File size: 37,046 Bytes
8065a26 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 |
2023-10-23 21:04:58,060 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-23 21:04:58,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
- NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /home/ubuntu/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
2023-10-23 21:04:58,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 Train: 3575 sentences
2023-10-23 21:04:58,061 (train_with_dev=False, train_with_test=False)
2023-10-23 21:04:58,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 Training Params:
2023-10-23 21:04:58,061 - learning_rate: "3e-05"
2023-10-23 21:04:58,061 - mini_batch_size: "8"
2023-10-23 21:04:58,061 - max_epochs: "10"
2023-10-23 21:04:58,061 - shuffle: "True"
2023-10-23 21:04:58,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 Plugins:
2023-10-23 21:04:58,061 - TensorboardLogger
2023-10-23 21:04:58,061 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 21:04:58,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,061 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 21:04:58,061 - metric: "('micro avg', 'f1-score')"
2023-10-23 21:04:58,062 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,062 Computation:
2023-10-23 21:04:58,062 - compute on device: cuda:0
2023-10-23 21:04:58,062 - embedding storage: none
2023-10-23 21:04:58,062 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,062 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-23 21:04:58,062 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,062 ----------------------------------------------------------------------------------------------------
2023-10-23 21:04:58,062 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 21:05:01,858 epoch 1 - iter 44/447 - loss 2.77674307 - time (sec): 3.80 - samples/sec: 2068.88 - lr: 0.000003 - momentum: 0.000000
2023-10-23 21:05:05,973 epoch 1 - iter 88/447 - loss 1.75956664 - time (sec): 7.91 - samples/sec: 2088.59 - lr: 0.000006 - momentum: 0.000000
2023-10-23 21:05:10,065 epoch 1 - iter 132/447 - loss 1.30302502 - time (sec): 12.00 - samples/sec: 2082.97 - lr: 0.000009 - momentum: 0.000000
2023-10-23 21:05:14,086 epoch 1 - iter 176/447 - loss 1.08268102 - time (sec): 16.02 - samples/sec: 2078.90 - lr: 0.000012 - momentum: 0.000000
2023-10-23 21:05:17,972 epoch 1 - iter 220/447 - loss 0.93560897 - time (sec): 19.91 - samples/sec: 2103.90 - lr: 0.000015 - momentum: 0.000000
2023-10-23 21:05:21,763 epoch 1 - iter 264/447 - loss 0.83394592 - time (sec): 23.70 - samples/sec: 2104.11 - lr: 0.000018 - momentum: 0.000000
2023-10-23 21:05:25,684 epoch 1 - iter 308/447 - loss 0.75226450 - time (sec): 27.62 - samples/sec: 2106.94 - lr: 0.000021 - momentum: 0.000000
2023-10-23 21:05:29,651 epoch 1 - iter 352/447 - loss 0.68133343 - time (sec): 31.59 - samples/sec: 2108.95 - lr: 0.000024 - momentum: 0.000000
2023-10-23 21:05:34,085 epoch 1 - iter 396/447 - loss 0.62684172 - time (sec): 36.02 - samples/sec: 2124.23 - lr: 0.000027 - momentum: 0.000000
