File size: 36,888 Bytes
373a3f3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 |
2023-10-23 19:29:49,570 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,571 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(1): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(2): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(3): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(4): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(5): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(6): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(7): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(8): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(9): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(10): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(11): BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-23 19:29:49,571 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,571 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /home/ubuntu/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-23 19:29:49,571 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,571 Train: 966 sentences
2023-10-23 19:29:49,571 (train_with_dev=False, train_with_test=False)
2023-10-23 19:29:49,571 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,571 Training Params:
2023-10-23 19:29:49,571 - learning_rate: "3e-05"
2023-10-23 19:29:49,571 - mini_batch_size: "4"
2023-10-23 19:29:49,571 - max_epochs: "10"
2023-10-23 19:29:49,571 - shuffle: "True"
2023-10-23 19:29:49,571 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,571 Plugins:
2023-10-23 19:29:49,571 - TensorboardLogger
2023-10-23 19:29:49,571 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 19:29:49,571 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,572 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 19:29:49,572 - metric: "('micro avg', 'f1-score')"
2023-10-23 19:29:49,572 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,572 Computation:
2023-10-23 19:29:49,572 - compute on device: cuda:0
2023-10-23 19:29:49,572 - embedding storage: none
2023-10-23 19:29:49,572 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,572 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-23 19:29:49,572 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,572 ----------------------------------------------------------------------------------------------------
2023-10-23 19:29:49,572 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 19:29:51,021 epoch 1 - iter 24/242 - loss 2.98287844 - time (sec): 1.45 - samples/sec: 1489.27 - lr: 0.000003 - momentum: 0.000000
2023-10-23 19:29:52,539 epoch 1 - iter 48/242 - loss 2.26117522 - time (sec): 2.97 - samples/sec: 1617.92 - lr: 0.000006 - momentum: 0.000000
2023-10-23 19:29:54,030 epoch 1 - iter 72/242 - loss 1.76926638 - time (sec): 4.46 - samples/sec: 1573.78 - lr: 0.000009 - momentum: 0.000000
2023-10-23 19:29:55,538 epoch 1 - iter 96/242 - loss 1.46345492 - time (sec): 5.97 - samples/sec: 1585.41 - lr: 0.000012 - momentum: 0.000000
2023-10-23 19:29:57,026 epoch 1 - iter 120/242 - loss 1.25478029 - time (sec): 7.45 - samples/sec: 1587.11 - lr: 0.000015 - momentum: 0.000000
2023-10-23 19:29:58,565 epoch 1 - iter 144/242 - loss 1.09074253 - time (sec): 8.99 - samples/sec: 1608.26 - lr: 0.000018 - momentum: 0.000000
2023-10-23 19:30:00,155 epoch 1 - iter 168/242 - loss 0.98154842 - time (sec): 10.58 - samples/sec: 1618.75 - lr: 0.000021 - momentum: 0.000000
