File size: 37,001 Bytes
02b3f9f |
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 |
2023-10-23 23:08:41,805 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 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 23:08:41,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 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 23:08:41,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 Train: 3575 sentences
2023-10-23 23:08:41,806 (train_with_dev=False, train_with_test=False)
2023-10-23 23:08:41,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 Training Params:
2023-10-23 23:08:41,806 - learning_rate: "5e-05"
2023-10-23 23:08:41,806 - mini_batch_size: "8"
2023-10-23 23:08:41,806 - max_epochs: "10"
2023-10-23 23:08:41,806 - shuffle: "True"
2023-10-23 23:08:41,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 Plugins:
2023-10-23 23:08:41,806 - TensorboardLogger
2023-10-23 23:08:41,806 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 23:08:41,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,806 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 23:08:41,806 - metric: "('micro avg', 'f1-score')"
2023-10-23 23:08:41,807 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,807 Computation:
2023-10-23 23:08:41,807 - compute on device: cuda:0
2023-10-23 23:08:41,807 - embedding storage: none
2023-10-23 23:08:41,807 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,807 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-23 23:08:41,807 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,807 ----------------------------------------------------------------------------------------------------
2023-10-23 23:08:41,807 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 23:08:45,910 epoch 1 - iter 44/447 - loss 2.27236626 - time (sec): 4.10 - samples/sec: 2190.58 - lr: 0.000005 - momentum: 0.000000
2023-10-23 23:08:49,783 epoch 1 - iter 88/447 - loss 1.47589070 - time (sec): 7.98 - samples/sec: 2169.45 - lr: 0.000010 - momentum: 0.000000
2023-10-23 23:08:53,739 epoch 1 - iter 132/447 - loss 1.14504662 - time (sec): 11.93 - samples/sec: 2183.40 - lr: 0.000015 - momentum: 0.000000
2023-10-23 23:08:57,410 epoch 1 - iter 176/447 - loss 0.95999639 - time (sec): 15.60 - samples/sec: 2206.17 - lr: 0.000020 - momentum: 0.000000
2023-10-23 23:09:01,957 epoch 1 - iter 220/447 - loss 0.81523692 - time (sec): 20.15 - samples/sec: 2172.46 - lr: 0.000024 - momentum: 0.000000
2023-10-23 23:09:05,690 epoch 1 - iter 264/447 - loss 0.73096034 - time (sec): 23.88 - samples/sec: 2166.16 - lr: 0.000029 - momentum: 0.000000
2023-10-23 23:09:09,636 epoch 1 - iter 308/447 - loss 0.66396424 - time (sec): 27.83 - samples/sec: 2151.96 - lr: 0.000034 - momentum: 0.000000
2023-10-23 23:09:13,363 epoch 1 - iter 352/447 - loss 0.61144599 - time (sec): 31.56 - samples/sec: 2134.84 - lr: 0.000039 - momentum: 0.000000
2023-10-23 23:09:17,484 epoch 1 - iter 396/447 - loss 0.56581415 - time (sec): 35.68 - samples/sec: 2144.61 - lr: 0.000044 - momentum: 0.000000