2023-10-23 21:05:37,891 epoch 1 - iter 440/447 - loss 0.58441638 - time (sec): 39.83 - samples/sec: 2137.74 - lr: 0.000029 - momentum: 0.000000
2023-10-23 21:05:38,506 ----------------------------------------------------------------------------------------------------
2023-10-23 21:05:38,506 EPOCH 1 done: loss 0.5780 - lr: 0.000029
2023-10-23 21:05:43,315 DEV : loss 0.15914756059646606 - f1-score (micro avg) 0.5805
2023-10-23 21:05:43,335 saving best model
2023-10-23 21:05:43,804 ----------------------------------------------------------------------------------------------------
2023-10-23 21:05:47,532 epoch 2 - iter 44/447 - loss 0.17247700 - time (sec): 3.73 - samples/sec: 2206.20 - lr: 0.000030 - momentum: 0.000000
2023-10-23 21:05:51,556 epoch 2 - iter 88/447 - loss 0.15210658 - time (sec): 7.75 - samples/sec: 2170.56 - lr: 0.000029 - momentum: 0.000000
2023-10-23 21:05:55,638 epoch 2 - iter 132/447 - loss 0.14415904 - time (sec): 11.83 - samples/sec: 2166.91 - lr: 0.000029 - momentum: 0.000000
2023-10-23 21:05:59,766 epoch 2 - iter 176/447 - loss 0.14348377 - time (sec): 15.96 - samples/sec: 2153.80 - lr: 0.000029 - momentum: 0.000000
2023-10-23 21:06:03,542 epoch 2 - iter 220/447 - loss 0.13706169 - time (sec): 19.74 - samples/sec: 2132.36 - lr: 0.000028 - momentum: 0.000000
2023-10-23 21:06:07,684 epoch 2 - iter 264/447 - loss 0.13827372 - time (sec): 23.88 - samples/sec: 2135.49 - lr: 0.000028 - momentum: 0.000000
2023-10-23 21:06:11,700 epoch 2 - iter 308/447 - loss 0.13580790 - time (sec): 27.90 - samples/sec: 2140.60 - lr: 0.000028 - momentum: 0.000000
2023-10-23 21:06:15,332 epoch 2 - iter 352/447 - loss 0.13583544 - time (sec): 31.53 - samples/sec: 2145.24 - lr: 0.000027 - momentum: 0.000000
2023-10-23 21:06:19,814 epoch 2 - iter 396/447 - loss 0.13680668 - time (sec): 36.01 - samples/sec: 2139.14 - lr: 0.000027 - momentum: 0.000000
2023-10-23 21:06:23,630 epoch 2 - iter 440/447 - loss 0.13356993 - time (sec): 39.83 - samples/sec: 2137.55 - lr: 0.000027 - momentum: 0.000000
2023-10-23 21:06:24,229 ----------------------------------------------------------------------------------------------------
2023-10-23 21:06:24,230 EPOCH 2 done: loss 0.1328 - lr: 0.000027
2023-10-23 21:06:30,708 DEV : loss 0.12941311299800873 - f1-score (micro avg) 0.7109
2023-10-23 21:06:30,728 saving best model
2023-10-23 21:06:31,322 ----------------------------------------------------------------------------------------------------
2023-10-23 21:06:35,398 epoch 3 - iter 44/447 - loss 0.05897943 - time (sec): 4.07 - samples/sec: 2144.85 - lr: 0.000026 - momentum: 0.000000
2023-10-23 21:06:39,480 epoch 3 - iter 88/447 - loss 0.07267667 - time (sec): 8.16 - samples/sec: 2139.98 - lr: 0.000026 - momentum: 0.000000
2023-10-23 21:06:43,614 epoch 3 - iter 132/447 - loss 0.07124643 - time (sec): 12.29 - samples/sec: 2163.33 - lr: 0.000026 - momentum: 0.000000
2023-10-23 21:06:47,536 epoch 3 - iter 176/447 - loss 0.06874515 - time (sec): 16.21 - samples/sec: 2126.51 - lr: 0.000025 - momentum: 0.000000
