jacobfortinmtl commited on
Commit
0ddf744
1 Parent(s): 50f41a8

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": "B-URL",
13
+ "1": "B-NEARBYGPSCOORDINATE",
14
+ "2": "I-FIRSTNAME",
15
+ "3": "B-IBAN",
16
+ "4": "B-JOBAREA",
17
+ "5": "B-JOBTYPE",
18
+ "6": "I-PHONEIMEI",
19
+ "7": "I-ACCOUNTNUMBER",
20
+ "8": "B-DATE",
21
+ "9": "I-PHONENUMBER",
22
+ "10": "B-MIDDLENAME",
23
+ "11": "I-PASSWORD",
24
+ "12": "B-PHONEIMEI",
25
+ "13": "B-COUNTY",
26
+ "14": "B-PASSWORD",
27
+ "15": "I-VEHICLEVIN",
28
+ "16": "B-AGE",
29
+ "17": "B-AMOUNT",
30
+ "18": "I-CREDITCARDNUMBER",
31
+ "19": "I-COUNTY",
32
+ "20": "I-USERNAME",
33
+ "21": "I-ZIPCODE",
34
+ "22": "I-CURRENCY",
35
+ "23": "I-SECONDARYADDRESS",
36
+ "24": "B-FIRSTNAME",
37
+ "25": "B-SEX",
38
+ "26": "I-ACCOUNTNAME",
39
+ "27": "I-CREDITCARDISSUER",
40
+ "28": "B-DOB",
41
+ "29": "I-IPV6",
42
+ "30": "B-IP",
43
+ "31": "B-VEHICLEVRM",
44
+ "32": "B-USERAGENT",
45
+ "33": "B-BITCOINADDRESS",
46
+ "34": "I-BIC",
47
+ "35": "I-CREDITCARDCVV",
48
+ "36": "I-STATE",
49
+ "37": "B-CREDITCARDISSUER",
50
+ "38": "B-GENDER",
51
+ "39": "I-LITECOINADDRESS",
52
+ "40": "B-SECONDARYADDRESS",
53
+ "41": "B-LITECOINADDRESS",
54
+ "42": "I-IBAN",
55
+ "43": "B-ACCOUNTNAME",
56
+ "44": "B-STATE",
57
+ "45": "I-TIME",
58
+ "46": "I-GENDER",
59
+ "47": "I-PIN",
60
+ "48": "I-JOBTYPE",
61
+ "49": "B-LASTNAME",
62
+ "50": "B-CREDITCARDNUMBER",
63
+ "51": "I-IPV4",
64
+ "52": "B-HEIGHT",
65
+ "53": "B-ETHEREUMADDRESS",
66
+ "54": "I-MASKEDNUMBER",
67
+ "55": "I-EYECOLOR",
68
+ "56": "B-USERNAME",
69
+ "57": "B-EYECOLOR",
70
+ "58": "B-JOBTITLE",
71
+ "59": "I-COMPANYNAME",
72
+ "60": "I-CURRENCYSYMBOL",
73
+ "61": "I-DOB",
74
+ "62": "I-LASTNAME",
75
+ "63": "B-VEHICLEVIN",
76
+ "64": "B-IPV4",
77
+ "65": "B-PHONENUMBER",
78
+ "66": "B-PREFIX",
79
+ "67": "I-URL",
80
+ "68": "B-CURRENCYNAME",
81
+ "69": "B-STREET",
82
+ "70": "B-SSN",
83
+ "71": "B-ZIPCODE",
84
+ "72": "I-STREET",
85
+ "73": "I-MIDDLENAME",
86
+ "74": "I-AMOUNT",
87
+ "75": "I-AGE",
88
+ "76": "I-PREFIX",
89
+ "77": "I-JOBTITLE",
90
+ "78": "I-IP",
91
+ "79": "O",
92
+ "80": "B-TIME",
93
+ "81": "I-JOBAREA",
94
+ "82": "I-NEARBYGPSCOORDINATE",
95
+ "83": "B-ACCOUNTNUMBER",
96
+ "84": "B-EMAIL",
97
+ "85": "B-PIN",
98
+ "86": "I-SSN",
99
