e5
1 Parent(s): 0112be8

Training in progress, epoch 1

Browse files
config.json ADDED
@@ -0,0 +1,250 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForTokenClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "O",
13
+ "1": "B-PHONEIMEI",
14
+ "2": "I-PHONEIMEI",
15
+ "3": "B-JOBAREA",
16
+ "4": "B-FIRSTNAME",
17
+ "5": "I-FIRSTNAME",
18
+ "6": "B-VEHICLEVIN",
19
+ "7": "I-VEHICLEVIN",
20
+ "8": "B-AGE",
21
+ "9": "B-GENDER",
22
+ "10": "I-GENDER",
23
+ "11": "B-HEIGHT",
24
+ "12": "I-HEIGHT",
25
+ "13": "B-BUILDINGNUMBER",
26
+ "14": "I-BUILDINGNUMBER",
27
+ "15": "B-MASKEDNUMBER",
28
+ "16": "I-MASKEDNUMBER",
29
+ "17": "B-PASSWORD",
30
+ "18": "I-PASSWORD",
31
+ "19": "B-DOB",
32
+ "20": "I-DOB",
33
+ "21": "B-IPV6",
34
+ "22": "I-IPV6",
35
+ "23": "B-NEARBYGPSCOORDINATE",
36
+ "24": "I-NEARBYGPSCOORDINATE",
37
+ "25": "B-USERAGENT",
38
+ "26": "I-USERAGENT",
39
+ "27": "B-TIME",
40
+ "28": "I-TIME",
41
+ "29": "B-JOBTITLE",
42
+ "30": "I-JOBTITLE",
43
+ "31": "B-COUNTY",
44
+ "32": "B-EMAIL",
45
+ "33": "I-EMAIL",
46
+ "34": "B-ACCOUNTNUMBER",
47
+ "35": "I-ACCOUNTNUMBER",
48
+ "36": "B-PIN",
49
+ "37": "I-PIN",
50
+ "38": "B-EYECOLOR",
51
+ "39": "I-EYECOLOR",
52
+ "40": "B-LASTNAME",
53
+ "41": "I-LASTNAME",
54
+ "42": "I-JOBAREA",
55
+ "43": "B-IPV4",
56
+ "44": "I-IPV4",
57
+ "45": "B-DATE",
58
+ "46": "I-DATE",
59
+ "47": "B-STREET",
60
+ "48": "I-STREET",
61
+ "49": "B-CITY",
62
+ "50": "I-CITY",
63
+ "51": "B-PREFIX",
64
+ "52": "I-PREFIX",
65
+ "53": "B-CREDITCARDISSUER",
66
+ "54": "B-CREDITCARDNUMBER",
67
+ "55": "I-CREDITCARDNUMBER",
68
+ "56": "I-CREDITCARDISSUER",
69
+ "57": "B-MIDDLENAME",
70
+ "58": "B-STATE",
71
+ "59": "I-STATE",
72
+ "60": "B-VEHICLEVRM",
73
+ "61": "I-VEHICLEVRM",
74
+ "62": "B-ORDINALDIRECTION",
75
+ "63": "B-SEX",
76
+ "64": "B-JOBTYPE",
77
+ "65": "I-JOBTYPE",
78
+ "66": "B-CURRENCYCODE",
79
+ "67": "I-CURRENCYCODE",
80
+ "68": "B-CURRENCYSYMBOL",
81
+ "69": "I-AMOUNT",
82
+ "70": "B-ACCOUNTNAME",
83
+ "71": "I-ACCOUNTNAME",
84
+ "72": "B-BITCOINADDRESS",
85
+ "73": "I-BITCOINADDRESS",
86
+ "74": "B-LITECOINADDRESS",
87
+ "75": "I-LITECOINADDRESS",
88
+ "76": "B-PHONENUMBER",
89
+ "77": "I-PHONENUMBER",
90
+ "78": "B-MAC",
91
+ "79": "I-MAC",
92
+ "80": "B-CURRENCY",
93
+ "81": "B-IBAN",