2023-10-23 23:09:21,621 epoch 1 - iter 440/447 - loss 0.52583326 - time (sec): 39.81 - samples/sec: 2143.06 - lr: 0.000049 - momentum: 0.000000
2023-10-23 23:09:22,210 ----------------------------------------------------------------------------------------------------
2023-10-23 23:09:22,211 EPOCH 1 done: loss 0.5212 - lr: 0.000049
2023-10-23 23:09:27,043 DEV : loss 0.1860317587852478 - f1-score (micro avg) 0.6471
2023-10-23 23:09:27,063 saving best model
2023-10-23 23:09:27,625 ----------------------------------------------------------------------------------------------------
2023-10-23 23:09:31,827 epoch 2 - iter 44/447 - loss 0.14761269 - time (sec): 4.20 - samples/sec: 2021.85 - lr: 0.000049 - momentum: 0.000000
2023-10-23 23:09:35,846 epoch 2 - iter 88/447 - loss 0.15284626 - time (sec): 8.22 - samples/sec: 2103.13 - lr: 0.000049 - momentum: 0.000000
2023-10-23 23:09:39,885 epoch 2 - iter 132/447 - loss 0.14429856 - time (sec): 12.26 - samples/sec: 2086.96 - lr: 0.000048 - momentum: 0.000000
2023-10-23 23:09:44,029 epoch 2 - iter 176/447 - loss 0.14832291 - time (sec): 16.40 - samples/sec: 2107.76 - lr: 0.000048 - momentum: 0.000000
2023-10-23 23:09:47,814 epoch 2 - iter 220/447 - loss 0.14470472 - time (sec): 20.19 - samples/sec: 2116.84 - lr: 0.000047 - momentum: 0.000000
2023-10-23 23:09:51,809 epoch 2 - iter 264/447 - loss 0.14200948 - time (sec): 24.18 - samples/sec: 2113.54 - lr: 0.000047 - momentum: 0.000000
2023-10-23 23:09:55,650 epoch 2 - iter 308/447 - loss 0.14010020 - time (sec): 28.02 - samples/sec: 2126.77 - lr: 0.000046 - momentum: 0.000000
2023-10-23 23:09:59,664 epoch 2 - iter 352/447 - loss 0.13626107 - time (sec): 32.04 - samples/sec: 2135.28 - lr: 0.000046 - momentum: 0.000000
2023-10-23 23:10:03,602 epoch 2 - iter 396/447 - loss 0.13432277 - time (sec): 35.98 - samples/sec: 2128.26 - lr: 0.000045 - momentum: 0.000000
2023-10-23 23:10:07,511 epoch 2 - iter 440/447 - loss 0.13033964 - time (sec): 39.89 - samples/sec: 2137.16 - lr: 0.000045 - momentum: 0.000000
2023-10-23 23:10:08,131 ----------------------------------------------------------------------------------------------------
2023-10-23 23:10:08,131 EPOCH 2 done: loss 0.1318 - lr: 0.000045
2023-10-23 23:10:14,620 DEV : loss 0.1406162679195404 - f1-score (micro avg) 0.6881
2023-10-23 23:10:14,640 saving best model
2023-10-23 23:10:15,340 ----------------------------------------------------------------------------------------------------
2023-10-23 23:10:19,694 epoch 3 - iter 44/447 - loss 0.08512096 - time (sec): 4.35 - samples/sec: 2195.81 - lr: 0.000044 - momentum: 0.000000
2023-10-23 23:10:23,624 epoch 3 - iter 88/447 - loss 0.07659362 - time (sec): 8.28 - samples/sec: 2246.50 - lr: 0.000043 - momentum: 0.000000
2023-10-23 23:10:27,642 epoch 3 - iter 132/447 - loss 0.08033167 - time (sec): 12.30 - samples/sec: 2212.91 - lr: 0.000043 - momentum: 0.000000
2023-10-23 23:10:31,335 epoch 3 - iter 176/447 - loss 0.07519298 - time (sec): 15.99 - samples/sec: 2203.20 - lr: 0.000042 - momentum: 0.000000
