Training done
Browse files- added_tokens.json +13 -7
- special_tokens_map.json +42 -6
- tokenizer.json +64 -8
- tokenizer_config.json +54 -6
added_tokens.json
CHANGED
|
@@ -2,29 +2,34 @@
|
|
| 2 |
"</s_acount>": 57544,
|
| 3 |
"</s_address>": 57528,
|
| 4 |
"</s_barcode>": 57540,
|
| 5 |
-
"</
|
|
|
|
|
|
|
| 6 |
"</s_city>": 57536,
|
| 7 |
"</s_colony>": 57532,
|
| 8 |
"</s_fullAddress>": 57526,
|
| 9 |
-
"</s_name>":
|
| 10 |
"</s_phone>": 57542,
|
| 11 |
"</s_reference>": 57546,
|
| 12 |
"</s_state>": 57538,
|
| 13 |
"</s_street>": 57530,
|
| 14 |
"</s_totalPaymentLower>": 57550,
|
| 15 |
"</s_totalPaymentTop>": 57548,
|
| 16 |
-
"</
|
|
|
|
| 17 |
"</s_zipCode>": 57534,
|
| 18 |
"<s_acount>": 57543,
|
| 19 |
"<s_address>": 57527,
|
| 20 |
"<s_barcode>": 57539,
|
| 21 |
-
"<
|
|
|
|
|
|
|
| 22 |
"<s_city>": 57535,
|
| 23 |
"<s_colony>": 57531,
|
| 24 |
-
"<s_cord-v2>":
|
| 25 |
"<s_fullAddress>": 57525,
|
| 26 |
"<s_iitcdip>": 57523,
|
| 27 |
-
"<s_name>":
|
| 28 |
"<s_phone>": 57541,
|
| 29 |
"<s_reference>": 57545,
|
| 30 |
"<s_state>": 57537,
|
|
@@ -32,7 +37,8 @@
|
|
| 32 |
"<s_synthdog>": 57524,
|
| 33 |
"<s_totalPaymentLower>": 57549,
|
| 34 |
"<s_totalPaymentTop>": 57547,
|
| 35 |
-
"<
|
|
|
|
| 36 |
"<s_zipCode>": 57533,
|
| 37 |
"<sep/>": 57522
|
| 38 |
}
|
|
|
|
| 2 |
"</s_acount>": 57544,
|
| 3 |
"</s_address>": 57528,
|
| 4 |
"</s_barcode>": 57540,
|
| 5 |
+
"</s_caption1>": 57556,
|
| 6 |
+
"</s_caption2>": 57558,
|
| 7 |
+
"</s_caption3>": 57560,
|
| 8 |
"</s_city>": 57536,
|
| 9 |
"</s_colony>": 57532,
|
| 10 |
"</s_fullAddress>": 57526,
|
| 11 |
+
"</s_name>": 57554,
|
| 12 |
"</s_phone>": 57542,
|
| 13 |
"</s_reference>": 57546,
|
| 14 |
"</s_state>": 57538,
|
| 15 |
"</s_street>": 57530,
|
| 16 |
"</s_totalPaymentLower>": 57550,
|
| 17 |
"</s_totalPaymentTop>": 57548,
|
| 18 |
+
"</s_totalPaymentletter>": 57552,
|
| 19 |
+
"</s_type>": 57562,
|
| 20 |
"</s_zipCode>": 57534,
|
| 21 |
"<s_acount>": 57543,
|
| 22 |
"<s_address>": 57527,
|
| 23 |
"<s_barcode>": 57539,
|
| 24 |
+
"<s_caption1>": 57555,
|
| 25 |
+
"<s_caption2>": 57557,
|
| 26 |
+
"<s_caption3>": 57559,
|
| 27 |
"<s_city>": 57535,
|
| 28 |
"<s_colony>": 57531,
|
| 29 |
+
"<s_cord-v2>": 57563,
|
| 30 |
"<s_fullAddress>": 57525,
|
| 31 |
"<s_iitcdip>": 57523,
|
| 32 |
+
"<s_name>": 57553,
|
| 33 |
"<s_phone>": 57541,
|
| 34 |
"<s_reference>": 57545,
|
| 35 |
"<s_state>": 57537,
|
|
|
|
| 37 |
"<s_synthdog>": 57524,
|
| 38 |
"<s_totalPaymentLower>": 57549,
|
| 39 |
"<s_totalPaymentTop>": 57547,
|
| 40 |
+
"<s_totalPaymentletter>": 57551,
|
| 41 |
+
"<s_type>": 57561,
|
| 42 |
"<s_zipCode>": 57533,
|
| 43 |
"<sep/>": 57522
|
| 44 |
}
|
special_tokens_map.json
CHANGED
|
@@ -3,9 +3,27 @@
|
|
| 3 |
"<s_iitcdip>",
|
| 4 |
"<s_synthdog>"
|
| 5 |
],
|
| 6 |
-
"bos_token":
|
| 7 |
-
|
| 8 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"mask_token": {
|
| 10 |
"content": "<mask>",
|
| 11 |
"lstrip": true,
|
|
@@ -13,7 +31,25 @@
|
|
| 13 |
"rstrip": false,
|
| 14 |
"single_word": false
|
| 15 |
},
|
| 16 |
-
"pad_token":
|
| 17 |
-
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
}
|
|
|
|
| 3 |
"<s_iitcdip>",
|
| 4 |
"<s_synthdog>"
|
| 5 |
],
|
| 6 |
+
"bos_token": {
|
| 7 |
+
"content": "<s>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"cls_token": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
},
|
| 20 |
+
"eos_token": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false
