arnabdhar commited on
Commit
881a246
1 Parent(s): 4731680

Training in progress, step 2500

Browse files
config.json ADDED
@@ -0,0 +1,263 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "prajjwal1/bert-tiny",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "O",
13
+ "1": "B-O",
14
+ "2": "I-O",
15
+ "3": "L-O",
16
+ "4": "U-O",
17
+ "5": "B-PERSON",
18
+ "6": "I-PERSON",
19
+ "7": "L-PERSON",
20
+ "8": "U-PERSON",
21
+ "9": "B-LOCATION",
22
+ "10": "I-LOCATION",
23
+ "11": "L-LOCATION",
24
+ "12": "U-LOCATION",
25
+ "13": "B-ORGANIZATION",
26
+ "14": "I-ORGANIZATION",
27
+ "15": "L-ORGANIZATION",
28
+ "16": "U-ORGANIZATION",
29
+ "17": "B-NRP",
30
+ "18": "I-NRP",
31
+ "19": "L-NRP",
32
+ "20": "U-NRP",
33
+ "21": "B-DATE_TIME",
34
+ "22": "I-DATE_TIME",
35
+ "23": "L-DATE_TIME",
36
+ "24": "U-DATE_TIME",
37
+ "25": "B-CREDIT_CARD",
38
+ "26": "I-CREDIT_CARD",
39
+ "27": "L-CREDIT_CARD",
40
+ "28": "U-CREDIT_CARD",
41
+ "29": "B-URL",
42
+ "30": "I-URL",
43
+ "31": "L-URL",
44
+ "32": "U-URL",
45
+ "33": "B-IBAN_CODE",
46
+ "34": "I-IBAN_CODE",
47
+ "35": "L-IBAN_CODE",
48
+ "36": "U-IBAN_CODE",
49
+ "37": "B-US_BANK_NUMBER",
50
+ "38": "I-US_BANK_NUMBER",
51
+ "39": "L-US_BANK_NUMBER",
52
+ "40": "U-US_BANK_NUMBER",
53
+ "41": "B-PHONE_NUMBER",
54
+ "42": "I-PHONE_NUMBER",
55
+ "43": "L-PHONE_NUMBER",
56
+ "44": "U-PHONE_NUMBER",
57
+ "45": "B-US_SSN",
58
+ "46": "I-US_SSN",
59
+ "47": "L-US_SSN",
60
+ "48": "U-US_SSN",
61
+ "49": "B-US_PASSPORT",
62
+ "50": "I-US_PASSPORT",
63
+ "51": "L-US_PASSPORT",
64
+ "52": "U-US_PASSPORT",
65
+ "53": "B-US_DRIVER_LICENSE",
66
+ "54": "I-US_DRIVER_LICENSE",
67
+ "55": "L-US_DRIVER_LICENSE",
68
+ "56": "U-US_DRIVER_LICENSE",
69
+ "57": "B-US_LICENSE_PLATE",
70
+ "58": "I-US_LICENSE_PLATE",
71
+ "59": "L-US_LICENSE_PLATE",
72
+ "60": "U-US_LICENSE_PLATE",
73
+ "61": "B-IP_ADDRESS",
74
+ "62": "I-IP_ADDRESS",
75
+ "63": "L-IP_ADDRESS",
76
+ "64": "U-IP_ADDRESS",
77
+ "65": "B-US_ITIN",
78
+ "66": "I-US_ITIN",
79
+ "67": "L-US_ITIN",
80
+ "68": "U-US_ITIN",
81
+ "69": "B-EMAIL_ADDRESS",
82
+ "70": "I-EMAIL_ADDRESS",
83
+ "71": "L-EMAIL_ADDRESS",
84
+ "72": "U-EMAIL_ADDRESS",
85
+ "73": "B-TITLE",
86
+ "74": "I-TITLE",
87
+ "75": "L-TITLE",
88
+ "76": "U-TITLE",
89
+ "77": "B-COORDINATE",
90
+ "78": "I-COORDINATE",
91
+ "79": "L-COORDINATE",
92
+ "80": "U-COORDINATE",
93
+ "81": "B-IMEI",
94
+ "82": "I-IMEI",
95
+ "83": "L-IMEI",
96
+ "84": "U-IMEI",
97
+ "85": "B-PASSWORD",
98
+ "86": "I-PASSWORD",
