Training in progress, step 2000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 966995080
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:703a6ac7121d730d9c0e42b1a06b45963892be5944000d1f0185ce7044bc3c9e
|
| 3 |
size 966995080
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1925064044
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddfbaef04fe7f922a8f6475913402834b772a6863710273bbd173e2839db0fd8
|
| 3 |
size 1925064044
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1877997138fd6c00a4ddcc0ec7e9c019b9f4ccb7a15031d5cedb28ccb7a2c96c
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29fb9e79fa30fbb431af919246a50a3118e2599b8f861d1f7ece53767b613869
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -296,6 +296,295 @@
|
|
| 296 |
"eval_samples_per_second": 2.257,
|
| 297 |
"eval_steps_per_second": 0.282,
|
| 298 |
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 299 |
}
|
| 300 |
],
|
| 301 |
"logging_steps": 25,
|
|
@@ -303,7 +592,7 @@
|
|
| 303 |
"num_input_tokens_seen": 0,
|
| 304 |
"num_train_epochs": 2,
|
| 305 |
"save_steps": 1000,
|
| 306 |
-
"total_flos":
|
| 307 |
"train_batch_size": 8,
|
| 308 |
"trial_name": null,
|
| 309 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 51.77491557370612,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-2000",
|
| 4 |
+
"epoch": 0.8,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 296 |
"eval_samples_per_second": 2.257,
|
| 297 |
"eval_steps_per_second": 0.282,
|
| 298 |
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.41,
|
| 302 |
+
"grad_norm": 11.293231964111328,
|
| 303 |
+
"learning_rate": 8.833333333333334e-06,
|
| 304 |
+
"loss": 1.153,
|
| 305 |
+
"step": 1025
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.42,
|
| 309 |
+
"grad_norm": 13.314165115356445,
|
| 310 |
+
"learning_rate": 8.777777777777778e-06,
|
| 311 |
+
"loss": 1.168,
|
| 312 |
+
"step": 1050
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.43,
|
| 316 |
+
"grad_norm": 13.231385231018066,
|
| 317 |
+
"learning_rate": 8.722222222222224e-06,
|
| 318 |
+
"loss": 1.1613,
|
| 319 |
+
"step": 1075
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.44,
|
| 323 |
+
"grad_norm": 13.21717643737793,
|
| 324 |
+
"learning_rate": 8.666666666666668e-06,
|
| 325 |
+
"loss": 1.1246,
|
| 326 |
+
"step": 1100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.45,
|
| 330 |
+
"grad_norm": 11.046935081481934,
|
| 331 |
+
"learning_rate": 8.611111111111112e-06,
|
| 332 |
+
"loss": 1.088,
|
| 333 |
+
"step": 1125
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.46,
|
| 337 |
+
"grad_norm": 14.906622886657715,
|
| 338 |
+
"learning_rate": 8.555555555555556e-06,
|
| 339 |
+
"loss": 1.19,
|
| 340 |
+
"step": 1150
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.47,
|
| 344 |
+
"grad_norm": 14.302517890930176,
|
| 345 |
+
"learning_rate": 8.5e-06,
|
| 346 |
+
"loss": 1.1351,
|
| 347 |
+
"step": 1175
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.48,
|
| 351 |
+
"grad_norm": 13.947770118713379,
|
| 352 |
+
"learning_rate": 8.444444444444446e-06,
|
| 353 |
+
"loss": 1.057,
|
| 354 |
+
"step": 1200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.49,
|
| 358 |
+
"grad_norm": 14.45609188079834,
|
| 359 |
+
"learning_rate": 8.38888888888889e-06,
|
| 360 |
+
"loss": 1.0993,
|
| 361 |
+
"step": 1225
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.5,
|
| 365 |
+
"grad_norm": 14.952827453613281,
|
| 366 |
+