2023-10-23 23:10:35,695 epoch 3 - iter 220/447 - loss 0.07314322 - time (sec): 20.35 - samples/sec: 2186.49 - lr: 0.000042 - momentum: 0.000000
2023-10-23 23:10:39,980 epoch 3 - iter 264/447 - loss 0.07593556 - time (sec): 24.64 - samples/sec: 2176.38 - lr: 0.000041 - momentum: 0.000000
2023-10-23 23:10:43,901 epoch 3 - iter 308/447 - loss 0.07629546 - time (sec): 28.56 - samples/sec: 2161.66 - lr: 0.000041 - momentum: 0.000000
2023-10-23 23:10:47,582 epoch 3 - iter 352/447 - loss 0.07758371 - time (sec): 32.24 - samples/sec: 2141.69 - lr: 0.000040 - momentum: 0.000000
2023-10-23 23:10:51,499 epoch 3 - iter 396/447 - loss 0.07667548 - time (sec): 36.16 - samples/sec: 2152.05 - lr: 0.000040 - momentum: 0.000000
2023-10-23 23:10:55,266 epoch 3 - iter 440/447 - loss 0.07715178 - time (sec): 39.93 - samples/sec: 2140.07 - lr: 0.000039 - momentum: 0.000000
2023-10-23 23:10:55,806 ----------------------------------------------------------------------------------------------------
2023-10-23 23:10:55,806 EPOCH 3 done: loss 0.0770 - lr: 0.000039
2023-10-23 23:11:02,275 DEV : loss 0.15030179917812347 - f1-score (micro avg) 0.7238
2023-10-23 23:11:02,295 saving best model
2023-10-23 23:11:02,975 ----------------------------------------------------------------------------------------------------
2023-10-23 23:11:06,783 epoch 4 - iter 44/447 - loss 0.04797574 - time (sec): 3.81 - samples/sec: 2181.58 - lr: 0.000038 - momentum: 0.000000
2023-10-23 23:11:11,093 epoch 4 - iter 88/447 - loss 0.04426022 - time (sec): 8.12 - samples/sec: 2169.33 - lr: 0.000038 - momentum: 0.000000
2023-10-23 23:11:15,295 epoch 4 - iter 132/447 - loss 0.04936236 - time (sec): 12.32 - samples/sec: 2146.96 - lr: 0.000037 - momentum: 0.000000
2023-10-23 23:11:19,019 epoch 4 - iter 176/447 - loss 0.04772664 - time (sec): 16.04 - samples/sec: 2124.81 - lr: 0.000037 - momentum: 0.000000
2023-10-23 23:11:23,603 epoch 4 - iter 220/447 - loss 0.05048456 - time (sec): 20.63 - samples/sec: 2113.87 - lr: 0.000036 - momentum: 0.000000
2023-10-23 23:11:27,386 epoch 4 - iter 264/447 - loss 0.04914078 - time (sec): 24.41 - samples/sec: 2109.21 - lr: 0.000036 - momentum: 0.000000
2023-10-23 23:11:31,436 epoch 4 - iter 308/447 - loss 0.05102896 - time (sec): 28.46 - samples/sec: 2130.52 - lr: 0.000035 - momentum: 0.000000
2023-10-23 23:11:35,388 epoch 4 - iter 352/447 - loss 0.04983912 - time (sec): 32.41 - samples/sec: 2128.90 - lr: 0.000035 - momentum: 0.000000
2023-10-23 23:11:39,226 epoch 4 - iter 396/447 - loss 0.04862449 - time (sec): 36.25 - samples/sec: 2128.71 - lr: 0.000034 - momentum: 0.000000
2023-10-23 23:11:43,063 epoch 4 - iter 440/447 - loss 0.04914043 - time (sec): 40.09 - samples/sec: 2126.14 - lr: 0.000033 - momentum: 0.000000
2023-10-23 23:11:43,676 ----------------------------------------------------------------------------------------------------
2023-10-23 23:11:43,676 EPOCH 4 done: loss 0.0491 - lr: 0.000033