99
+ "87": "L-PASSWORD",
100
+ "88": "U-PASSWORD",
101
+ "89": "B-LICENSE_PLATE",
102
+ "90": "I-LICENSE_PLATE",
103
+ "91": "L-LICENSE_PLATE",
104
+ "92": "U-LICENSE_PLATE",
105
+ "93": "B-CURRENCY",
106
+ "94": "I-CURRENCY",
107
+ "95": "L-CURRENCY",
108
+ "96": "U-CURRENCY",
109
+ "97": "B-FINANCIAL",
110
+ "98": "I-FINANCIAL",
111
+ "99": "L-FINANCIAL",
112
+ "100": "U-FINANCIAL",
113
+ "101": "B-ROUTING_NUMBER",
114
+ "102": "I-ROUTING_NUMBER",
115
+ "103": "L-ROUTING_NUMBER",
116
+ "104": "U-ROUTING_NUMBER",
117
+ "105": "B-SWIFT_CODE",
118
+ "106": "I-SWIFT_CODE",
119
+ "107": "L-SWIFT_CODE",
120
+ "108": "U-SWIFT_CODE",
121
+ "109": "B-MAC_ADDRESS",
122
+ "110": "I-MAC_ADDRESS",
123
+ "111": "L-MAC_ADDRESS",
124
+ "112": "U-MAC_ADDRESS",
125
+ "113": "B-AGE",
126
+ "114": "I-AGE",
127
+ "115": "L-AGE",
128
+ "116": "U-AGE"
129
+ },
130
+ "initializer_range": 0.02,
131
+ "intermediate_size": 512,
132
+ "label2id": {
133
+ "B-AGE": 113,
134
+ "B-COORDINATE": 77,
135
+ "B-CREDIT_CARD": 25,
136
+ "B-CURRENCY": 93,
137
+ "B-DATE_TIME": 21,
138
+ "B-EMAIL_ADDRESS": 69,
139
+ "B-FINANCIAL": 97,
140
+ "B-IBAN_CODE": 33,
141
+ "B-IMEI": 81,
142
+ "B-IP_ADDRESS": 61,
143
+ "B-LICENSE_PLATE": 89,
144
+ "B-LOCATION": 9,
145
+ "B-MAC_ADDRESS": 109,
146
+ "B-NRP": 17,
147
+ "B-O": 1,
148
+ "B-ORGANIZATION": 13,
149
+ "B-PASSWORD": 85,
150
+ "B-PERSON": 5,
151
+ "B-PHONE_NUMBER": 41,
152
+ "B-ROUTING_NUMBER": 101,
153
+ "B-SWIFT_CODE": 105,
154
+ "B-TITLE": 73,
155
+ "B-URL": 29,
156
+ "B-US_BANK_NUMBER": 37,
157
+ "B-US_DRIVER_LICENSE": 53,
158
+ "B-US_ITIN": 65,
159
+ "B-US_LICENSE_PLATE": 57,
160
+ "B-US_PASSPORT": 49,
161
+ "B-US_SSN": 45,
162
+ "I-AGE": 114,
163
+ "I-COORDINATE": 78,
164
+ "I-CREDIT_CARD": 26,
165
+ "I-CURRENCY": 94,
166
+ "I-DATE_TIME": 22,
167
+ "I-EMAIL_ADDRESS": 70,
168
+ "I-FINANCIAL": 98,
169
+ "I-IBAN_CODE": 34,
170
+ "I-IMEI": 82,
171
+ "I-IP_ADDRESS": 62,
172
+ "I-LICENSE_PLATE": 90,
173
+ "I-LOCATION": 10,
174
+ "I-MAC_ADDRESS": 110,
175
+ "I-NRP": 18,
176
+ "I-O": 2,
177
+ "I-ORGANIZATION": 14,
178
+ "I-PASSWORD": 86,
179
+ "I-PERSON": 6,
180
+ "I-PHONE_NUMBER": 42,
181
+ "I-ROUTING_NUMBER": 102,
182
+ "I-SWIFT_CODE": 106,
183
+ "I-TITLE": 74,
184
+ "I-URL": 30,
185
+ "I-US_BANK_NUMBER": 38,
186
+ "I-US_DRIVER_LICENSE": 54,
187
+ "I-US_ITIN": 66,
188
+ "I-US_LICENSE_PLATE": 58,
189
+ "I-US_PASSPORT": 50,
190
+ "I-US_SSN": 46,
191
+ "L-AGE": 115,
192
+ "L-COORDINATE": 79,
193
+ "L-CREDIT_CARD": 27,
194
