priteshkj commited on
Commit
4bc48a7
·
1 Parent(s): e5462aa

Training done

Browse files
Files changed (3) hide show
  1. added_tokens.json +31 -7
  2. tokenizer.json +222 -6
  3. tokenizer_config.json +198 -6
added_tokens.json CHANGED
@@ -1,12 +1,36 @@
1
  {
2
- "</s_Assets>": 57528,
3
- "</s_Liabilities>": 57526,
4
- "</s_Misc>": 57530,
5
- "<s_Assets>": 57527,
6
- "<s_Liabilities>": 57525,
7
- "<s_Misc>": 57529,
8
- "<s_cord-v2>": 57531,
 
 
 
 
 
 
 
 
 
 
9
  "<s_iitcdip>": 57523,
 
 
 
 
 
 
10
  "<s_synthdog>": 57524,
 
 
 
 
 
 
 
 
11
  "<sep/>": 57522
12
  }
 
1
  {
2
+ "</s_height>": 57538,
3
+ "</s_image_id>": 57532,
4
+ "</s_image_size>": 57534,
5
+ "</s_label>": 57554,
6
+ "</s_meta>": 57526,
7
+ "</s_rect>": 57542,
8
+ "</s_split>": 57530,
9
+ "</s_value>": 57552,
10
+ "</s_version>": 57528,
11
+ "</s_width>": 57536,
12
+ "</s_words>": 57540,
13
+ "</s_x1>": 57544,
14
+ "</s_x2>": 57548,
15
+ "</s_y1>": 57546,
16
+ "</s_y2>": 57550,
17
+ "<s_cord-v2>": 57555,
18
+ "<s_height>": 57537,
19
  "<s_iitcdip>": 57523,
20
+ "<s_image_id>": 57531,
21
+ "<s_image_size>": 57533,
22
+ "<s_label>": 57553,
23
+ "<s_meta>": 57525,
24
+ "<s_rect>": 57541,
25
+ "<s_split>": 57529,
26
  "<s_synthdog>": 57524,
27
+ "<s_value>": 57551,
28
+ "<s_version>": 57527,
29
+ "<s_width>": 57535,
30
+ "<s_words>": 57539,
31
+ "<s_x1>": 57543,
32
+ "<s_x2>": 57547,
33
+ "<s_y1>": 57545,
34
+ "<s_y2>": 57549,
35
  "<sep/>": 57522
36
  }
tokenizer.json CHANGED
@@ -91,7 +91,7 @@
91
  },
92
  {
93
  "id": 57525,
94
- "content": "<s_Liabilities>",
95
  "single_word": false,
96
  "lstrip": false,
97
  "rstrip": false,
@@ -100,7 +100,7 @@
100
  },
101
  {
102
  "id": 57526,
103
- "content": "</s_Liabilities>",
104
  "single_word": false,
105
  "lstrip": false,
106
  "rstrip": false,
@@ -109,7 +109,7 @@
109
  },
110
  {
111
  "id": 57527,
112
- "content": "<s_Assets>",
113
  "single_word": false,
114
  "lstrip": false,
115
  "rstrip": false,
@@ -118,7 +118,7 @@
118
  },
119
  {
120
  "id": 57528,
121
- "content": "</s_Assets>",
122
  "single_word": false,
123
  "lstrip": false,
124
  "rstrip": false,
@@ -127,7 +127,7 @@
127
  },
128
  {
129
  "id": 57529,
130
- "content": "<s_Misc>",
131
  "single_word": false,
132
  "lstrip": false,
133
  "rstrip": false,
@@ -136,7 +136,7 @@
136
  },
137
  {
138
  "id": 57530,
139
- "content": "</s_Misc>",
140
  "single_word": false,
141
  "lstrip": false,
142
  "rstrip": false,
@@ -145,6 +145,222 @@
145
  },
146
  {
147
  "id": 57531,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
148
  "content": "<s_cord-v2>",
149
  "single_word": false,
150
  "lstrip": false,
 