2023-10-23 23:11:50,165 DEV : loss 0.15656068921089172 - f1-score (micro avg) 0.7269
2023-10-23 23:11:50,185 saving best model
2023-10-23 23:11:50,988 ----------------------------------------------------------------------------------------------------
2023-10-23 23:11:54,902 epoch 5 - iter 44/447 - loss 0.03002117 - time (sec): 3.91 - samples/sec: 2187.36 - lr: 0.000033 - momentum: 0.000000
2023-10-23 23:11:59,212 epoch 5 - iter 88/447 - loss 0.02728072 - time (sec): 8.22 - samples/sec: 2120.86 - lr: 0.000032 - momentum: 0.000000
2023-10-23 23:12:03,032 epoch 5 - iter 132/447 - loss 0.02895759 - time (sec): 12.04 - samples/sec: 2136.02 - lr: 0.000032 - momentum: 0.000000
2023-10-23 23:12:06,776 epoch 5 - iter 176/447 - loss 0.03074911 - time (sec): 15.79 - samples/sec: 2136.86 - lr: 0.000031 - momentum: 0.000000
2023-10-23 23:12:10,831 epoch 5 - iter 220/447 - loss 0.03293494 - time (sec): 19.84 - samples/sec: 2125.55 - lr: 0.000031 - momentum: 0.000000
2023-10-23 23:12:14,878 epoch 5 - iter 264/447 - loss 0.03246686 - time (sec): 23.89 - samples/sec: 2117.56 - lr: 0.000030 - momentum: 0.000000
2023-10-23 23:12:18,517 epoch 5 - iter 308/447 - loss 0.03300365 - time (sec): 27.53 - samples/sec: 2127.25 - lr: 0.000030 - momentum: 0.000000
2023-10-23 23:12:22,916 epoch 5 - iter 352/447 - loss 0.03468769 - time (sec): 31.93 - samples/sec: 2128.62 - lr: 0.000029 - momentum: 0.000000
2023-10-23 23:12:26,745 epoch 5 - iter 396/447 - loss 0.03430249 - time (sec): 35.76 - samples/sec: 2140.22 - lr: 0.000028 - momentum: 0.000000
2023-10-23 23:12:30,698 epoch 5 - iter 440/447 - loss 0.03391188 - time (sec): 39.71 - samples/sec: 2143.57 - lr: 0.000028 - momentum: 0.000000
2023-10-23 23:12:31,298 ----------------------------------------------------------------------------------------------------
2023-10-23 23:12:31,298 EPOCH 5 done: loss 0.0339 - lr: 0.000028
2023-10-23 23:12:37,775 DEV : loss 0.20451626181602478 - f1-score (micro avg) 0.7428
2023-10-23 23:12:37,795 saving best model
2023-10-23 23:12:38,499 ----------------------------------------------------------------------------------------------------
2023-10-23 23:12:42,599 epoch 6 - iter 44/447 - loss 0.02170057 - time (sec): 4.10 - samples/sec: 2012.29 - lr: 0.000027 - momentum: 0.000000
2023-10-23 23:12:46,760 epoch 6 - iter 88/447 - loss 0.02124938 - time (sec): 8.26 - samples/sec: 2053.64 - lr: 0.000027 - momentum: 0.000000
2023-10-23 23:12:51,324 epoch 6 - iter 132/447 - loss 0.01971043 - time (sec): 12.82 - samples/sec: 2073.96 - lr: 0.000026 - momentum: 0.000000
2023-10-23 23:12:55,094 epoch 6 - iter 176/447 - loss 0.02017089 - time (sec): 16.59 - samples/sec: 2093.75 - lr: 0.000026 - momentum: 0.000000
2023-10-23 23:12:58,991 epoch 6 - iter 220/447 - loss 0.02245611 - time (sec): 20.49 - samples/sec: 2107.08 - lr: 0.000025 - momentum: 0.000000
2023-10-23 23:13:02,855 epoch 6 - iter 264/447 - loss 0.02192478 - time (sec): 24.36 - samples/sec: 2114.08 - lr: 0.000025 - momentum: 0.000000