|
| 26 |
+
},
|
| 27 |
"mask_token": {
|
| 28 |
"content": "<mask>",
|
| 29 |
"lstrip": true,
|
|
|
|
| 31 |
"rstrip": false,
|
| 32 |
"single_word": false
|
| 33 |
},
|
| 34 |
+
"pad_token": {
|
| 35 |
+
"content": "<pad>",
|
| 36 |
+
"lstrip": false,
|
| 37 |
+
"normalized": false,
|
| 38 |
+
"rstrip": false,
|
| 39 |
+
"single_word": false
|
| 40 |
+
},
|
| 41 |
+
"sep_token": {
|
| 42 |
+
"content": "</s>",
|
| 43 |
+
"lstrip": false,
|
| 44 |
+
"normalized": false,
|
| 45 |
+
"rstrip": false,
|
| 46 |
+
"single_word": false
|
| 47 |
+
},
|
| 48 |
+
"unk_token": {
|
| 49 |
+
"content": "<unk>",
|
| 50 |
+
"lstrip": false,
|
| 51 |
+
"normalized": false,
|
| 52 |
+
"rstrip": false,
|
| 53 |
+
"single_word": false
|
| 54 |
+
}
|
| 55 |
}
|
tokenizer.json
CHANGED
|
@@ -325,7 +325,7 @@
|
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"id": 57551,
|
| 328 |
-
"content": "<
|
| 329 |
"single_word": false,
|
| 330 |
"lstrip": false,
|
| 331 |
"rstrip": false,
|
|
@@ -334,7 +334,7 @@
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"id": 57552,
|
| 337 |
-
"content": "</
|
| 338 |
"single_word": false,
|
| 339 |
"lstrip": false,
|
| 340 |
"rstrip": false,
|
|
@@ -343,7 +343,7 @@
|
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"id": 57553,
|
| 346 |
-
"content": "<
|
| 347 |
"single_word": false,
|
| 348 |
"lstrip": false,
|
| 349 |
"rstrip": false,
|
|
@@ -352,7 +352,7 @@
|
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"id": 57554,
|
| 355 |
-
"content": "</
|
| 356 |
"single_word": false,
|
| 357 |
"lstrip": false,
|
| 358 |
"rstrip": false,
|
|
@@ -361,7 +361,7 @@
|
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"id": 57555,
|
| 364 |
-
"content": "<
|
| 365 |
"single_word": false,
|
| 366 |
"lstrip": false,
|
| 367 |
"rstrip": false,
|
|
@@ -370,7 +370,7 @@
|
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"id": 57556,
|
| 373 |
-
"content": "</
|
| 374 |
"single_word": false,
|
| 375 |
"lstrip": false,
|
| 376 |
"rstrip": false,
|
|
@@ -379,6 +379,60 @@
|
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"id": 57557,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
"content": "<s_cord-v2>",
|
| 383 |
"single_word": false,
|
| 384 |
"lstrip": false,
|
|
@@ -406,7 +460,8 @@
|
|
| 406 |
"pre_tokenizer": {
|
| 407 |
"type": "Metaspace",
|
| 408 |
"replacement": "▁",
|
| 409 |
-
"add_prefix_space": true
|
|
|
|
| 410 |
},
|
| 411 |
"post_processor": {
|
| 412 |
"type": "TemplateProcessing",
|
|
@@ -492,7 +547,8 @@
|
|
| 492 |
"decoder": {
|
| 493 |
"type": "Metaspace",
|
| 494 |
"replacement": "▁",
|
| 495 |
-
"add_prefix_space": true
|
|
|
|
| 496 |
},
|
| 497 |
"model": {
|
| 498 |
"type": "Unigram",
|
|
|
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"id": 57551,
|
| 328 |
+
"content": "<s_totalPaymentletter>",
|
| 329 |
"single_word": false,
|
| 330 |
"lstrip": false,
|
| 331 |
"rstrip": false,
|
|
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"id": 57552,
|
| 337 |
+
"content": "</s_totalPaymentletter>",
|
| 338 |
"single_word": false,
|
| 339 |
"lstrip": false,
|
| 340 |
"rstrip": false,
|
|
|
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"id": 57553,
|
| 346 |
+
"content": "<s_name>",
|
| 347 |
"single_word": false,
|
| 348 |
"lstrip": false,
|
| 349 |
"rstrip": false,
|
|
|
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"id": 57554,
|
| 355 |
+