+ "87": "B-COMPANYNAME",
100
+ "88": "B-MASKEDNUMBER",
101
+ "89": "I-ETHEREUMADDRESS",
102
+ "90": "I-DATE",
103
+ "91": "I-HEIGHT",
104
+ "92": "B-BIC",
105
+ "93": "I-EMAIL",
106
+ "94": "B-CITY",
107
+ "95": "B-ORDINALDIRECTION",
108
+ "96": "I-BUILDINGNUMBER",
109
+ "97": "I-MAC",
110
+ "98": "I-USERAGENT",
111
+ "99": "I-BITCOINADDRESS",
112
+ "100": "I-CITY",
113
+ "101": "B-CURRENCYCODE",
114
+ "102": "B-CURRENCY",
115
+ "103": "B-MAC",
116
+ "104": "I-CURRENCYCODE",
117
+ "105": "B-CREDITCARDCVV",
118
+ "106": "B-CURRENCYSYMBOL",
119
+ "107": "B-IPV6",
120
+ "108": "I-VEHICLEVRM",
121
+ "109": "B-BUILDINGNUMBER",
122
+ "110": "I-CURRENCYNAME"
123
+ },
124
+ "initializer_range": 0.02,
125
+ "label2id": {
126
+ "B-ACCOUNTNAME": 43,
127
+ "B-ACCOUNTNUMBER": 83,
128
+ "B-AGE": 16,
129
+ "B-AMOUNT": 17,
130
+ "B-BIC": 92,
131
+ "B-BITCOINADDRESS": 33,
132
+ "B-BUILDINGNUMBER": 109,
133
+ "B-CITY": 94,
134
+ "B-COMPANYNAME": 87,
135
+ "B-COUNTY": 13,
136
+ "B-CREDITCARDCVV": 105,
137
+ "B-CREDITCARDISSUER": 37,
138
+ "B-CREDITCARDNUMBER": 50,
139
+ "B-CURRENCY": 102,
140
+ "B-CURRENCYCODE": 101,
141
+ "B-CURRENCYNAME": 68,
142
+ "B-CURRENCYSYMBOL": 106,
143
+ "B-DATE": 8,
144
+ "B-DOB": 28,
145
+ "B-EMAIL": 84,
146
+ "B-ETHEREUMADDRESS": 53,
147
+ "B-EYECOLOR": 57,
148
+ "B-FIRSTNAME": 24,
149
+ "B-GENDER": 38,
150
+ "B-HEIGHT": 52,
151
+ "B-IBAN": 3,
152
+ "B-IP": 30,
153
+ "B-IPV4": 64,
154
+ "B-IPV6": 107,
155
+ "B-JOBAREA": 4,
156
+ "B-JOBTITLE": 58,
157
+ "B-JOBTYPE": 5,
158
+ "B-LASTNAME": 49,
159
+ "B-LITECOINADDRESS": 41,
160
+ "B-MAC": 103,
161
+ "B-MASKEDNUMBER": 88,
162
+ "B-MIDDLENAME": 10,
163
+ "B-NEARBYGPSCOORDINATE": 1,
164
+ "B-ORDINALDIRECTION": 95,
165
+ "B-PASSWORD": 14,
166
+ "B-PHONEIMEI": 12,
167
+ "B-PHONENUMBER": 65,
168
+ "B-PIN": 85,
169
+ "B-PREFIX": 66,
170
+ "B-SECONDARYADDRESS": 40,
171
+ "B-SEX": 25,
172
+ "B-SSN": 70,
173
+ "B-STATE": 44,
174
+ "B-STREET": 69,
175
+ "B-TIME": 80,
176
+ "B-URL": 0,
177
+ "B-USERAGENT": 32,
178
+ "B-USERNAME": 56,
179
+ "B-VEHICLEVIN": 63,
180
+ "B-VEHICLEVRM": 31,
181
+ "B-ZIPCODE": 71,
182
+ "I-ACCOUNTNAME": 26,
183
+ "I-ACCOUNTNUMBER": 7,
184
+ "I-AGE": 75,
185
+ "I-AMOUNT": 74,
186
+ "I-BIC": 34,
187
+ "I-BITCOINADDRESS": 99,
188
+ "I-BUILDINGNUMBER": 96,
189
+ "I-CITY": 100,
190
+ "I-COMPANYNAME": 59,