2023-10-23 23:13:06,613 epoch 6 - iter 308/447 - loss 0.02597977 - time (sec): 28.11 - samples/sec: 2110.28 - lr: 0.000024 - momentum: 0.000000
2023-10-23 23:13:10,309 epoch 6 - iter 352/447 - loss 0.02523379 - time (sec): 31.81 - samples/sec: 2110.92 - lr: 0.000023 - momentum: 0.000000
2023-10-23 23:13:14,459 epoch 6 - iter 396/447 - loss 0.02528112 - time (sec): 35.96 - samples/sec: 2117.03 - lr: 0.000023 - momentum: 0.000000
2023-10-23 23:13:18,376 epoch 6 - iter 440/447 - loss 0.02605289 - time (sec): 39.88 - samples/sec: 2136.22 - lr: 0.000022 - momentum: 0.000000
2023-10-23 23:13:19,036 ----------------------------------------------------------------------------------------------------
2023-10-23 23:13:19,037 EPOCH 6 done: loss 0.0258 - lr: 0.000022
2023-10-23 23:13:25,516 DEV : loss 0.2170478105545044 - f1-score (micro avg) 0.7614
2023-10-23 23:13:25,535 saving best model
2023-10-23 23:13:26,188 ----------------------------------------------------------------------------------------------------
2023-10-23 23:13:30,487 epoch 7 - iter 44/447 - loss 0.01538206 - time (sec): 4.30 - samples/sec: 2125.61 - lr: 0.000022 - momentum: 0.000000
2023-10-23 23:13:34,505 epoch 7 - iter 88/447 - loss 0.01267065 - time (sec): 8.32 - samples/sec: 2125.87 - lr: 0.000021 - momentum: 0.000000
2023-10-23 23:13:38,226 epoch 7 - iter 132/447 - loss 0.01358387 - time (sec): 12.04 - samples/sec: 2144.81 - lr: 0.000021 - momentum: 0.000000
2023-10-23 23:13:42,418 epoch 7 - iter 176/447 - loss 0.01343420 - time (sec): 16.23 - samples/sec: 2174.12 - lr: 0.000020 - momentum: 0.000000
2023-10-23 23:13:46,425 epoch 7 - iter 220/447 - loss 0.01768025 - time (sec): 20.24 - samples/sec: 2148.41 - lr: 0.000020 - momentum: 0.000000
2023-10-23 23:13:50,570 epoch 7 - iter 264/447 - loss 0.01641940 - time (sec): 24.38 - samples/sec: 2131.58 - lr: 0.000019 - momentum: 0.000000
2023-10-23 23:13:54,456 epoch 7 - iter 308/447 - loss 0.01606435 - time (sec): 28.27 - samples/sec: 2137.45 - lr: 0.000018 - momentum: 0.000000
2023-10-23 23:13:58,669 epoch 7 - iter 352/447 - loss 0.01613984 - time (sec): 32.48 - samples/sec: 2138.73 - lr: 0.000018 - momentum: 0.000000
2023-10-23 23:14:02,734 epoch 7 - iter 396/447 - loss 0.01570408 - time (sec): 36.55 - samples/sec: 2134.97 - lr: 0.000017 - momentum: 0.000000
2023-10-23 23:14:06,299 epoch 7 - iter 440/447 - loss 0.01613572 - time (sec): 40.11 - samples/sec: 2127.49 - lr: 0.000017 - momentum: 0.000000
2023-10-23 23:14:06,864 ----------------------------------------------------------------------------------------------------
2023-10-23 23:14:06,864 EPOCH 7 done: loss 0.0161 - lr: 0.000017
2023-10-23 23:14:13,333 DEV : loss 0.23187175393104553 - f1-score (micro avg) 0.7806
2023-10-23 23:14:13,353 saving best model
2023-10-23 23:14:14,060 ----------------------------------------------------------------------------------------------------
2023-10-23 23:14:17,973 epoch 8 - iter 44/447 - loss 0.01336597 - time (sec): 3.91 - samples/sec: 2170.87 - lr: 0.000016 - momentum: 0.000000