2023-10-23 19:30:01,639 epoch 1 - iter 192/242 - loss 0.89632668 - time (sec): 12.07 - samples/sec: 1618.33 - lr: 0.000024 - momentum: 0.000000
2023-10-23 19:30:03,136 epoch 1 - iter 216/242 - loss 0.82169295 - time (sec): 13.56 - samples/sec: 1612.00 - lr: 0.000027 - momentum: 0.000000
2023-10-23 19:30:04,692 epoch 1 - iter 240/242 - loss 0.75910623 - time (sec): 15.12 - samples/sec: 1620.78 - lr: 0.000030 - momentum: 0.000000
2023-10-23 19:30:04,820 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:04,820 EPOCH 1 done: loss 0.7531 - lr: 0.000030
2023-10-23 19:30:05,632 DEV : loss 0.1827384978532791 - f1-score (micro avg) 0.6213
2023-10-23 19:30:05,636 saving best model
2023-10-23 19:30:06,104 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:07,610 epoch 2 - iter 24/242 - loss 0.13604524 - time (sec): 1.51 - samples/sec: 1685.70 - lr: 0.000030 - momentum: 0.000000
2023-10-23 19:30:09,092 epoch 2 - iter 48/242 - loss 0.16051416 - time (sec): 2.99 - samples/sec: 1606.50 - lr: 0.000029 - momentum: 0.000000
2023-10-23 19:30:10,571 epoch 2 - iter 72/242 - loss 0.16266687 - time (sec): 4.47 - samples/sec: 1587.93 - lr: 0.000029 - momentum: 0.000000
2023-10-23 19:30:12,084 epoch 2 - iter 96/242 - loss 0.16915945 - time (sec): 5.98 - samples/sec: 1586.67 - lr: 0.000029 - momentum: 0.000000
2023-10-23 19:30:13,664 epoch 2 - iter 120/242 - loss 0.16054725 - time (sec): 7.56 - samples/sec: 1614.35 - lr: 0.000028 - momentum: 0.000000
2023-10-23 19:30:15,218 epoch 2 - iter 144/242 - loss 0.15751912 - time (sec): 9.11 - samples/sec: 1619.90 - lr: 0.000028 - momentum: 0.000000
2023-10-23 19:30:16,764 epoch 2 - iter 168/242 - loss 0.15934129 - time (sec): 10.66 - samples/sec: 1625.26 - lr: 0.000028 - momentum: 0.000000
2023-10-23 19:30:18,272 epoch 2 - iter 192/242 - loss 0.15641142 - time (sec): 12.17 - samples/sec: 1617.13 - lr: 0.000027 - momentum: 0.000000
2023-10-23 19:30:19,801 epoch 2 - iter 216/242 - loss 0.15506669 - time (sec): 13.70 - samples/sec: 1623.99 - lr: 0.000027 - momentum: 0.000000
2023-10-23 19:30:21,317 epoch 2 - iter 240/242 - loss 0.15220355 - time (sec): 15.21 - samples/sec: 1617.97 - lr: 0.000027 - momentum: 0.000000
2023-10-23 19:30:21,431 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:21,431 EPOCH 2 done: loss 0.1516 - lr: 0.000027
2023-10-23 19:30:22,122 DEV : loss 0.12770959734916687 - f1-score (micro avg) 0.7935
2023-10-23 19:30:22,125 saving best model
2023-10-23 19:30:22,824 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:24,302 epoch 3 - iter 24/242 - loss 0.12719161 - time (sec): 1.48 - samples/sec: 1577.87 - lr: 0.000026 - momentum: 0.000000
2023-10-23 19:30:25,824 epoch 3 - iter 48/242 - loss 0.11257977 - time (sec): 3.00 - samples/sec: 1578.28 - lr: 0.000026 - momentum: 0.000000
2023-10-23 19:30:27,346 epoch 3 - iter 72/242 - loss 0.11151491 - time (sec): 4.52 - samples/sec: 1628.09 - lr: 0.000026 - momentum: 0.000000
2023-10-23 19:30:28,909 epoch 3 - iter 96/242 - loss 0.10559088 - time (sec): 6.08 - samples/sec: 1644.85 - lr: 0.000025 - momentum: 0.000000
2023-10-23 19:30:30,427 epoch 3 - iter 120/242 - loss 0.09317932 - time (sec): 7.60 - samples/sec: 1671.53 - lr: 0.000025 - momentum: 0.000000
2023-10-23 19:30:31,909 epoch 3 - iter 144/242 - loss 0.09460460 - time (sec): 9.08 - samples/sec: 1652.42 - lr: 0.000025 - momentum: 0.000000
