End of training
Browse files
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 4.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 481,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 0.
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 2528,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 12.66,
|
| 3 |
+
"eval_cer": 0.9608244326394233,
|
| 4 |
+
"eval_loss": 4.192136287689209,
|
| 5 |
+
"eval_runtime": 52.7328,
|
| 6 |
"eval_samples": 481,
|
| 7 |
+
"eval_samples_per_second": 9.121,
|
| 8 |
+
"eval_steps_per_second": 1.157,
|
| 9 |
+
"eval_wer": 0.9294849931787176,
|
| 10 |
+
"train_loss": 1.9340453338623047,
|
| 11 |
+
"train_runtime": 2875.5301,
|
| 12 |
"train_samples": 2528,
|
| 13 |
+
"train_samples_per_second": 11.128,
|
| 14 |
+
"train_steps_per_second": 0.348
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_cer": 0.
|
| 4 |
-
"eval_loss": 4.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 481,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 1.
|
| 9 |
-
"eval_wer": 0.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 12.66,
|
| 3 |
+
"eval_cer": 0.9608244326394233,
|
| 4 |
+
"eval_loss": 4.192136287689209,
|
| 5 |
+
"eval_runtime": 52.7328,
|
| 6 |
"eval_samples": 481,
|
| 7 |
+
"eval_samples_per_second": 9.121,
|
| 8 |
+
"eval_steps_per_second": 1.157,
|
| 9 |
+
"eval_wer": 0.9294849931787176
|
| 10 |
}
|
runs/Dec18_17-32-59_129-146-179-188/events.out.tfevents.1671388061.129-146-179-188.141056.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1daf1563a93e258ca38036d9bbede0dd1eb72b9269b014b7ca7ddc67dd8eeb64
|
| 3 |
+
size 405
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 2528,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 12.66,
|
| 3 |
+
"train_loss": 1.9340453338623047,
|
| 4 |
+
"train_runtime": 2875.5301,
|
| 5 |
"train_samples": 2528,
|
| 6 |
+
"train_samples_per_second": 11.128,
|
| 7 |
+
"train_steps_per_second": 0.348
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 4.
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -358,18 +358,328 @@
|
|
| 358 |
"step": 500
|
| 359 |
},
|
| 360 |
{
|
| 361 |
-
"epoch": 6.
|
| 362 |
-
"
|
| 363 |
-
"
|
| 364 |
-
"
|
| 365 |
-
|
| 366 |
-
|
| 367 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
}
|
| 369 |
],
|
| 370 |
-
"max_steps":
|
| 371 |
-
"num_train_epochs":
|
| 372 |
-
"total_flos":
|
| 373 |
"trial_name": null,
|
| 374 |
"trial_params": null
|
| 375 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 4.192136287689209,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-1000",
|
| 4 |
+
"epoch": 12.658227848101266,
|
| 5 |
+
"global_step": 1000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 358 |
"step": 500
|
| 359 |
},
|
| 360 |
{
|
| 361 |
+
"epoch": 6.46,
|
| 362 |
+
"learning_rate": 3.8025000000000003e-07,
|
| 363 |
+
"loss": 4.3992,
|
| 364 |
+
"step": 510
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 6.58,
|
| 368 |
+
"learning_rate": 3.8775e-07,
|
| 369 |
+
"loss": 4.1942,
|
| 370 |
+
"step": 520
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 6.71,
|
| 374 |
+
"learning_rate": 3.9525000000000005e-07,
|
| 375 |
+
"loss": 4.339,
|
| 376 |
+
"step": 530
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 6.84,
|
| 380 |
+
"learning_rate": 4.0275000000000003e-07,
|
| 381 |
+
"loss": 4.2826,
|
| 382 |
+
"step": 540
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 6.96,
|
| 386 |
+
"learning_rate": 4.1025000000000006e-07,
|
| 387 |
+
"loss": 4.3752,
|
| 388 |
+
"step": 550
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 7.09,
|
| 392 |
+
"learning_rate": 4.1775000000000004e-07,
|
| 393 |
+
"loss": 3.9576,
|
| 394 |
+
"step": 560
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.22,
|
| 398 |
+
"learning_rate": 4.2524999999999997e-07,
|
| 399 |
+
"loss": 4.0814,
|
| 400 |
+
"step": 570
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 7.34,
|
| 404 |
+
"learning_rate": 4.3274999999999995e-07,
|
| 405 |
+
"loss": 4.3274,
|
| 406 |
+
"step": 580
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 7.47,
|
| 410 |
+
"learning_rate": 4.4025e-07,