2023-10-23 21:06:51,418 epoch 3 - iter 220/447 - loss 0.06845317 - time (sec): 20.10 - samples/sec: 2144.00 - lr: 0.000025 - momentum: 0.000000
2023-10-23 21:06:55,177 epoch 3 - iter 264/447 - loss 0.06741457 - time (sec): 23.85 - samples/sec: 2150.15 - lr: 0.000025 - momentum: 0.000000
2023-10-23 21:06:59,046 epoch 3 - iter 308/447 - loss 0.06763183 - time (sec): 27.72 - samples/sec: 2142.44 - lr: 0.000024 - momentum: 0.000000
2023-10-23 21:07:03,214 epoch 3 - iter 352/447 - loss 0.06611837 - time (sec): 31.89 - samples/sec: 2146.71 - lr: 0.000024 - momentum: 0.000000
2023-10-23 21:07:07,031 epoch 3 - iter 396/447 - loss 0.06604370 - time (sec): 35.71 - samples/sec: 2149.25 - lr: 0.000024 - momentum: 0.000000
2023-10-23 21:07:11,156 epoch 3 - iter 440/447 - loss 0.06746680 - time (sec): 39.83 - samples/sec: 2134.13 - lr: 0.000023 - momentum: 0.000000
2023-10-23 21:07:11,816 ----------------------------------------------------------------------------------------------------
2023-10-23 21:07:11,816 EPOCH 3 done: loss 0.0677 - lr: 0.000023
2023-10-23 21:07:18,319 DEV : loss 0.13155309855937958 - f1-score (micro avg) 0.7518
2023-10-23 21:07:18,339 saving best model
2023-10-23 21:07:18,912 ----------------------------------------------------------------------------------------------------
2023-10-23 21:07:22,645 epoch 4 - iter 44/447 - loss 0.04956695 - time (sec): 3.73 - samples/sec: 2133.43 - lr: 0.000023 - momentum: 0.000000
2023-10-23 21:07:26,708 epoch 4 - iter 88/447 - loss 0.04093136 - time (sec): 7.79 - samples/sec: 2113.74 - lr: 0.000023 - momentum: 0.000000
2023-10-23 21:07:30,759 epoch 4 - iter 132/447 - loss 0.03912973 - time (sec): 11.85 - samples/sec: 2133.26 - lr: 0.000022 - momentum: 0.000000
2023-10-23 21:07:34,940 epoch 4 - iter 176/447 - loss 0.03954112 - time (sec): 16.03 - samples/sec: 2116.49 - lr: 0.000022 - momentum: 0.000000
2023-10-23 21:07:39,131 epoch 4 - iter 220/447 - loss 0.03912372 - time (sec): 20.22 - samples/sec: 2109.00 - lr: 0.000022 - momentum: 0.000000
2023-10-23 21:07:43,162 epoch 4 - iter 264/447 - loss 0.04029854 - time (sec): 24.25 - samples/sec: 2116.55 - lr: 0.000021 - momentum: 0.000000
2023-10-23 21:07:47,407 epoch 4 - iter 308/447 - loss 0.04017848 - time (sec): 28.49 - samples/sec: 2117.43 - lr: 0.000021 - momentum: 0.000000
2023-10-23 21:07:51,304 epoch 4 - iter 352/447 - loss 0.03995350 - time (sec): 32.39 - samples/sec: 2121.72 - lr: 0.000021 - momentum: 0.000000
2023-10-23 21:07:55,232 epoch 4 - iter 396/447 - loss 0.04119580 - time (sec): 36.32 - samples/sec: 2122.18 - lr: 0.000020 - momentum: 0.000000
2023-10-23 21:07:59,018 epoch 4 - iter 440/447 - loss 0.04224669 - time (sec): 40.10 - samples/sec: 2126.79 - lr: 0.000020 - momentum: 0.000000
2023-10-23 21:07:59,609 ----------------------------------------------------------------------------------------------------
2023-10-23 21:07:59,609 EPOCH 4 done: loss 0.0431 - lr: 0.000020
2023-10-23 21:08:06,099 DEV : loss 0.17578744888305664 - f1-score (micro avg) 0.764