"content": "</s_name>",
|
| 356 |
"single_word": false,
|
| 357 |
"lstrip": false,
|
| 358 |
"rstrip": false,
|
|
|
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"id": 57555,
|
| 364 |
+
"content": "<s_caption1>",
|
| 365 |
"single_word": false,
|
| 366 |
"lstrip": false,
|
| 367 |
"rstrip": false,
|
|
|
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"id": 57556,
|
| 373 |
+
"content": "</s_caption1>",
|
| 374 |
"single_word": false,
|
| 375 |
"lstrip": false,
|
| 376 |
"rstrip": false,
|
|
|
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"id": 57557,
|
| 382 |
+
"content": "<s_caption2>",
|
| 383 |
+
"single_word": false,
|
| 384 |
+
"lstrip": false,
|
| 385 |
+
"rstrip": false,
|
| 386 |
+
"normalized": true,
|
| 387 |
+
"special": false
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"id": 57558,
|
| 391 |
+
"content": "</s_caption2>",
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"lstrip": false,
|
| 394 |
+
"rstrip": false,
|
| 395 |
+
"normalized": true,
|
| 396 |
+
"special": false
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"id": 57559,
|
| 400 |
+
"content": "<s_caption3>",
|
| 401 |
+
"single_word": false,
|
| 402 |
+
"lstrip": false,
|
| 403 |
+
"rstrip": false,
|
| 404 |
+
"normalized": true,
|
| 405 |
+
"special": false
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"id": 57560,
|
| 409 |
+
"content": "</s_caption3>",
|
| 410 |
+
"single_word": false,
|
| 411 |
+
"lstrip": false,
|
| 412 |
+
"rstrip": false,
|
| 413 |
+
"normalized": true,
|
| 414 |
+
"special": false
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"id": 57561,
|
| 418 |
+
"content": "<s_type>",
|
| 419 |
+
"single_word": false,
|
| 420 |
+
"lstrip": false,
|
| 421 |
+
"rstrip": false,
|
| 422 |
+
"normalized": true,
|
| 423 |
+
"special": false
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"id": 57562,
|
| 427 |
+
"content": "</s_type>",
|
| 428 |
+
"single_word": false,
|
| 429 |
+
"lstrip": false,
|
| 430 |
+
"rstrip": false,
|
| 431 |
+
"normalized": true,
|
| 432 |
+
"special": false
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"id": 57563,
|
| 436 |
"content": "<s_cord-v2>",
|
| 437 |
"single_word": false,
|
| 438 |
"lstrip": false,
|
|
|
|
| 460 |
"pre_tokenizer": {
|
| 461 |
"type": "Metaspace",
|
| 462 |
"replacement": "▁",
|
| 463 |
+
"add_prefix_space": true,
|
| 464 |
+
"prepend_scheme": "always"
|
| 465 |
},
|
| 466 |
"post_processor": {
|
| 467 |
"type": "TemplateProcessing",
|
|
|
|
| 547 |
"decoder": {
|
| 548 |
"type": "Metaspace",
|
| 549 |
"replacement": "▁",
|
| 550 |
+
"add_prefix_space": true,
|
| 551 |
+
"prepend_scheme": "always"
|
| 552 |
},
|
| 553 |
"model": {
|
| 554 |
"type": "Unigram",
|
tokenizer_config.json
CHANGED
|
@@ -273,7 +273,7 @@
|
|
| 273 |
"special": false
|
| 274 |
},
|
| 275 |
"57551": {
|
| 276 |
-
"content": "<
|
| 277 |
"lstrip": false,
|
| 278 |
"normalized": true,
|
| 279 |
"rstrip": false,
|
|
@@ -281,7 +281,7 @@
|
|
| 281 |
"special": false
|
| 282 |
},
|
| 283 |
"57552": {
|
| 284 |
-
"content": "</
|
| 285 |
"lstrip": false,
|
| 286 |
"normalized": true,
|
| 287 |
"rstrip": false,
|
|
@@ -289,7 +289,7 @@
|
|
| 289 |
"special": false
|
| 290 |
},
|
| 291 |
"57553": {
|
| 292 |
-
"content": "<
|
| 293 |
"lstrip": false,
|
| 294 |
"normalized": true,
|
| 295 |
"rstrip": false,
|
|
@@ -297,7 +297,7 @@
|
|
| 297 |
"special": false
|
| 298 |
},
|
| 299 |
"57554": {