94
+ "82": "I-IBAN",
95
+ "83": "B-COMPANYNAME",
96
+ "84": "I-COMPANYNAME",
97
+ "85": "B-CURRENCYNAME",
98
+ "86": "I-CURRENCYNAME",
99
+ "87": "I-CURRENCYSYMBOL",
100
+ "88": "B-ZIPCODE",
101
+ "89": "I-ZIPCODE",
102
+ "90": "B-SSN",
103
+ "91": "I-SSN",
104
+ "92": "B-AMOUNT",
105
+ "93": "I-CURRENCY",
106
+ "94": "B-URL",
107
+ "95": "I-URL",
108
+ "96": "B-IP",
109
+ "97": "I-IP",
110
+ "98": "B-SECONDARYADDRESS",
111
+ "99": "I-SECONDARYADDRESS",
112
+ "100": "B-USERNAME",
113
+ "101": "I-USERNAME",
114
+ "102": "B-ETHEREUMADDRESS",
115
+ "103": "I-ETHEREUMADDRESS",
116
+ "104": "B-CREDITCARDCVV",
117
+ "105": "I-CREDITCARDCVV",
118
+ "106": "I-COUNTY",
119
+ "107": "I-AGE",
120
+ "108": "I-MIDDLENAME",
121
+ "109": "B-BIC",
122
+ "110": "I-BIC"
123
+ },
124
+ "initializer_range": 0.02,
125
+ "label2id": {
126
+ "B-ACCOUNTNAME": 70,
127
+ "B-ACCOUNTNUMBER": 34,
128
+ "B-AGE": 8,
129
+ "B-AMOUNT": 92,
130
+ "B-BIC": 109,
131
+ "B-BITCOINADDRESS": 72,
132
+ "B-BUILDINGNUMBER": 13,
133
+ "B-CITY": 49,
134
+ "B-COMPANYNAME": 83,
135
+ "B-COUNTY": 31,
136
+ "B-CREDITCARDCVV": 104,
137
+ "B-CREDITCARDISSUER": 53,
138
+ "B-CREDITCARDNUMBER": 54,
139
+ "B-CURRENCY": 80,
140
+ "B-CURRENCYCODE": 66,
141
+ "B-CURRENCYNAME": 85,
142
+ "B-CURRENCYSYMBOL": 68,
143
+ "B-DATE": 45,
144
+ "B-DOB": 19,
145
+ "B-EMAIL": 32,
146
+ "B-ETHEREUMADDRESS": 102,
147
+ "B-EYECOLOR": 38,
148
+ "B-FIRSTNAME": 4,
149
+ "B-GENDER": 9,
150
+ "B-HEIGHT": 11,
151
+ "B-IBAN": 81,
152
+ "B-IP": 96,
153
+ "B-IPV4": 43,
154
+ "B-IPV6": 21,
155
+ "B-JOBAREA": 3,
156
+ "B-JOBTITLE": 29,
157
+ "B-JOBTYPE": 64,
158
+ "B-LASTNAME": 40,
159
+ "B-LITECOINADDRESS": 74,
160
+ "B-MAC": 78,
161
+ "B-MASKEDNUMBER": 15,
162
+ "B-MIDDLENAME": 57,
163
+ "B-NEARBYGPSCOORDINATE": 23,
164
+ "B-ORDINALDIRECTION": 62,
165
+ "B-PASSWORD": 17,
166
+ "B-PHONEIMEI": 1,
167
+ "B-PHONENUMBER": 76,
168
+ "B-PIN": 36,
169
+ "B-PREFIX": 51,
170
+ "B-SECONDARYADDRESS": 98,
171
+ "B-SEX": 63,
172
+ "B-SSN": 90,
173
+ "B-STATE": 58,
174
+ "B-STREET": 47,
175
+ "B-TIME": 27,
176
+ "B-URL": 94,
177
+ "B-USERAGENT": 25,
178
+ "B-USERNAME": 100,
179
+ "B-VEHICLEVIN": 6,
180
+ "B-VEHICLEVRM": 60,
181
+ "B-ZIPCODE": 88,
182
+ "I-ACCOUNTNAME": 71,
183
+ "I-ACCOUNTNUMBER": 35,
184
+ "I-AGE": 107,
185