2023-10-23 19:30:33,459 epoch 3 - iter 168/242 - loss 0.09435856 - time (sec): 10.63 - samples/sec: 1643.72 - lr: 0.000024 - momentum: 0.000000
2023-10-23 19:30:34,949 epoch 3 - iter 192/242 - loss 0.09314101 - time (sec): 12.12 - samples/sec: 1621.29 - lr: 0.000024 - momentum: 0.000000
2023-10-23 19:30:36,529 epoch 3 - iter 216/242 - loss 0.10155585 - time (sec): 13.70 - samples/sec: 1619.78 - lr: 0.000024 - momentum: 0.000000
2023-10-23 19:30:38,045 epoch 3 - iter 240/242 - loss 0.09838725 - time (sec): 15.22 - samples/sec: 1619.20 - lr: 0.000023 - momentum: 0.000000
2023-10-23 19:30:38,158 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:38,159 EPOCH 3 done: loss 0.0980 - lr: 0.000023
2023-10-23 19:30:38,852 DEV : loss 0.1285228729248047 - f1-score (micro avg) 0.8362
2023-10-23 19:30:38,856 saving best model
2023-10-23 19:30:39,547 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:41,015 epoch 4 - iter 24/242 - loss 0.04685614 - time (sec): 1.47 - samples/sec: 1580.95 - lr: 0.000023 - momentum: 0.000000
2023-10-23 19:30:42,569 epoch 4 - iter 48/242 - loss 0.08110896 - time (sec): 3.02 - samples/sec: 1600.25 - lr: 0.000023 - momentum: 0.000000
2023-10-23 19:30:44,072 epoch 4 - iter 72/242 - loss 0.07997580 - time (sec): 4.52 - samples/sec: 1640.82 - lr: 0.000022 - momentum: 0.000000
2023-10-23 19:30:45,644 epoch 4 - iter 96/242 - loss 0.07417430 - time (sec): 6.10 - samples/sec: 1628.90 - lr: 0.000022 - momentum: 0.000000
2023-10-23 19:30:47,137 epoch 4 - iter 120/242 - loss 0.06583957 - time (sec): 7.59 - samples/sec: 1608.25 - lr: 0.000022 - momentum: 0.000000
2023-10-23 19:30:48,700 epoch 4 - iter 144/242 - loss 0.06891605 - time (sec): 9.15 - samples/sec: 1633.63 - lr: 0.000021 - momentum: 0.000000
2023-10-23 19:30:50,232 epoch 4 - iter 168/242 - loss 0.06741911 - time (sec): 10.68 - samples/sec: 1628.92 - lr: 0.000021 - momentum: 0.000000
2023-10-23 19:30:51,785 epoch 4 - iter 192/242 - loss 0.06826652 - time (sec): 12.24 - samples/sec: 1620.78 - lr: 0.000021 - momentum: 0.000000
2023-10-23 19:30:53,271 epoch 4 - iter 216/242 - loss 0.06909254 - time (sec): 13.72 - samples/sec: 1608.79 - lr: 0.000020 - momentum: 0.000000
2023-10-23 19:30:54,785 epoch 4 - iter 240/242 - loss 0.06791650 - time (sec): 15.24 - samples/sec: 1612.41 - lr: 0.000020 - momentum: 0.000000
2023-10-23 19:30:54,901 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:54,902 EPOCH 4 done: loss 0.0675 - lr: 0.000020
2023-10-23 19:30:55,598 DEV : loss 0.16268931329250336 - f1-score (micro avg) 0.8259
2023-10-23 19:30:55,602 ----------------------------------------------------------------------------------------------------
2023-10-23 19:30:57,140 epoch 5 - iter 24/242 - loss 0.05705890 - time (sec): 1.54 - samples/sec: 1606.57 - lr: 0.000020 - momentum: 0.000000
2023-10-23 19:30:58,668 epoch 5 - iter 48/242 - loss 0.05415032 - time (sec): 3.07 - samples/sec: 1623.01 - lr: 0.000019 - momentum: 0.000000
2023-10-23 19:31:00,159 epoch 5 - iter 72/242 - loss 0.05768220 - time (sec): 4.56 - samples/sec: 1624.60 - lr: 0.000019 - momentum: 0.000000
2023-10-23 19:31:01,659 epoch 5 - iter 96/242 - loss 0.05428573 - time (sec): 6.06 - samples/sec: 1634.34 - lr: 0.000019 - momentum: 0.000000
2023-10-23 19:31:03,167 epoch 5 - iter 120/242 - loss 0.05240841 - time (sec): 7.56 - samples/sec: 1637.80 - lr: 0.000018 - momentum: 0.000000
2023-10-23 19:31:04,711 epoch 5 - iter 144/242 - loss 0.04914866 - time (sec): 9.11 - samples/sec: 1636.15 - lr: 0.000018 - momentum: 0.000000