191
+ "I-COUNTY": 19,
192
+ "I-CREDITCARDCVV": 35,
193
+ "I-CREDITCARDISSUER": 27,
194
+ "I-CREDITCARDNUMBER": 18,
195
+ "I-CURRENCY": 22,
196
+ "I-CURRENCYCODE": 104,
197
+ "I-CURRENCYNAME": 110,
198
+ "I-CURRENCYSYMBOL": 60,
199
+ "I-DATE": 90,
200
+ "I-DOB": 61,
201
+ "I-EMAIL": 93,
202
+ "I-ETHEREUMADDRESS": 89,
203
+ "I-EYECOLOR": 55,
204
+ "I-FIRSTNAME": 2,
205
+ "I-GENDER": 46,
206
+ "I-HEIGHT": 91,
207
+ "I-IBAN": 42,
208
+ "I-IP": 78,
209
+ "I-IPV4": 51,
210
+ "I-IPV6": 29,
211
+ "I-JOBAREA": 81,
212
+ "I-JOBTITLE": 77,
213
+ "I-JOBTYPE": 48,
214
+ "I-LASTNAME": 62,
215
+ "I-LITECOINADDRESS": 39,
216
+ "I-MAC": 97,
217
+ "I-MASKEDNUMBER": 54,
218
+ "I-MIDDLENAME": 73,
219
+ "I-NEARBYGPSCOORDINATE": 82,
220
+ "I-PASSWORD": 11,
221
+ "I-PHONEIMEI": 6,
222
+ "I-PHONENUMBER": 9,
223
+ "I-PIN": 47,
224
+ "I-PREFIX": 76,
225
+ "I-SECONDARYADDRESS": 23,
226
+ "I-SSN": 86,
227
+ "I-STATE": 36,
228
+ "I-STREET": 72,
229
+ "I-TIME": 45,
230
+ "I-URL": 67,
231
+ "I-USERAGENT": 98,
232
+ "I-USERNAME": 20,
233
+ "I-VEHICLEVIN": 15,
234
+ "I-VEHICLEVRM": 108,
235
+ "I-ZIPCODE": 21,
236
+ "O": 79
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": "4.39.3",
249
+ "vocab_size": 30522
250
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75465580446bbc0a2f112f661e400484c3ca3ebaf2f0046618febd6e6567ddc7
3
+ size 265805308
runs/Apr04_15-42-01_5ec3c10e95d4/events.out.tfevents.1712245322.5ec3c10e95d4.1851.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ecd22239ab60ff5fc7688f77897c8340e95326403bcb3dff1214fae78cfa0b6
3
+ size 9689
runs/Apr04_15-42-36_5ec3c10e95d4/events.out.tfevents.1712245357.5ec3c10e95d4.1851.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3e01644e72ab3820aa599d8afe6d1dca148c36a4e7b7d814af5a7360c18f40e
3
+ size 9689
runs/Apr04_15-48-13_5ec3c10e95d4/events.out.tfevents.1712245699.5ec3c10e95d4.1851.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de33b88a8c3214e9d97c9d306d8cae4e2e3d7fd532e267b0989b8f53f342b77c
3
+ size 11004
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": 1000000000000000019884624838656,
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 sha256:0553c454a0507f9c7065de830b81b13f68b5e5837fd64aeea027379331240111
3
+ size 4920
vocab.txt ADDED
The diff for this file is too large to render. See raw diff