+ "L-CURRENCY": 95,
195
+ "L-DATE_TIME": 23,
196
+ "L-EMAIL_ADDRESS": 71,
197
+ "L-FINANCIAL": 99,
198
+ "L-IBAN_CODE": 35,
199
+ "L-IMEI": 83,
200
+ "L-IP_ADDRESS": 63,
201
+ "L-LICENSE_PLATE": 91,
202
+ "L-LOCATION": 11,
203
+ "L-MAC_ADDRESS": 111,
204
+ "L-NRP": 19,
205
+ "L-O": 3,
206
+ "L-ORGANIZATION": 15,
207
+ "L-PASSWORD": 87,
208
+ "L-PERSON": 7,
209
+ "L-PHONE_NUMBER": 43,
210
+ "L-ROUTING_NUMBER": 103,
211
+ "L-SWIFT_CODE": 107,
212
+ "L-TITLE": 75,
213
+ "L-URL": 31,
214
+ "L-US_BANK_NUMBER": 39,
215
+ "L-US_DRIVER_LICENSE": 55,
216
+ "L-US_ITIN": 67,
217
+ "L-US_LICENSE_PLATE": 59,
218
+ "L-US_PASSPORT": 51,
219
+ "L-US_SSN": 47,
220
+ "O": 0,
221
+ "U-AGE": 116,
222
+ "U-COORDINATE": 80,
223
+ "U-CREDIT_CARD": 28,
224
+ "U-CURRENCY": 96,
225
+ "U-DATE_TIME": 24,
226
+ "U-EMAIL_ADDRESS": 72,
227
+ "U-FINANCIAL": 100,
228
+ "U-IBAN_CODE": 36,
229
+ "U-IMEI": 84,
230
+ "U-IP_ADDRESS": 64,
231
+ "U-LICENSE_PLATE": 92,
232
+ "U-LOCATION": 12,
233
+ "U-MAC_ADDRESS": 112,
234
+ "U-NRP": 20,
235
+ "U-O": 4,
236
+ "U-ORGANIZATION": 16,
237
+ "U-PASSWORD": 88,
238
+ "U-PERSON": 8,
239
+ "U-PHONE_NUMBER": 44,
240
+ "U-ROUTING_NUMBER": 104,
241
+ "U-SWIFT_CODE": 108,
242
+ "U-TITLE": 76,
243
+ "U-URL": 32,
244
+ "U-US_BANK_NUMBER": 40,
245
+ "U-US_DRIVER_LICENSE": 56,
246
+ "U-US_ITIN": 68,
247
+ "U-US_LICENSE_PLATE": 60,
248
+ "U-US_PASSPORT": 52,
249
+ "U-US_SSN": 48
250
+ },
251
+ "layer_norm_eps": 1e-12,
252
+ "max_position_embeddings": 512,
253
+ "model_type": "bert",
254
+ "num_attention_heads": 2,
255
+ "num_hidden_layers": 2,
256
+ "pad_token_id": 0,
257
+ "position_embedding_type": "absolute",
258
+ "torch_dtype": "float32",
259
+ "transformers_version": "4.36.2",
260
+ "type_vocab_size": 2,
261
+ "use_cache": true,
262
+ "vocab_size": 30522
263
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f2c76963dcebaeb8523bec2519fe7e1d95ed2338d9417d3ad0257de6b1038bc
3
+ size 17542420
runs/Dec23_04-07-26_cf6a4d141451/events.out.tfevents.1703304455.cf6a4d141451.215.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e3720a616b4b8dd0dad5255e779f714b21215cb54a806a9fe58e72093461fe3
3
+ size 14154
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 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 1000000000000000019884624838656,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8527e2a9d91876e83de5c50d587215e135a646595405f24d62a011982f0513d
3
+ size 4728
vocab.txt ADDED
The diff for this file is too large to render. See raw diff