2023-10-23 19:31:06,237 epoch 5 - iter 168/242 - loss 0.04959201 - time (sec): 10.64 - samples/sec: 1630.73 - lr: 0.000018 - momentum: 0.000000
2023-10-23 19:31:07,753 epoch 5 - iter 192/242 - loss 0.04995992 - time (sec): 12.15 - samples/sec: 1605.06 - lr: 0.000017 - momentum: 0.000000
2023-10-23 19:31:09,284 epoch 5 - iter 216/242 - loss 0.04891403 - time (sec): 13.68 - samples/sec: 1612.94 - lr: 0.000017 - momentum: 0.000000
2023-10-23 19:31:10,852 epoch 5 - iter 240/242 - loss 0.04560757 - time (sec): 15.25 - samples/sec: 1615.49 - lr: 0.000017 - momentum: 0.000000
2023-10-23 19:31:10,962 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:10,962 EPOCH 5 done: loss 0.0457 - lr: 0.000017
2023-10-23 19:31:11,659 DEV : loss 0.1693263053894043 - f1-score (micro avg) 0.8425
2023-10-23 19:31:11,663 saving best model
2023-10-23 19:31:12,458 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:13,964 epoch 6 - iter 24/242 - loss 0.00754029 - time (sec): 1.50 - samples/sec: 1492.40 - lr: 0.000016 - momentum: 0.000000
2023-10-23 19:31:15,436 epoch 6 - iter 48/242 - loss 0.02913193 - time (sec): 2.98 - samples/sec: 1546.80 - lr: 0.000016 - momentum: 0.000000
2023-10-23 19:31:16,992 epoch 6 - iter 72/242 - loss 0.03202316 - time (sec): 4.53 - samples/sec: 1616.66 - lr: 0.000016 - momentum: 0.000000
2023-10-23 19:31:18,538 epoch 6 - iter 96/242 - loss 0.03013955 - time (sec): 6.08 - samples/sec: 1599.96 - lr: 0.000015 - momentum: 0.000000
2023-10-23 19:31:20,088 epoch 6 - iter 120/242 - loss 0.03080415 - time (sec): 7.63 - samples/sec: 1645.91 - lr: 0.000015 - momentum: 0.000000
2023-10-23 19:31:21,638 epoch 6 - iter 144/242 - loss 0.03003144 - time (sec): 9.18 - samples/sec: 1655.31 - lr: 0.000015 - momentum: 0.000000
2023-10-23 19:31:23,126 epoch 6 - iter 168/242 - loss 0.03283920 - time (sec): 10.67 - samples/sec: 1630.94 - lr: 0.000014 - momentum: 0.000000
2023-10-23 19:31:24,642 epoch 6 - iter 192/242 - loss 0.03395577 - time (sec): 12.18 - samples/sec: 1627.18 - lr: 0.000014 - momentum: 0.000000
2023-10-23 19:31:26,111 epoch 6 - iter 216/242 - loss 0.03345099 - time (sec): 13.65 - samples/sec: 1613.56 - lr: 0.000014 - momentum: 0.000000
2023-10-23 19:31:27,662 epoch 6 - iter 240/242 - loss 0.03181677 - time (sec): 15.20 - samples/sec: 1615.60 - lr: 0.000013 - momentum: 0.000000
2023-10-23 19:31:27,785 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:27,786 EPOCH 6 done: loss 0.0322 - lr: 0.000013
2023-10-23 19:31:28,484 DEV : loss 0.1637829840183258 - f1-score (micro avg) 0.8688
2023-10-23 19:31:28,488 saving best model
2023-10-23 19:31:29,085 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:30,628 epoch 7 - iter 24/242 - loss 0.01992873 - time (sec): 1.54 - samples/sec: 1671.67 - lr: 0.000013 - momentum: 0.000000
2023-10-23 19:31:32,109 epoch 7 - iter 48/242 - loss 0.01958366 - time (sec): 3.02 - samples/sec: 1562.88 - lr: 0.000013 - momentum: 0.000000
2023-10-23 19:31:33,628 epoch 7 - iter 72/242 - loss 0.02160298 - time (sec): 4.54 - samples/sec: 1544.62 - lr: 0.000012 - momentum: 0.000000
2023-10-23 19:31:35,152 epoch 7 - iter 96/242 - loss 0.03093927 - time (sec): 6.07 - samples/sec: 1567.90 - lr: 0.000012 - momentum: 0.000000
2023-10-23 19:31:36,645 epoch 7 - iter 120/242 - loss 0.02876334 - time (sec): 7.56 - samples/sec: 1534.47 - lr: 0.000012 - momentum: 0.000000