"learning_rate": 8.333333333333334e-06,
|
| 367 |
+
"loss": 1.1626,
|
| 368 |
+
"step": 1250
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.51,
|
| 372 |
+
"grad_norm": 16.128353118896484,
|
| 373 |
+
"learning_rate": 8.277777777777778e-06,
|
| 374 |
+
"loss": 1.1082,
|
| 375 |
+
"step": 1275
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.52,
|
| 379 |
+
"grad_norm": 13.550396919250488,
|
| 380 |
+
"learning_rate": 8.222222222222222e-06,
|
| 381 |
+
"loss": 1.183,
|
| 382 |
+
"step": 1300
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.53,
|
| 386 |
+
"grad_norm": 14.400228500366211,
|
| 387 |
+
"learning_rate": 8.166666666666668e-06,
|
| 388 |
+
"loss": 1.1988,
|
| 389 |
+
"step": 1325
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.54,
|
| 393 |
+
"grad_norm": 13.9801607131958,
|
| 394 |
+
"learning_rate": 8.111111111111112e-06,
|
| 395 |
+
"loss": 1.1314,
|
| 396 |
+
"step": 1350
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.55,
|
| 400 |
+
"grad_norm": 12.84874439239502,
|
| 401 |
+
"learning_rate": 8.055555555555557e-06,
|
| 402 |
+
"loss": 1.1411,
|
| 403 |
+
"step": 1375
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.56,
|
| 407 |
+
"grad_norm": 14.126324653625488,
|
| 408 |
+
"learning_rate": 8.000000000000001e-06,
|
| 409 |
+
"loss": 1.1314,
|
| 410 |
+
"step": 1400
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.57,
|
| 414 |
+
"grad_norm": 12.402750015258789,
|
| 415 |
+
"learning_rate": 7.944444444444445e-06,
|
| 416 |
+
"loss": 1.1071,
|
| 417 |
+
"step": 1425
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.58,
|
| 421 |
+
"grad_norm": 13.835284233093262,
|
| 422 |
+
"learning_rate": 7.88888888888889e-06,
|
| 423 |
+
"loss": 1.1393,
|
| 424 |
+
"step": 1450
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.59,
|
| 428 |
+
"grad_norm": 12.414569854736328,
|
| 429 |
+
"learning_rate": 7.833333333333333e-06,
|
| 430 |
+
"loss": 1.1026,
|
| 431 |
+
"step": 1475
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.6,
|
| 435 |
+
"grad_norm": 15.43626880645752,
|
| 436 |
+
"learning_rate": 7.77777777777778e-06,
|
| 437 |
+
"loss": 1.164,
|
| 438 |
+
"step": 1500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.61,
|
| 442 |
+
"grad_norm": 13.067487716674805,
|
| 443 |
+
"learning_rate": 7.722222222222223e-06,
|
| 444 |
+
"loss": 1.0448,
|
| 445 |
+
"step": 1525
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.62,
|
| 449 |
+
"grad_norm": 14.158551216125488,
|
| 450 |
+
"learning_rate": 7.666666666666667e-06,
|
| 451 |
+
"loss": 1.1674,
|
| 452 |
+
"step": 1550
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.63,
|
| 456 |
+
"grad_norm": 13.062005996704102,
|
| 457 |
+
"learning_rate": 7.611111111111111e-06,
|
| 458 |
+
"loss": 1.0916,
|
| 459 |
+
"step": 1575
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.64,
|
| 463 |
+
"grad_norm": 13.6104736328125,
|
| 464 |
+
"learning_rate": 7.555555555555556e-06,
|
| 465 |
+
"loss": 1.0424,
|
| 466 |
+
"step": 1600
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.65,
|
| 470 |
+
"grad_norm": 11.52835750579834,
|
| 471 |
+
"learning_rate": 7.500000000000001e-06,
|
| 472 |
+
"loss": 1.0196,
|
| 473 |
+
"step": 1625
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.66,
|
| 477 |
+
"grad_norm": 14.118935585021973,
|
| 478 |
+
"learning_rate": 7.444444444444445e-06,
|
| 479 |
+
"loss": 1.1502,
|
| 480 |
+
"step": 1650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.67,