91
  },
92
  {
93
  "id": 57525,
94
+ "content": "<s_meta>",
95
  "single_word": false,
96
  "lstrip": false,
97
  "rstrip": false,
 
100
  },
101
  {
102
  "id": 57526,
103
+ "content": "</s_meta>",
104
  "single_word": false,
105
  "lstrip": false,
106
  "rstrip": false,
 
109
  },
110
  {
111
  "id": 57527,
112
+ "content": "<s_version>",
113
  "single_word": false,
114
  "lstrip": false,
115
  "rstrip": false,
 
118
  },
119
  {
120
  "id": 57528,
121
+ "content": "</s_version>",
122
  "single_word": false,
123
  "lstrip": false,
124
  "rstrip": false,
 
127
  },
128
  {
129
  "id": 57529,
130
+ "content": "<s_split>",
131
  "single_word": false,
132
  "lstrip": false,
133
  "rstrip": false,
 
136
  },
137
  {
138
  "id": 57530,
139
+ "content": "</s_split>",
140
  "single_word": false,
141
  "lstrip": false,
142
  "rstrip": false,
 
145
  },
146
  {
147
  "id": 57531,
148
+ "content": "<s_image_id>",
149
+ "single_word": false,
150
+ "lstrip": false,
151
+ "rstrip": false,
152
+ "normalized": true,
153
+ "special": false
154
+ },
155
+ {
156
+ "id": 57532,
157
+ "content": "</s_image_id>",
158
+ "single_word": false,
159
+ "lstrip": false,
160
+ "rstrip": false,
161
+ "normalized": true,
162
+ "special": false
163
+ },
164
+ {
165
+ "id": 57533,
166
+ "content": "<s_image_size>",
167
+ "single_word": false,
168
+ "lstrip": false,
169
+ "rstrip": false,
170
+ "normalized": true,
171
+ "special": false
172
+ },
173
+ {
174
+ "id": 57534,
175
+ "content": "</s_image_size>",
176
+ "single_word": false,
177
+ "lstrip": false,
178
+ "rstrip": false,
179
+ "normalized": true,
180
+ "special": false
181
+ },
182
+ {
183
+ "id": 57535,
184
+ "content": "<s_width>",
185
+ "single_word": false,
186
+ "lstrip": false,
187
+ "rstrip": false,
188
+ "normalized": true,
189
+ "special": false
190
+ },
191
+ {
192
+ "id": 57536,
193
+ "content": "</s_width>",
194
+ "single_word": false,
195
+ "lstrip": false,
196
+ "rstrip": false,
197
+ "normalized": true,
198
+ "special": false
199
+ },
200
+ {
201
+ "id": 57537,
202
+ "content": "<s_height>",
203
+ "single_word": false,
204
+ "lstrip": false,
205
+ "rstrip": false,
206
+ "normalized": true,
207
+ "special": false
208
+ },
209
+ {
210
+ "id": 57538,
211
+ "content": "</s_height>",
212
+ "single_word": false,
213
+ "lstrip": false,
214
+ "rstrip": false,
215
+ "normalized": true,
216
+ "special": false
217
+ },
218
+ {
219
+ "id": 57539,
220
+ "content": "<s_words>",
221
+ "single_word": false,
222
+ "lstrip": false,
223
+ "rstrip": false,
224
+ "normalized": true,
225
+ "special": false
226
+ },
227
+ {
228
+ "id": 57540,
229
+ "content": "</s_words>",
230
+ "single_word": false,
231
+ "lstrip": false,
232
+ "rstrip": false,
233
+ "normalized": true,
234
+ "special": false
235
+ },
236
+ {
237
+ "id": 57541,
238
+ "content": "<s_rect>",
239
+ "single_word": false,
240
+ "lstrip": false,
241
+ "rstrip": false,
242
+ "normalized": true,
243
+ "special": false
244
+ },
245
+ {
246
+ "id": 57542,
247
+ "content": "</s_rect>",
248
+ "single_word": false,
249
+ "lstrip": false,
250
+ "rstrip": false,
251
+ "normalized": true,
252
+ "special": false
253
+ },
254
+ {
255
+ "id": 57543,
256
+ "content": "<s_x1>",
257
+ "single_word": false,
258
+ "lstrip": false,
259
+ "rstrip": false,
260
+ "normalized": true,
261
+ "special": false
262
+ },
263
+ {
264
+ "id": 57544,
265