+ "I-AMOUNT": 69,
186
+ "I-BIC": 110,
187
+ "I-BITCOINADDRESS": 73,
188
+ "I-BUILDINGNUMBER": 14,
189
+ "I-CITY": 50,
190
+ "I-COMPANYNAME": 84,
191
+ "I-COUNTY": 106,
192
+ "I-CREDITCARDCVV": 105,
193
+ "I-CREDITCARDISSUER": 56,
194
+ "I-CREDITCARDNUMBER": 55,
195
+ "I-CURRENCY": 93,
196
+ "I-CURRENCYCODE": 67,
197
+ "I-CURRENCYNAME": 86,
198
+ "I-CURRENCYSYMBOL": 87,
199
+ "I-DATE": 46,
200
+ "I-DOB": 20,
201
+ "I-EMAIL": 33,
202
+ "I-ETHEREUMADDRESS": 103,
203
+ "I-EYECOLOR": 39,
204
+ "I-FIRSTNAME": 5,
205
+ "I-GENDER": 10,
206
+ "I-HEIGHT": 12,
207
+ "I-IBAN": 82,
208
+ "I-IP": 97,
209
+ "I-IPV4": 44,
210
+ "I-IPV6": 22,
211
+ "I-JOBAREA": 42,
212
+ "I-JOBTITLE": 30,
213
+ "I-JOBTYPE": 65,
214
+ "I-LASTNAME": 41,
215
+ "I-LITECOINADDRESS": 75,
216
+ "I-MAC": 79,
217
+ "I-MASKEDNUMBER": 16,
218
+ "I-MIDDLENAME": 108,
219
+ "I-NEARBYGPSCOORDINATE": 24,
220
+ "I-PASSWORD": 18,
221
+ "I-PHONEIMEI": 2,
222
+ "I-PHONENUMBER": 77,
223
+ "I-PIN": 37,
224
+ "I-PREFIX": 52,
225
+ "I-SECONDARYADDRESS": 99,
226
+ "I-SSN": 91,
227
+ "I-STATE": 59,
228
+ "I-STREET": 48,
229
+ "I-TIME": 28,
230
+ "I-URL": 95,
231
+ "I-USERAGENT": 26,
232
+ "I-USERNAME": 101,
233
+ "I-VEHICLEVIN": 7,
234
+ "I-VEHICLEVRM": 61,
235
+ "I-ZIPCODE": 89,
236
+ "O": 0
237
+ },
238
+ "max_position_embeddings": 512,
239
+ "model_type": "distilbert",
240
+ "n_heads": 12,
241
+ "n_layers": 6,
242
+ "pad_token_id": 0,
243
+ "qa_dropout": 0.1,
244
+ "seq_classif_dropout": 0.2,
245
+ "sinusoidal_pos_embds": false,
246
+ "tie_weights_": true,
247
+ "torch_dtype": "float32",
248
+ "transformers_version": "4.35.0",
249
+ "vocab_size": 30522
250
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2921989c9a27fc5fda3fd0e8557327f8ce4e9484a92ea885b21f273af836896f
3
+ size 265805308
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "mask_token": "[MASK]",
48
+ "model_max_length": 512,
49
+ "pad_token": "[PAD]",
50
+ "sep_token": "[SEP]",
51
+ "strip_accents": null,
52
+ "tokenize_chinese_chars": true,
53
+ "tokenizer_class": "DistilBertTokenizer",
54
+ "unk_token": "[UNK]"
55
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9abef3463a596aed47802635090c13f894fedef7c57929c0dde92f409fc4953e
3
+ size 4219
vocab.txt ADDED
The diff for this file is too large to render. See raw diff