|
| 300 |
-
"content": "</
|
| 301 |
"lstrip": false,
|
| 302 |
"normalized": true,
|
| 303 |
"rstrip": false,
|
|
@@ -305,7 +305,7 @@
|
|
| 305 |
"special": false
|
| 306 |
},
|
| 307 |
"57555": {
|
| 308 |
-
"content": "<
|
| 309 |
"lstrip": false,
|
| 310 |
"normalized": true,
|
| 311 |
"rstrip": false,
|
|
@@ -313,7 +313,7 @@
|
|
| 313 |
"special": false
|
| 314 |
},
|
| 315 |
"57556": {
|
| 316 |
-
"content": "</
|
| 317 |
"lstrip": false,
|
| 318 |
"normalized": true,
|
| 319 |
"rstrip": false,
|
|
@@ -321,6 +321,54 @@
|
|
| 321 |
"special": false
|
| 322 |
},
|
| 323 |
"57557": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"content": "<s_cord-v2>",
|
| 325 |
"lstrip": false,
|
| 326 |
"normalized": true,
|
|
|
|
| 273 |
"special": false
|
| 274 |
},
|
| 275 |
"57551": {
|
| 276 |
+
"content": "<s_totalPaymentletter>",
|
| 277 |
"lstrip": false,
|
| 278 |
"normalized": true,
|
| 279 |
"rstrip": false,
|
|
|
|
| 281 |
"special": false
|
| 282 |
},
|
| 283 |
"57552": {
|
| 284 |
+
"content": "</s_totalPaymentletter>",
|
| 285 |
"lstrip": false,
|
| 286 |
"normalized": true,
|
| 287 |
"rstrip": false,
|
|
|
|
| 289 |
"special": false
|
| 290 |
},
|
| 291 |
"57553": {
|
| 292 |
+
"content": "<s_name>",
|
| 293 |
"lstrip": false,
|
| 294 |
"normalized": true,
|
| 295 |
"rstrip": false,
|
|
|
|
| 297 |
"special": false
|
| 298 |
},
|
| 299 |
"57554": {
|
| 300 |
+
"content": "</s_name>",
|
| 301 |
"lstrip": false,
|
| 302 |
"normalized": true,
|
| 303 |
"rstrip": false,
|
|
|
|
| 305 |
"special": false
|
| 306 |
},
|
| 307 |
"57555": {
|
| 308 |
+
"content": "<s_caption1>",
|
| 309 |
"lstrip": false,
|
| 310 |
"normalized": true,
|
| 311 |
"rstrip": false,
|
|
|
|
| 313 |
"special": false
|
| 314 |
},
|
| 315 |
"57556": {
|
| 316 |
+
"content": "</s_caption1>",
|
| 317 |
"lstrip": false,
|
| 318 |
"normalized": true,
|
| 319 |
"rstrip": false,
|
|
|
|
| 321 |
"special": false
|
| 322 |
},
|
| 323 |
"57557": {
|
| 324 |
+
"content": "<s_caption2>",
|
| 325 |
+
"lstrip": false,
|
| 326 |
+
"normalized": true,
|
| 327 |
+
"rstrip": false,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": false
|
| 330 |
+
},
|
| 331 |
+
"57558": {
|
| 332 |
+
"content": "</s_caption2>",
|
| 333 |
+
"lstrip": false,
|
| 334 |
+
"normalized": true,
|
| 335 |
+
"rstrip": false,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": false
|
| 338 |
+
},
|
| 339 |
+
"57559": {
|
| 340 |
+
"content": "<s_caption3>",
|
| 341 |
+
"lstrip": false,
|
| 342 |
+
"normalized": true,
|
| 343 |
+
"rstrip": false,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": false
|
| 346 |
+
},
|
| 347 |
+
"57560": {
|
| 348 |
+
"content": "</s_caption3>",
|
| 349 |
+
"lstrip": false,
|
| 350 |
+
"normalized": true,
|
| 351 |
+
"rstrip": false,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": false
|
| 354 |
+
},
|
| 355 |
+
"57561": {
|
| 356 |
+
"content": "<s_type>",
|
| 357 |
+
"lstrip": false,
|
| 358 |
+
"normalized": true,
|
| 359 |
+
"rstrip": false,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": false
|
| 362 |
+
},
|
| 363 |
+
"57562": {
|
| 364 |
+
"content": "</s_type>",
|
| 365 |
+
"lstrip": false,
|
| 366 |
+
"normalized": true,
|
| 367 |
+
"rstrip": false,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": false
|
| 370 |
+
},
|
| 371 |
+
"57563": {
|
| 372 |
"content": "<s_cord-v2>",
|
| 373 |
"lstrip": false,
|
| 374 |
"normalized": true,
|