2023-10-23 21:08:06,120 saving best model
2023-10-23 21:08:06,714 ----------------------------------------------------------------------------------------------------
2023-10-23 21:08:10,724 epoch 5 - iter 44/447 - loss 0.03173418 - time (sec): 4.01 - samples/sec: 2167.42 - lr: 0.000020 - momentum: 0.000000
2023-10-23 21:08:14,826 epoch 5 - iter 88/447 - loss 0.03285878 - time (sec): 8.11 - samples/sec: 2075.18 - lr: 0.000019 - momentum: 0.000000
2023-10-23 21:08:18,587 epoch 5 - iter 132/447 - loss 0.03247897 - time (sec): 11.87 - samples/sec: 2091.33 - lr: 0.000019 - momentum: 0.000000
2023-10-23 21:08:22,960 epoch 5 - iter 176/447 - loss 0.03016416 - time (sec): 16.24 - samples/sec: 2100.92 - lr: 0.000019 - momentum: 0.000000
2023-10-23 21:08:26,816 epoch 5 - iter 220/447 - loss 0.02778420 - time (sec): 20.10 - samples/sec: 2102.65 - lr: 0.000018 - momentum: 0.000000
2023-10-23 21:08:30,624 epoch 5 - iter 264/447 - loss 0.02814833 - time (sec): 23.91 - samples/sec: 2103.17 - lr: 0.000018 - momentum: 0.000000
2023-10-23 21:08:35,077 epoch 5 - iter 308/447 - loss 0.02586079 - time (sec): 28.36 - samples/sec: 2110.52 - lr: 0.000018 - momentum: 0.000000
2023-10-23 21:08:39,022 epoch 5 - iter 352/447 - loss 0.02544490 - time (sec): 32.31 - samples/sec: 2113.41 - lr: 0.000017 - momentum: 0.000000
2023-10-23 21:08:42,951 epoch 5 - iter 396/447 - loss 0.02670970 - time (sec): 36.24 - samples/sec: 2126.29 - lr: 0.000017 - momentum: 0.000000
2023-10-23 21:08:46,739 epoch 5 - iter 440/447 - loss 0.02604997 - time (sec): 40.02 - samples/sec: 2134.99 - lr: 0.000017 - momentum: 0.000000
2023-10-23 21:08:47,301 ----------------------------------------------------------------------------------------------------
2023-10-23 21:08:47,301 EPOCH 5 done: loss 0.0260 - lr: 0.000017
2023-10-23 21:08:53,795 DEV : loss 0.19835765659809113 - f1-score (micro avg) 0.7738
2023-10-23 21:08:53,815 saving best model
2023-10-23 21:08:54,418 ----------------------------------------------------------------------------------------------------
2023-10-23 21:08:58,311 epoch 6 - iter 44/447 - loss 0.02184566 - time (sec): 3.89 - samples/sec: 2032.43 - lr: 0.000016 - momentum: 0.000000
2023-10-23 21:09:02,255 epoch 6 - iter 88/447 - loss 0.02189035 - time (sec): 7.84 - samples/sec: 2042.74 - lr: 0.000016 - momentum: 0.000000
2023-10-23 21:09:06,400 epoch 6 - iter 132/447 - loss 0.01858513 - time (sec): 11.98 - samples/sec: 2069.63 - lr: 0.000016 - momentum: 0.000000
2023-10-23 21:09:10,451 epoch 6 - iter 176/447 - loss 0.01839336 - time (sec): 16.03 - samples/sec: 2120.22 - lr: 0.000015 - momentum: 0.000000
2023-10-23 21:09:14,439 epoch 6 - iter 220/447 - loss 0.01779606 - time (sec): 20.02 - samples/sec: 2132.02 - lr: 0.000015 - momentum: 0.000000
2023-10-23 21:09:18,507 epoch 6 - iter 264/447 - loss 0.01809152 - time (sec): 24.09 - samples/sec: 2112.60 - lr: 0.000015 - momentum: 0.000000