2023-10-23 19:31:38,197 epoch 7 - iter 144/242 - loss 0.02510561 - time (sec): 9.11 - samples/sec: 1585.38 - lr: 0.000011 - momentum: 0.000000
2023-10-23 19:31:39,738 epoch 7 - iter 168/242 - loss 0.02481182 - time (sec): 10.65 - samples/sec: 1606.81 - lr: 0.000011 - momentum: 0.000000
2023-10-23 19:31:41,220 epoch 7 - iter 192/242 - loss 0.02407380 - time (sec): 12.13 - samples/sec: 1599.71 - lr: 0.000011 - momentum: 0.000000
2023-10-23 19:31:42,729 epoch 7 - iter 216/242 - loss 0.02445811 - time (sec): 13.64 - samples/sec: 1601.31 - lr: 0.000010 - momentum: 0.000000
2023-10-23 19:31:44,297 epoch 7 - iter 240/242 - loss 0.02432614 - time (sec): 15.21 - samples/sec: 1611.23 - lr: 0.000010 - momentum: 0.000000
2023-10-23 19:31:44,421 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:44,422 EPOCH 7 done: loss 0.0241 - lr: 0.000010
2023-10-23 19:31:45,119 DEV : loss 0.17772217094898224 - f1-score (micro avg) 0.86
2023-10-23 19:31:45,122 ----------------------------------------------------------------------------------------------------
2023-10-23 19:31:46,627 epoch 8 - iter 24/242 - loss 0.03288147 - time (sec): 1.50 - samples/sec: 1679.19 - lr: 0.000010 - momentum: 0.000000
2023-10-23 19:31:48,138 epoch 8 - iter 48/242 - loss 0.01853878 - time (sec): 3.01 - samples/sec: 1642.54 - lr: 0.000009 - momentum: 0.000000
2023-10-23 19:31:49,644 epoch 8 - iter 72/242 - loss 0.01634984 - time (sec): 4.52 - samples/sec: 1632.94 - lr: 0.000009 - momentum: 0.000000
2023-10-23 19:31:51,166 epoch 8 - iter 96/242 - loss 0.01817815 - time (sec): 6.04 - samples/sec: 1617.32 - lr: 0.000009 - momentum: 0.000000
2023-10-23 19:31:52,747 epoch 8 - iter 120/242 - loss 0.02189944 - time (sec): 7.62 - samples/sec: 1621.72 - lr: 0.000008 - momentum: 0.000000
2023-10-23 19:31:54,346 epoch 8 - iter 144/242 - loss 0.01859034 - time (sec): 9.22 - samples/sec: 1630.24 - lr: 0.000008 - momentum: 0.000000
2023-10-23 19:31:55,804 epoch 8 - iter 168/242 - loss 0.02102774 - time (sec): 10.68 - samples/sec: 1616.58 - lr: 0.000008 - momentum: 0.000000
2023-10-23 19:31:57,342 epoch 8 - iter 192/242 - loss 0.01985651 - time (sec): 12.22 - samples/sec: 1609.44 - lr: 0.000007 - momentum: 0.000000
2023-10-23 19:31:58,848 epoch 8 - iter 216/242 - loss 0.01865202 - time (sec): 13.73 - samples/sec: 1609.88 - lr: 0.000007 - momentum: 0.000000
2023-10-23 19:32:00,386 epoch 8 - iter 240/242 - loss 0.01800372 - time (sec): 15.26 - samples/sec: 1606.30 - lr: 0.000007 - momentum: 0.000000
2023-10-23 19:32:00,506 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:00,507 EPOCH 8 done: loss 0.0178 - lr: 0.000007
2023-10-23 19:32:01,209 DEV : loss 0.194981187582016 - f1-score (micro avg) 0.8373
2023-10-23 19:32:01,213 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:02,754 epoch 9 - iter 24/242 - loss 0.02859587 - time (sec): 1.54 - samples/sec: 1628.93 - lr: 0.000006 - momentum: 0.000000
2023-10-23 19:32:04,317 epoch 9 - iter 48/242 - loss 0.02766799 - time (sec): 3.10 - samples/sec: 1579.06 - lr: 0.000006 - momentum: 0.000000
2023-10-23 19:32:05,850 epoch 9 - iter 72/242 - loss 0.02277408 - time (sec): 4.64 - samples/sec: 1589.03 - lr: 0.000006 - momentum: 0.000000
2023-10-23 19:32:07,389 epoch 9 - iter 96/242 - loss 0.01930538 - time (sec): 6.18 - samples/sec: 1617.33 - lr: 0.000005 - momentum: 0.000000
2023-10-23 19:32:08,917 epoch 9 - iter 120/242 - loss 0.01681611 - time (sec): 7.70 - samples/sec: 1623.29 - lr: 0.000005 - momentum: 0.000000