2023-10-23 23:14:22,346 epoch 8 - iter 88/447 - loss 0.01477991 - time (sec): 8.29 - samples/sec: 2121.51 - lr: 0.000016 - momentum: 0.000000
2023-10-23 23:14:26,105 epoch 8 - iter 132/447 - loss 0.01321960 - time (sec): 12.04 - samples/sec: 2137.93 - lr: 0.000015 - momentum: 0.000000
2023-10-23 23:14:30,061 epoch 8 - iter 176/447 - loss 0.01275806 - time (sec): 16.00 - samples/sec: 2113.90 - lr: 0.000015 - momentum: 0.000000
2023-10-23 23:14:33,963 epoch 8 - iter 220/447 - loss 0.01276995 - time (sec): 19.90 - samples/sec: 2119.87 - lr: 0.000014 - momentum: 0.000000
2023-10-23 23:14:37,589 epoch 8 - iter 264/447 - loss 0.01257530 - time (sec): 23.53 - samples/sec: 2133.43 - lr: 0.000013 - momentum: 0.000000
2023-10-23 23:14:41,523 epoch 8 - iter 308/447 - loss 0.01188743 - time (sec): 27.46 - samples/sec: 2139.33 - lr: 0.000013 - momentum: 0.000000
2023-10-23 23:14:46,168 epoch 8 - iter 352/447 - loss 0.01152702 - time (sec): 32.11 - samples/sec: 2127.33 - lr: 0.000012 - momentum: 0.000000
2023-10-23 23:14:50,037 epoch 8 - iter 396/447 - loss 0.01253502 - time (sec): 35.98 - samples/sec: 2145.76 - lr: 0.000012 - momentum: 0.000000
2023-10-23 23:14:53,846 epoch 8 - iter 440/447 - loss 0.01242547 - time (sec): 39.79 - samples/sec: 2145.24 - lr: 0.000011 - momentum: 0.000000
2023-10-23 23:14:54,454 ----------------------------------------------------------------------------------------------------
2023-10-23 23:14:54,455 EPOCH 8 done: loss 0.0126 - lr: 0.000011
2023-10-23 23:15:00,672 DEV : loss 0.25415274500846863 - f1-score (micro avg) 0.7639
2023-10-23 23:15:00,692 ----------------------------------------------------------------------------------------------------
2023-10-23 23:15:04,596 epoch 9 - iter 44/447 - loss 0.00714680 - time (sec): 3.90 - samples/sec: 2135.73 - lr: 0.000011 - momentum: 0.000000
2023-10-23 23:15:08,853 epoch 9 - iter 88/447 - loss 0.00594912 - time (sec): 8.16 - samples/sec: 2108.50 - lr: 0.000010 - momentum: 0.000000
2023-10-23 23:15:13,106 epoch 9 - iter 132/447 - loss 0.00597966 - time (sec): 12.41 - samples/sec: 2114.70 - lr: 0.000010 - momentum: 0.000000
2023-10-23 23:15:16,844 epoch 9 - iter 176/447 - loss 0.00580727 - time (sec): 16.15 - samples/sec: 2123.16 - lr: 0.000009 - momentum: 0.000000
2023-10-23 23:15:20,658 epoch 9 - iter 220/447 - loss 0.00487805 - time (sec): 19.97 - samples/sec: 2143.39 - lr: 0.000008 - momentum: 0.000000
2023-10-23 23:15:24,215 epoch 9 - iter 264/447 - loss 0.00589291 - time (sec): 23.52 - samples/sec: 2136.72 - lr: 0.000008 - momentum: 0.000000
2023-10-23 23:15:28,396 epoch 9 - iter 308/447 - loss 0.00563785 - time (sec): 27.70 - samples/sec: 2132.94 - lr: 0.000007 - momentum: 0.000000
2023-10-23 23:15:32,702 epoch 9 - iter 352/447 - loss 0.00592797 - time (sec): 32.01 - samples/sec: 2146.78 - lr: 0.000007 - momentum: 0.000000
2023-10-23 23:15:36,654 epoch 9 - iter 396/447 - loss 0.00641546 - time (sec): 35.96 - samples/sec: 2132.53 - lr: 0.000006 - momentum: 0.000000