2023-10-23 21:09:22,335 epoch 6 - iter 308/447 - loss 0.01799876 - time (sec): 27.92 - samples/sec: 2123.56 - lr: 0.000014 - momentum: 0.000000
2023-10-23 21:09:26,257 epoch 6 - iter 352/447 - loss 0.01963981 - time (sec): 31.84 - samples/sec: 2130.81 - lr: 0.000014 - momentum: 0.000000
2023-10-23 21:09:30,544 epoch 6 - iter 396/447 - loss 0.01948104 - time (sec): 36.13 - samples/sec: 2122.38 - lr: 0.000014 - momentum: 0.000000
2023-10-23 21:09:34,423 epoch 6 - iter 440/447 - loss 0.01934598 - time (sec): 40.00 - samples/sec: 2135.33 - lr: 0.000013 - momentum: 0.000000
2023-10-23 21:09:34,989 ----------------------------------------------------------------------------------------------------
2023-10-23 21:09:34,990 EPOCH 6 done: loss 0.0195 - lr: 0.000013
2023-10-23 21:09:41,480 DEV : loss 0.2243068516254425 - f1-score (micro avg) 0.7653
2023-10-23 21:09:41,500 ----------------------------------------------------------------------------------------------------
2023-10-23 21:09:45,224 epoch 7 - iter 44/447 - loss 0.00920856 - time (sec): 3.72 - samples/sec: 2231.28 - lr: 0.000013 - momentum: 0.000000
2023-10-23 21:09:49,298 epoch 7 - iter 88/447 - loss 0.00662161 - time (sec): 7.80 - samples/sec: 2168.79 - lr: 0.000013 - momentum: 0.000000
2023-10-23 21:09:53,791 epoch 7 - iter 132/447 - loss 0.00770613 - time (sec): 12.29 - samples/sec: 2141.83 - lr: 0.000012 - momentum: 0.000000
2023-10-23 21:09:57,731 epoch 7 - iter 176/447 - loss 0.00876968 - time (sec): 16.23 - samples/sec: 2139.89 - lr: 0.000012 - momentum: 0.000000
2023-10-23 21:10:01,668 epoch 7 - iter 220/447 - loss 0.01105371 - time (sec): 20.17 - samples/sec: 2133.73 - lr: 0.000012 - momentum: 0.000000
2023-10-23 21:10:05,750 epoch 7 - iter 264/447 - loss 0.01232071 - time (sec): 24.25 - samples/sec: 2134.11 - lr: 0.000011 - momentum: 0.000000
2023-10-23 21:10:09,797 epoch 7 - iter 308/447 - loss 0.01229655 - time (sec): 28.30 - samples/sec: 2129.39 - lr: 0.000011 - momentum: 0.000000
2023-10-23 21:10:13,591 epoch 7 - iter 352/447 - loss 0.01195873 - time (sec): 32.09 - samples/sec: 2131.21 - lr: 0.000011 - momentum: 0.000000
2023-10-23 21:10:17,520 epoch 7 - iter 396/447 - loss 0.01218580 - time (sec): 36.02 - samples/sec: 2138.12 - lr: 0.000010 - momentum: 0.000000
2023-10-23 21:10:21,420 epoch 7 - iter 440/447 - loss 0.01243524 - time (sec): 39.92 - samples/sec: 2141.02 - lr: 0.000010 - momentum: 0.000000
2023-10-23 21:10:21,970 ----------------------------------------------------------------------------------------------------
2023-10-23 21:10:21,971 EPOCH 7 done: loss 0.0123 - lr: 0.000010
2023-10-23 21:10:28,450 DEV : loss 0.23942111432552338 - f1-score (micro avg) 0.7782
2023-10-23 21:10:28,471 saving best model
2023-10-23 21:10:29,061 ----------------------------------------------------------------------------------------------------
2023-10-23 21:10:32,915 epoch 8 - iter 44/447 - loss 0.01227698 - time (sec): 3.85 - samples/sec: 2174.91 - lr: 0.000010 - momentum: 0.000000
2023-10-23 21:10:36,826 epoch 8 - iter 88/447 - loss 0.01241903 - time (sec): 7.76 - samples/sec: 2170.56 - lr: 0.000009 - momentum: 0.000000
