Isotonic commited on
Commit
fe7983b
1 Parent(s): e839713

Training in progress, epoch 1

Browse files
config.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-german-cased",
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-ORDINALDIRECTION",
14
+ "2": "I-ORDINALDIRECTION",
15
+ "3": "B-NEARBYGPSCOORDINATE",
16
+ "4": "I-NEARBYGPSCOORDINATE",
17
+ "5": "B-PREFIX",
18
+ "6": "I-PREFIX",
19
+ "7": "B-LASTNAME",
20
+ "8": "I-LASTNAME",
21
+ "9": "B-SECONDARYADDRESS",
22
+ "10": "I-SECONDARYADDRESS",
23
+ "11": "B-PHONENUMBER",
24
+ "12": "I-PHONENUMBER",
25
+ "13": "B-JOBTITLE",
26
+ "14": "I-JOBTITLE",
27
+ "15": "B-BIC",
28
+ "16": "I-BIC",
29
+ "17": "B-IBAN",
30
+ "18": "I-IBAN",
31
+ "19": "B-CURRENCYNAME",
32
+ "20": "I-CURRENCYNAME",
33
+ "21": "B-CURRENCYSYMBOL",
34
+ "22": "B-FIRSTNAME",
35
+ "23": "I-FIRSTNAME",
36
+ "24": "B-PHONEIMEI",
37
+ "25": "I-PHONEIMEI",
38
+ "26": "B-AMOUNT",
39
+ "27": "I-AMOUNT",
40
+ "28": "B-CREDITCARDISSUER",
41
+ "29": "I-CREDITCARDISSUER",
42
+ "30": "B-CREDITCARDNUMBER",
43
+ "31": "I-CREDITCARDNUMBER",
44
+ "32": "B-EMAIL",
45
+ "33": "I-EMAIL",
46
+ "34": "B-BITCOINADDRESS",
47
+ "35": "I-BITCOINADDRESS",
48
+ "36": "B-STATE",
49
+ "37": "I-STATE",
50
+ "38": "B-JOBTYPE",
51
+ "39": "B-MASKEDNUMBER",
52
+ "40": "I-MASKEDNUMBER",
53
+ "41": "B-PASSWORD",
54
+ "42": "I-PASSWORD",
55
+ "43": "B-MAC",
56
+ "44": "I-MAC",
57
+ "45": "B-JOBAREA",
58
+ "46": "I-JOBAREA",
59
+ "47": "B-EYECOLOR",
60
+ "48": "I-EYECOLOR",
61
+ "49": "B-USERNAME",
62
+ "50": "I-USERNAME",
63
+ "51": "B-DOB",
64
+ "52": "I-DOB",
65
+ "53": "B-PIN",
66
+ "54": "I-PIN",
67
+ "55": "B-SEX",
68
+ "56": "I-SEX",
69
+ "57": "B-TIME",
70
+ "58": "I-TIME",
71
+ "59": "B-ETHEREUMADDRESS",
72
+ "60": "I-ETHEREUMADDRESS",
73
+ "61": "B-COMPANYNAME",
74
+ "62": "I-COMPANYNAME",
75
+ "63": "B-IPV6",
76
+ "64": "I-IPV6",
77
+ "65": "B-GENDER",
78
+ "66": "I-GENDER",
79
+ "67": "B-CURRENCY",
80
+ "68": "I-CURRENCY",
81
+ "69": "B-USERAGENT",
82
+ "70": "I-USERAGENT",
83
+ "71": "B-IP",
84
+ "72": "I-IP",
85
+ "73": "B-BUILDINGNUMBER",
86
+ "74": "I-BUILDINGNUMBER",
87
+ "75": "B-CITY",
88
+ "76": "I-CITY",
89
+ "77": "I-JOBTYPE",
90
+ "78": "B-IPV4",
91
+ "79": "I-IPV4",
92
+ "80": "B-COUNTY",
93
+ "81": "I-COUNTY",
94
+ "82": "I-CURRENCYSYMBOL",
95
+ "83": "B-CREDITCARDCVV",