|
| 484 |
+
"grad_norm": 13.2473726272583,
|
| 485 |
+
"learning_rate": 7.38888888888889e-06,
|
| 486 |
+
"loss": 1.0562,
|
| 487 |
+
"step": 1675
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.68,
|
| 491 |
+
"grad_norm": 13.026944160461426,
|
| 492 |
+
"learning_rate": 7.333333333333333e-06,
|
| 493 |
+
"loss": 1.0391,
|
| 494 |
+
"step": 1700
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.69,
|
| 498 |
+
"grad_norm": 11.923539161682129,
|
| 499 |
+
"learning_rate": 7.277777777777778e-06,
|
| 500 |
+
"loss": 1.063,
|
| 501 |
+
"step": 1725
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.7,
|
| 505 |
+
"grad_norm": 10.581581115722656,
|
| 506 |
+
"learning_rate": 7.222222222222223e-06,
|
| 507 |
+
"loss": 1.0504,
|
| 508 |
+
"step": 1750
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.71,
|
| 512 |
+
"grad_norm": 13.761798858642578,
|
| 513 |
+
"learning_rate": 7.166666666666667e-06,
|
| 514 |
+
"loss": 1.1781,
|
| 515 |
+
"step": 1775
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.72,
|
| 519 |
+
"grad_norm": 13.440286636352539,
|
| 520 |
+
"learning_rate": 7.111111111111112e-06,
|
| 521 |
+
"loss": 1.088,
|
| 522 |
+
"step": 1800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.73,
|
| 526 |
+
"grad_norm": 11.378331184387207,
|
| 527 |
+
"learning_rate": 7.055555555555557e-06,
|
| 528 |
+
"loss": 1.017,
|
| 529 |
+
"step": 1825
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.74,
|
| 533 |
+
"grad_norm": 16.24916648864746,
|
| 534 |
+
"learning_rate": 7e-06,
|
| 535 |
+
"loss": 1.0669,
|
| 536 |
+
"step": 1850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.75,
|
| 540 |
+
"grad_norm": 14.499041557312012,
|
| 541 |
+
"learning_rate": 6.944444444444445e-06,
|
| 542 |
+
"loss": 1.023,
|
| 543 |
+
"step": 1875
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.76,
|
| 547 |
+
"grad_norm": 14.587787628173828,
|
| 548 |
+
"learning_rate": 6.88888888888889e-06,
|
| 549 |
+
"loss": 1.1128,
|
| 550 |
+
"step": 1900
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.77,
|
| 554 |
+
"grad_norm": 14.249890327453613,
|
| 555 |
+
"learning_rate": 6.833333333333334e-06,
|
| 556 |
+
"loss": 1.0462,
|
| 557 |
+
"step": 1925
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.78,
|
| 561 |
+
"grad_norm": 13.22544002532959,
|
| 562 |
+
"learning_rate": 6.777777777777779e-06,
|
| 563 |
+
"loss": 1.0564,
|
| 564 |
+
"step": 1950
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.79,
|
| 568 |
+
"grad_norm": 13.404162406921387,
|
| 569 |
+
"learning_rate": 6.7222222222222235e-06,
|
| 570 |
+
"loss": 1.0517,
|
| 571 |
+
"step": 1975
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.8,
|
| 575 |
+
"grad_norm": 13.87370491027832,
|
| 576 |
+
"learning_rate": 6.666666666666667e-06,
|
| 577 |
+
"loss": 1.0556,
|
| 578 |
+
"step": 2000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.8,
|
| 582 |
+
"eval_cer": 51.77491557370612,
|
| 583 |
+
"eval_loss": 1.0214924812316895,
|
| 584 |
+
"eval_runtime": 1738.5549,
|
| 585 |
+
"eval_samples_per_second": 2.264,
|
| 586 |
+
"eval_steps_per_second": 0.283,
|
| 587 |
+
"step": 2000
|
| 588 |
}
|
| 589 |
],
|
| 590 |
"logging_steps": 25,
|
|
|
|
| 592 |
"num_input_tokens_seen": 0,
|
| 593 |
"num_train_epochs": 2,
|
| 594 |
"save_steps": 1000,
|
| 595 |
+
"total_flos": 9.23473281024e+18,
|
| 596 |
"train_batch_size": 8,
|
| 597 |
"trial_name": null,
|
| 598 |
"trial_params": null
|