2023-10-23 19:32:10,404 epoch 9 - iter 144/242 - loss 0.01500288 - time (sec): 9.19 - samples/sec: 1618.61 - lr: 0.000005 - momentum: 0.000000
2023-10-23 19:32:11,935 epoch 9 - iter 168/242 - loss 0.01351396 - time (sec): 10.72 - samples/sec: 1600.59 - lr: 0.000004 - momentum: 0.000000
2023-10-23 19:32:13,428 epoch 9 - iter 192/242 - loss 0.01220532 - time (sec): 12.21 - samples/sec: 1592.90 - lr: 0.000004 - momentum: 0.000000
2023-10-23 19:32:14,944 epoch 9 - iter 216/242 - loss 0.01149296 - time (sec): 13.73 - samples/sec: 1606.89 - lr: 0.000004 - momentum: 0.000000
2023-10-23 19:32:16,481 epoch 9 - iter 240/242 - loss 0.01091838 - time (sec): 15.27 - samples/sec: 1613.65 - lr: 0.000003 - momentum: 0.000000
2023-10-23 19:32:16,595 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:16,596 EPOCH 9 done: loss 0.0109 - lr: 0.000003
2023-10-23 19:32:17,294 DEV : loss 0.1826126128435135 - f1-score (micro avg) 0.8607
2023-10-23 19:32:17,298 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:18,855 epoch 10 - iter 24/242 - loss 0.01137724 - time (sec): 1.56 - samples/sec: 1703.46 - lr: 0.000003 - momentum: 0.000000
2023-10-23 19:32:20,347 epoch 10 - iter 48/242 - loss 0.00797534 - time (sec): 3.05 - samples/sec: 1654.01 - lr: 0.000003 - momentum: 0.000000
2023-10-23 19:32:21,927 epoch 10 - iter 72/242 - loss 0.00713636 - time (sec): 4.63 - samples/sec: 1662.10 - lr: 0.000002 - momentum: 0.000000
2023-10-23 19:32:23,402 epoch 10 - iter 96/242 - loss 0.00601590 - time (sec): 6.10 - samples/sec: 1594.00 - lr: 0.000002 - momentum: 0.000000
2023-10-23 19:32:24,907 epoch 10 - iter 120/242 - loss 0.00677766 - time (sec): 7.61 - samples/sec: 1612.58 - lr: 0.000002 - momentum: 0.000000
2023-10-23 19:32:26,407 epoch 10 - iter 144/242 - loss 0.00631534 - time (sec): 9.11 - samples/sec: 1625.46 - lr: 0.000001 - momentum: 0.000000
2023-10-23 19:32:27,926 epoch 10 - iter 168/242 - loss 0.00589143 - time (sec): 10.63 - samples/sec: 1622.92 - lr: 0.000001 - momentum: 0.000000
2023-10-23 19:32:29,420 epoch 10 - iter 192/242 - loss 0.00766039 - time (sec): 12.12 - samples/sec: 1617.90 - lr: 0.000001 - momentum: 0.000000
2023-10-23 19:32:31,004 epoch 10 - iter 216/242 - loss 0.00876882 - time (sec): 13.71 - samples/sec: 1621.94 - lr: 0.000000 - momentum: 0.000000
2023-10-23 19:32:32,544 epoch 10 - iter 240/242 - loss 0.00939067 - time (sec): 15.25 - samples/sec: 1610.64 - lr: 0.000000 - momentum: 0.000000
2023-10-23 19:32:32,664 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:32,665 EPOCH 10 done: loss 0.0093 - lr: 0.000000
2023-10-23 19:32:33,365 DEV : loss 0.18982915580272675 - f1-score (micro avg) 0.8546
2023-10-23 19:32:33,839 ----------------------------------------------------------------------------------------------------
2023-10-23 19:32:33,840 Loading model from best epoch ...
2023-10-23 19:32:35,389 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-23 19:32:36,253
Results:
- F-score (micro) 0.8154
- F-score (macro) 0.5535
- Accuracy 0.7064
By class:
precision recall f1-score support
pers 0.8777 0.8777 0.8777 139
scope 0.8421 0.8682 0.8550 129
work 0.6593 0.7500 0.7018 80
loc 0.6667 0.2222 0.3333 9
date 0.0000 0.0000 0.0000 3
micro avg 0.8087 0.8222 0.8154 360
macro avg 0.6092 0.5436 0.5535 360
weighted avg 0.8038 0.8222 0.8095 360
2023-10-23 19:32:36,253 ----------------------------------------------------------------------------------------------------
|