96
+ "84": "I-CREDITCARDCVV",
97
+ "85": "B-CURRENCYCODE",
98
+ "86": "I-CURRENCYCODE",
99
+ "87": "B-VEHICLEVRM",
100
+ "88": "I-VEHICLEVRM",
101
+ "89": "B-STREET",
102
+ "90": "I-STREET",
103
+ "91": "B-URL",
104
+ "92": "I-URL",
105
+ "93": "B-DATE",
106
+ "94": "I-DATE",
107
+ "95": "B-ZIPCODE",
108
+ "96": "I-ZIPCODE",
109
+ "97": "B-ACCOUNTNAME",
110
+ "98": "I-ACCOUNTNAME",
111
+ "99": "B-MIDDLENAME",
112
+ "100": "B-SSN",
113
+ "101": "I-SSN",
114
+ "102": "B-AGE",
115
+ "103": "I-AGE",
116
+ "104": "B-ACCOUNTNUMBER",
117
+ "105": "I-ACCOUNTNUMBER",
118
+ "106": "I-MIDDLENAME",
119
+ "107": "B-LITECOINADDRESS",
120
+ "108": "I-LITECOINADDRESS",
121
+ "109": "B-VEHICLEVIN",
122
+ "110": "I-VEHICLEVIN",
123
+ "111": "B-HEIGHT",
124
+ "112": "I-HEIGHT"
125
+ },
126
+ "initializer_range": 0.02,
127
+ "label2id": {
128
+ "B-ACCOUNTNAME": 97,
129
+ "B-ACCOUNTNUMBER": 104,
130
+ "B-AGE": 102,
131
+ "B-AMOUNT": 26,
132
+ "B-BIC": 15,
133
+ "B-BITCOINADDRESS": 34,
134
+ "B-BUILDINGNUMBER": 73,
135
+ "B-CITY": 75,
136
+ "B-COMPANYNAME": 61,
137
+ "B-COUNTY": 80,
138
+ "B-CREDITCARDCVV": 83,
139
+ "B-CREDITCARDISSUER": 28,
140
+ "B-CREDITCARDNUMBER": 30,
141
+ "B-CURRENCY": 67,
142
+ "B-CURRENCYCODE": 85,
143
+ "B-CURRENCYNAME": 19,
144
+ "B-CURRENCYSYMBOL": 21,
145
+ "B-DATE": 93,
146
+ "B-DOB": 51,
147
+ "B-EMAIL": 32,
148
+ "B-ETHEREUMADDRESS": 59,
149
+ "B-EYECOLOR": 47,
150
+ "B-FIRSTNAME": 22,
151
+ "B-GENDER": 65,
152
+ "B-HEIGHT": 111,
153
+ "B-IBAN": 17,
154
+ "B-IP": 71,
155
+ "B-IPV4": 78,
156
+ "B-IPV6": 63,
157
+ "B-JOBAREA": 45,
158
+ "B-JOBTITLE": 13,
159
+ "B-JOBTYPE": 38,
160
+ "B-LASTNAME": 7,
161
+ "B-LITECOINADDRESS": 107,
162
+ "B-MAC": 43,
163
+ "B-MASKEDNUMBER": 39,
164
+ "B-MIDDLENAME": 99,
165
+ "B-NEARBYGPSCOORDINATE": 3,
166
+ "B-ORDINALDIRECTION": 1,
167
+ "B-PASSWORD": 41,
168
+ "B-PHONEIMEI": 24,
169
+ "B-PHONENUMBER": 11,
170
+ "B-PIN": 53,
171
+ "B-PREFIX": 5,
172
+ "B-SECONDARYADDRESS": 9,
173
+ "B-SEX": 55,
174
+ "B-SSN": 100,
175
+ "B-STATE": 36,
176
+ "B-STREET": 89,
177
+ "B-TIME": 57,
178
+ "B-URL": 91,
179
+ "B-USERAGENT": 69,
180
+ "B-USERNAME": 49,
181
+ "B-VEHICLEVIN": 109,
182
+ "B-VEHICLEVRM": 87,
183
+ "B-ZIPCODE": 95,
184
+ "I-ACCOUNTNAME": 98,
185
+ "I-ACCOUNTNUMBER": 105,
186
+ "I-AGE": 103,
187
+ "I-AMOUNT": 27,
188
+ "I-BIC": 16,