+ "content": "</s_x1>",
266
+ "single_word": false,
267
+ "lstrip": false,
268
+ "rstrip": false,
269
+ "normalized": true,
270
+ "special": false
271
+ },
272
+ {
273
+ "id": 57545,
274
+ "content": "<s_y1>",
275
+ "single_word": false,
276
+ "lstrip": false,
277
+ "rstrip": false,
278
+ "normalized": true,
279
+ "special": false
280
+ },
281
+ {
282
+ "id": 57546,
283
+ "content": "</s_y1>",
284
+ "single_word": false,
285
+ "lstrip": false,
286
+ "rstrip": false,
287
+ "normalized": true,
288
+ "special": false
289
+ },
290
+ {
291
+ "id": 57547,
292
+ "content": "<s_x2>",
293
+ "single_word": false,
294
+ "lstrip": false,
295
+ "rstrip": false,
296
+ "normalized": true,
297
+ "special": false
298
+ },
299
+ {
300
+ "id": 57548,
301
+ "content": "</s_x2>",
302
+ "single_word": false,
303
+ "lstrip": false,
304
+ "rstrip": false,
305
+ "normalized": true,
306
+ "special": false
307
+ },
308
+ {
309
+ "id": 57549,
310
+ "content": "<s_y2>",
311
+ "single_word": false,
312
+ "lstrip": false,
313
+ "rstrip": false,
314
+ "normalized": true,
315
+ "special": false
316
+ },
317
+ {
318
+ "id": 57550,
319
+ "content": "</s_y2>",
320
+ "single_word": false,
321
+ "lstrip": false,
322
+ "rstrip": false,
323
+ "normalized": true,
324
+ "special": false
325
+ },
326
+ {
327
+ "id": 57551,
328
+ "content": "<s_value>",
329
+ "single_word": false,
330
+ "lstrip": false,
331
+ "rstrip": false,
332
+ "normalized": true,
333
+ "special": false
334
+ },
335
+ {
336
+ "id": 57552,
337
+ "content": "</s_value>",
338
+ "single_word": false,
339
+ "lstrip": false,
340
+ "rstrip": false,
341
+ "normalized": true,
342
+ "special": false
343
+ },
344
+ {
345
+ "id": 57553,
346
+ "content": "<s_label>",
347
+ "single_word": false,
348
+ "lstrip": false,
349
+ "rstrip": false,
350
+ "normalized": true,
351
+ "special": false
352
+ },
353
+ {
354
+ "id": 57554,
355
+ "content": "</s_label>",
356
+ "single_word": false,
357
+ "lstrip": false,
358
+ "rstrip": false,
359
+ "normalized": true,
360
+ "special": false
361
+ },
362
+ {
363
+ "id": 57555,
364
  "content": "<s_cord-v2>",
365
  "single_word": false,
366
  "lstrip": false,
tokenizer_config.json CHANGED
@@ -65,7 +65,7 @@
65
  "special": true
66
  },
67
  "57525": {
68
- "content": "<s_Liabilities>",
69
  "lstrip": false,
70
  "normalized": true,
71
  "rstrip": false,
@@ -73,7 +73,7 @@
73
  "special": false
74
  },
75
  "57526": {
76
- "content": "</s_Liabilities>",
77
  "lstrip": false,
78
  "normalized": true,
79
  "rstrip": false,
@@ -81,7 +81,7 @@
81
  "special": false
82
  },
83
  "57527": {
84
- "content": "<s_Assets>",
85
  "lstrip": false,
86
  "normalized": true,
87
  "rstrip": false,
@@ -89,7 +89,7 @@
89
  "special": false
90
  },
91
  "57528": {
92
- "content": "</s_Assets>",
93
  "lstrip": false,
94
  "normalized": true,
95
  "rstrip": false,
@@ -97,7 +97,7 @@
97
  "special": false
98
  },
99
  "57529": {
100
- "content": "<s_Misc>",
101
  "lstrip": false,
102
  "normalized": true,
103
  "rstrip": false,
@@ -105,7 +105,7 @@
105
  "special": false
106
  },
107
  "57530": {
108
- "content": "</s_Misc>",
109
  "lstrip": false,
110
  "normalized": true,
111
  "rstrip": false,
@@ -113,6 +113,198 @@
113
  "special": false
114
  },
115
  "57531": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
116
  "content": "<s_cord-v2>",
117
  "lstrip": false,
118
  "normalized": true,
 