2023-10-23 21:10:40,673 epoch 8 - iter 132/447 - loss 0.01170822 - time (sec): 11.61 - samples/sec: 2125.15 - lr: 0.000009 - momentum: 0.000000
2023-10-23 21:10:45,301 epoch 8 - iter 176/447 - loss 0.00886420 - time (sec): 16.24 - samples/sec: 2140.47 - lr: 0.000009 - momentum: 0.000000
2023-10-23 21:10:49,255 epoch 8 - iter 220/447 - loss 0.00834151 - time (sec): 20.19 - samples/sec: 2146.67 - lr: 0.000008 - momentum: 0.000000
2023-10-23 21:10:52,900 epoch 8 - iter 264/447 - loss 0.00734419 - time (sec): 23.84 - samples/sec: 2125.14 - lr: 0.000008 - momentum: 0.000000
2023-10-23 21:10:57,066 epoch 8 - iter 308/447 - loss 0.00699004 - time (sec): 28.00 - samples/sec: 2125.54 - lr: 0.000008 - momentum: 0.000000
2023-10-23 21:11:01,039 epoch 8 - iter 352/447 - loss 0.00726347 - time (sec): 31.98 - samples/sec: 2127.42 - lr: 0.000007 - momentum: 0.000000
2023-10-23 21:11:05,458 epoch 8 - iter 396/447 - loss 0.00762904 - time (sec): 36.40 - samples/sec: 2122.16 - lr: 0.000007 - momentum: 0.000000
2023-10-23 21:11:09,201 epoch 8 - iter 440/447 - loss 0.00722230 - time (sec): 40.14 - samples/sec: 2121.45 - lr: 0.000007 - momentum: 0.000000
2023-10-23 21:11:09,820 ----------------------------------------------------------------------------------------------------
2023-10-23 21:11:09,820 EPOCH 8 done: loss 0.0072 - lr: 0.000007
2023-10-23 21:11:16,024 DEV : loss 0.24051210284233093 - f1-score (micro avg) 0.7845
2023-10-23 21:11:16,045 saving best model
2023-10-23 21:11:16,944 ----------------------------------------------------------------------------------------------------
2023-10-23 21:11:20,570 epoch 9 - iter 44/447 - loss 0.00666362 - time (sec): 3.62 - samples/sec: 2217.26 - lr: 0.000006 - momentum: 0.000000
2023-10-23 21:11:24,590 epoch 9 - iter 88/447 - loss 0.00577915 - time (sec): 7.65 - samples/sec: 2129.85 - lr: 0.000006 - momentum: 0.000000
2023-10-23 21:11:28,832 epoch 9 - iter 132/447 - loss 0.00485046 - time (sec): 11.89 - samples/sec: 2118.85 - lr: 0.000006 - momentum: 0.000000
2023-10-23 21:11:32,648 epoch 9 - iter 176/447 - loss 0.00491996 - time (sec): 15.70 - samples/sec: 2142.21 - lr: 0.000005 - momentum: 0.000000
2023-10-23 21:11:36,618 epoch 9 - iter 220/447 - loss 0.00513214 - time (sec): 19.67 - samples/sec: 2151.50 - lr: 0.000005 - momentum: 0.000000
2023-10-23 21:11:40,891 epoch 9 - iter 264/447 - loss 0.00479915 - time (sec): 23.95 - samples/sec: 2148.31 - lr: 0.000005 - momentum: 0.000000
2023-10-23 21:11:45,144 epoch 9 - iter 308/447 - loss 0.00461094 - time (sec): 28.20 - samples/sec: 2147.40 - lr: 0.000004 - momentum: 0.000000
2023-10-23 21:11:48,902 epoch 9 - iter 352/447 - loss 0.00508793 - time (sec): 31.96 - samples/sec: 2144.23 - lr: 0.000004 - momentum: 0.000000
2023-10-23 21:11:52,648 epoch 9 - iter 396/447 - loss 0.00553986 - time (sec): 35.70 - samples/sec: 2147.82 - lr: 0.000004 - momentum: 0.000000
2023-10-23 21:11:56,680 epoch 9 - iter 440/447 - loss 0.00517049 - time (sec): 39.74 - samples/sec: 2149.08 - lr: 0.000003 - momentum: 0.000000