189
+ "I-BITCOINADDRESS": 35,
190
+ "I-BUILDINGNUMBER": 74,
191
+ "I-CITY": 76,
192
+ "I-COMPANYNAME": 62,
193
+ "I-COUNTY": 81,
194
+ "I-CREDITCARDCVV": 84,
195
+ "I-CREDITCARDISSUER": 29,
196
+ "I-CREDITCARDNUMBER": 31,
197
+ "I-CURRENCY": 68,
198
+ "I-CURRENCYCODE": 86,
199
+ "I-CURRENCYNAME": 20,
200
+ "I-CURRENCYSYMBOL": 82,
201
+ "I-DATE": 94,
202
+ "I-DOB": 52,
203
+ "I-EMAIL": 33,
204
+ "I-ETHEREUMADDRESS": 60,
205
+ "I-EYECOLOR": 48,
206
+ "I-FIRSTNAME": 23,
207
+ "I-GENDER": 66,
208
+ "I-HEIGHT": 112,
209
+ "I-IBAN": 18,
210
+ "I-IP": 72,
211
+ "I-IPV4": 79,
212
+ "I-IPV6": 64,
213
+ "I-JOBAREA": 46,
214
+ "I-JOBTITLE": 14,
215
+ "I-JOBTYPE": 77,
216
+ "I-LASTNAME": 8,
217
+ "I-LITECOINADDRESS": 108,
218
+ "I-MAC": 44,
219
+ "I-MASKEDNUMBER": 40,
220
+ "I-MIDDLENAME": 106,
221
+ "I-NEARBYGPSCOORDINATE": 4,
222
+ "I-ORDINALDIRECTION": 2,
223
+ "I-PASSWORD": 42,
224
+ "I-PHONEIMEI": 25,
225
+ "I-PHONENUMBER": 12,
226
+ "I-PIN": 54,
227
+ "I-PREFIX": 6,
228
+ "I-SECONDARYADDRESS": 10,
229
+ "I-SEX": 56,
230
+ "I-SSN": 101,
231
+ "I-STATE": 37,
232
+ "I-STREET": 90,
233
+ "I-TIME": 58,
234
+ "I-URL": 92,
235
+ "I-USERAGENT": 70,
236
+ "I-USERNAME": 50,
237
+ "I-VEHICLEVIN": 110,
238
+ "I-VEHICLEVRM": 88,
239
+ "I-ZIPCODE": 96,
240
+ "O": 0
241
+ },
242
+ "max_position_embeddings": 512,
243
+ "model_type": "distilbert",
244
+ "n_heads": 12,
245
+ "n_layers": 6,
246
+ "output_past": true,
247
+ "pad_token_id": 0,
248
+ "qa_dropout": 0.1,
249
+ "seq_classif_dropout": 0.2,
250
+ "sinusoidal_pos_embds": true,
251
+ "tie_weights_": true,
252
+ "torch_dtype": "float32",
253
+ "transformers_version": "4.35.2",
254
+ "vocab_size": 31102
255
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ff1c732205e7aacab291c6429be44e47b61f415397bde05227678150ba8e2f0
3
+ size 267593220
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
+ "101": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "102": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "103": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "104": {
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": false,
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 sha256:55bd8d3e1df96e82f606773a554292f0d81f9d9e4b834f51d8d4437da35f1237
3
+ size 4664
vocab.txt ADDED
The diff for this file is too large to render. See raw diff