65
  "special": true
66
  },
67
  "57525": {
68
+ "content": "<s_meta>",
69
  "lstrip": false,
70
  "normalized": true,
71
  "rstrip": false,
 
73
  "special": false
74
  },
75
  "57526": {
76
+ "content": "</s_meta>",
77
  "lstrip": false,
78
  "normalized": true,
79
  "rstrip": false,
 
81
  "special": false
82
  },
83
  "57527": {
84
+ "content": "<s_version>",
85
  "lstrip": false,
86
  "normalized": true,
87
  "rstrip": false,
 
89
  "special": false
90
  },
91
  "57528": {
92
+ "content": "</s_version>",
93
  "lstrip": false,
94
  "normalized": true,
95
  "rstrip": false,
 
97
  "special": false
98
  },
99
  "57529": {
100
+ "content": "<s_split>",
101
  "lstrip": false,
102
  "normalized": true,
103
  "rstrip": false,
 
105
  "special": false
106
  },
107
  "57530": {
108
+ "content": "</s_split>",
109
  "lstrip": false,
110
  "normalized": true,
111
  "rstrip": false,
 
113
  "special": false
114
  },
115
  "57531": {
116
+ "content": "<s_image_id>",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "57532": {
124
+ "content": "</s_image_id>",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "57533": {
132
+ "content": "<s_image_size>",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "57534": {
140
+ "content": "</s_image_size>",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "57535": {
148
+ "content": "<s_width>",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "57536": {
156
+ "content": "</s_width>",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "57537": {
164
+ "content": "<s_height>",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "57538": {
172
+ "content": "</s_height>",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "57539": {
180
+ "content": "<s_words>",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "57540": {
188
+ "content": "</s_words>",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "57541": {
196
+ "content": "<s_rect>",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "57542": {
204
+ "content": "</s_rect>",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "57543": {
212
+ "content": "<s_x1>",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "57544": {
220
+ "content": "</s_x1>",
221
+ "lstrip": false,
222
+ "normalized": true,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": false
226
+ },
227
+ "57545": {
228
+ "content": "<s_y1>",
229
+ "lstrip": false,
230
+ "normalized": true,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": false
234
+ },
235
+ "57546": {
236
+ "content": "</s_y1>",
237
+ "lstrip": false,
238
+ "normalized": true,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": false
242
+ },
243
+ "57547": {
244
+ "content": "<s_x2>",
245
+ "lstrip": false,
246
+ "normalized": true,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": false
250
+ },
251
+ "57548": {
252
+ "content": "</s_x2>",
253
+ "lstrip": false,
254
+ "normalized": true,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": false
258
+ },
259
+ "57549": {
260
+ "content": "<s_y2>",
261
+ "lstrip": false,
262
+ "normalized": true,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": false
266
+ },
267
+ "57550": {
268
+ "content": "</s_y2>",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "57551": {
276
+ "content": "<s_value>",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "57552": {
284
+ "content": "</s_value>",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "57553": {
292
+ "content": "<s_label>",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "57554": {
300
+ "content": "</s_label>",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "57555": {
308
  "content": "<s_cord-v2>",
309
  "lstrip": false,
310
  "normalized": true,