2023-10-23 23:15:40,759 epoch 9 - iter 440/447 - loss 0.00638558 - time (sec): 40.07 - samples/sec: 2125.63 - lr: 0.000006 - momentum: 0.000000
2023-10-23 23:15:41,327 ----------------------------------------------------------------------------------------------------
2023-10-23 23:15:41,327 EPOCH 9 done: loss 0.0065 - lr: 0.000006
2023-10-23 23:15:47,534 DEV : loss 0.2673643231391907 - f1-score (micro avg) 0.7574
2023-10-23 23:15:47,554 ----------------------------------------------------------------------------------------------------
2023-10-23 23:15:51,278 epoch 10 - iter 44/447 - loss 0.00836690 - time (sec): 3.72 - samples/sec: 2141.97 - lr: 0.000005 - momentum: 0.000000
2023-10-23 23:15:55,016 epoch 10 - iter 88/447 - loss 0.00437091 - time (sec): 7.46 - samples/sec: 2131.00 - lr: 0.000005 - momentum: 0.000000
2023-10-23 23:15:59,096 epoch 10 - iter 132/447 - loss 0.00373131 - time (sec): 11.54 - samples/sec: 2151.94 - lr: 0.000004 - momentum: 0.000000
2023-10-23 23:16:03,686 epoch 10 - iter 176/447 - loss 0.00354035 - time (sec): 16.13 - samples/sec: 2099.39 - lr: 0.000003 - momentum: 0.000000
2023-10-23 23:16:07,914 epoch 10 - iter 220/447 - loss 0.00388026 - time (sec): 20.36 - samples/sec: 2107.69 - lr: 0.000003 - momentum: 0.000000
2023-10-23 23:16:11,759 epoch 10 - iter 264/447 - loss 0.00333544 - time (sec): 24.20 - samples/sec: 2114.67 - lr: 0.000002 - momentum: 0.000000
2023-10-23 23:16:15,479 epoch 10 - iter 308/447 - loss 0.00354977 - time (sec): 27.92 - samples/sec: 2121.71 - lr: 0.000002 - momentum: 0.000000
2023-10-23 23:16:19,304 epoch 10 - iter 352/447 - loss 0.00329614 - time (sec): 31.75 - samples/sec: 2122.26 - lr: 0.000001 - momentum: 0.000000
2023-10-23 23:16:23,342 epoch 10 - iter 396/447 - loss 0.00360615 - time (sec): 35.79 - samples/sec: 2130.55 - lr: 0.000001 - momentum: 0.000000
2023-10-23 23:16:27,685 epoch 10 - iter 440/447 - loss 0.00375059 - time (sec): 40.13 - samples/sec: 2118.59 - lr: 0.000000 - momentum: 0.000000
2023-10-23 23:16:28,315 ----------------------------------------------------------------------------------------------------
2023-10-23 23:16:28,315 EPOCH 10 done: loss 0.0037 - lr: 0.000000
2023-10-23 23:16:34,547 DEV : loss 0.257240891456604 - f1-score (micro avg) 0.7644
2023-10-23 23:16:35,123 ----------------------------------------------------------------------------------------------------
2023-10-23 23:16:35,124 Loading model from best epoch ...
2023-10-23 23:16:36,865 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 23:16:41,683
Results:
- F-score (micro) 0.746
- F-score (macro) 0.6591
- Accuracy 0.6121
By class:
precision recall f1-score support
loc 0.8388 0.8641 0.8512 596
pers 0.6378 0.7297 0.6807 333
org 0.5726 0.5076 0.5382 132
prod 0.6800 0.5152 0.5862 66
time 0.6458 0.6327 0.6392 49
micro avg 0.7355 0.7568 0.7460 1176
macro avg 0.6750 0.6498 0.6591 1176
weighted avg 0.7350 0.7568 0.7441 1176
2023-10-23 23:16:41,683 ----------------------------------------------------------------------------------------------------
|