2023-10-23 21:11:57,308 ----------------------------------------------------------------------------------------------------
2023-10-23 21:11:57,309 EPOCH 9 done: loss 0.0051 - lr: 0.000003
2023-10-23 21:12:03,519 DEV : loss 0.2497478574514389 - f1-score (micro avg) 0.7909
2023-10-23 21:12:03,540 saving best model
2023-10-23 21:12:04,111 ----------------------------------------------------------------------------------------------------
2023-10-23 21:12:08,002 epoch 10 - iter 44/447 - loss 0.00252055 - time (sec): 3.89 - samples/sec: 2207.00 - lr: 0.000003 - momentum: 0.000000
2023-10-23 21:12:12,159 epoch 10 - iter 88/447 - loss 0.00205105 - time (sec): 8.05 - samples/sec: 2163.44 - lr: 0.000003 - momentum: 0.000000
2023-10-23 21:12:16,370 epoch 10 - iter 132/447 - loss 0.00162416 - time (sec): 12.26 - samples/sec: 2100.99 - lr: 0.000002 - momentum: 0.000000
2023-10-23 21:12:20,040 epoch 10 - iter 176/447 - loss 0.00314123 - time (sec): 15.93 - samples/sec: 2135.69 - lr: 0.000002 - momentum: 0.000000
2023-10-23 21:12:24,341 epoch 10 - iter 220/447 - loss 0.00360926 - time (sec): 20.23 - samples/sec: 2144.50 - lr: 0.000002 - momentum: 0.000000
2023-10-23 21:12:28,106 epoch 10 - iter 264/447 - loss 0.00340675 - time (sec): 23.99 - samples/sec: 2135.76 - lr: 0.000001 - momentum: 0.000000
2023-10-23 21:12:31,919 epoch 10 - iter 308/447 - loss 0.00353423 - time (sec): 27.81 - samples/sec: 2147.17 - lr: 0.000001 - momentum: 0.000000
2023-10-23 21:12:36,013 epoch 10 - iter 352/447 - loss 0.00314415 - time (sec): 31.90 - samples/sec: 2139.65 - lr: 0.000001 - momentum: 0.000000
2023-10-23 21:12:40,324 epoch 10 - iter 396/447 - loss 0.00312445 - time (sec): 36.21 - samples/sec: 2122.16 - lr: 0.000000 - momentum: 0.000000
2023-10-23 21:12:44,382 epoch 10 - iter 440/447 - loss 0.00304078 - time (sec): 40.27 - samples/sec: 2116.62 - lr: 0.000000 - momentum: 0.000000
2023-10-23 21:12:45,004 ----------------------------------------------------------------------------------------------------
2023-10-23 21:12:45,004 EPOCH 10 done: loss 0.0030 - lr: 0.000000
2023-10-23 21:12:51,224 DEV : loss 0.25497499108314514 - f1-score (micro avg) 0.7901
2023-10-23 21:12:51,722 ----------------------------------------------------------------------------------------------------
2023-10-23 21:12:51,723 Loading model from best epoch ...
2023-10-23 21:12:53,466 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
2023-10-23 21:12:58,280
Results:
- F-score (micro) 0.7524
- F-score (macro) 0.665
- Accuracy 0.6214
By class:
precision recall f1-score support
loc 0.8280 0.8641 0.8456 596
pers 0.7064 0.7658 0.7349 333
org 0.4706 0.4848 0.4776 132
prod 0.6071 0.5152 0.5574 66
time 0.7500 0.6735 0.7097 49
micro avg 0.7391 0.7662 0.7524 1176
macro avg 0.6724 0.6607 0.6650 1176
weighted avg 0.7378 0.7662 0.7511 1176
2023-10-23 21:12:58,280 ----------------------------------------------------------------------------------------------------
|