|
| 411 |
+
"loss": 3.9749,
|
| 412 |
+
"step": 590
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 7.59,
|
| 416 |
+
"learning_rate": 4.4774999999999997e-07,
|
| 417 |
+
"loss": 3.8373,
|
| 418 |
+
"step": 600
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 7.72,
|
| 422 |
+
"learning_rate": 4.5525e-07,
|
| 423 |
+
"loss": 4.1076,
|
| 424 |
+
"step": 610
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 7.85,
|
| 428 |
+
"learning_rate": 4.6275e-07,
|
| 429 |
+
"loss": 3.8279,
|
| 430 |
+
"step": 620
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 7.97,
|
| 434 |
+
"learning_rate": 4.7025e-07,
|
| 435 |
+
"loss": 4.0263,
|
| 436 |
+
"step": 630
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 8.1,
|
| 440 |
+
"learning_rate": 4.7775e-07,
|
| 441 |
+
"loss": 3.8141,
|
| 442 |
+
"step": 640
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 8.23,
|
| 446 |
+
"learning_rate": 4.8525e-07,
|
| 447 |
+
"loss": 3.8791,
|
| 448 |
+
"step": 650
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 8.35,
|
| 452 |
+
"learning_rate": 4.927500000000001e-07,
|
| 453 |
+
"loss": 3.6906,
|
| 454 |
+
"step": 660
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 8.48,
|
| 458 |
+
"learning_rate": 5.0025e-07,
|
| 459 |
+
"loss": 4.0741,
|
| 460 |
+
"step": 670
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 8.61,
|
| 464 |
+
"learning_rate": 5.0775e-07,
|
| 465 |
+
"loss": 3.7322,
|
| 466 |
+
"step": 680
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 8.73,
|
| 470 |
+
"learning_rate": 5.152500000000001e-07,
|
| 471 |
+
"loss": 4.1834,
|
| 472 |
+
"step": 690
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 8.86,
|
| 476 |
+
"learning_rate": 5.2275e-07,
|
| 477 |
+
"loss": 3.8571,
|
| 478 |
+
"step": 700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 8.99,
|
| 482 |
+
"learning_rate": 5.3025e-07,
|
| 483 |
+
"loss": 3.9223,
|
| 484 |
+
"step": 710
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 9.11,
|
| 488 |
+
"learning_rate": 5.3775e-07,
|
| 489 |
+
"loss": 3.6997,
|
| 490 |
+
"step": 720
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 9.24,
|
| 494 |
+
"learning_rate": 5.4525e-07,
|
| 495 |
+
"loss": 4.0024,
|
| 496 |
+
"step": 730
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 9.37,
|
| 500 |
+
"learning_rate": 5.5275e-07,
|
| 501 |
+
"loss": 3.7653,
|
| 502 |
+
"step": 740
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 9.49,
|
| 506 |
+
"learning_rate": 5.602500000000001e-07,
|
| 507 |
+
"loss": 3.873,
|
| 508 |
+
"step": 750
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 9.62,
|
| 512 |
+
"learning_rate": 5.6775e-07,
|
| 513 |
+
"loss": 3.597,
|
| 514 |
+
"step": 760
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 9.75,
|
| 518 |
+
"learning_rate": 5.7525e-07,
|
| 519 |
+
"loss": 3.7902,
|
| 520 |
+
"step": 770
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 9.87,
|
| 524 |
+
"learning_rate": 5.827500000000001e-07,
|
| 525 |
+
"loss": 3.8314,
|
| 526 |
+
"step": 780
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 10.0,
|
| 530 |
+
"learning_rate": 5.902500000000001e-07,
|
| 531 |
+
"loss": 3.956,
|
| 532 |
+
"step": 790
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 10.13,
|
| 536 |
+
"learning_rate": 5.9775e-07,
|
| 537 |
+
"loss": 3.6297,
|
| 538 |
+
"step": 800
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 10.25,
|
| 542 |
+
"learning_rate": 6.0525e-07,
|
| 543 |
+
"loss": 4.0449,
|
| 544 |
+
"step": 810
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 10.38,
|
| 548 |
+
"learning_rate": 6.1275e-07,
|
| 549 |
+
"loss": 3.5669,
|
| 550 |
+
"step": 820
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 10.51,
|
| 554 |
+
"learning_rate": 6.2025e-07,
|
| 555 |
+
"loss": 3.6994,
|
| 556 |
+
"step": 830
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 10.63,
|
| 560 |
+
"learning_rate": 6.277499999999999e-07,
|
| 561 |
+
"loss": 3.6146,
|
| 562 |
+
"step": 840
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 10.76,
|
| 566 |
+
"learning_rate": 6.3525e-07,
|
| 567 |
+
"loss": 3.8637,
|
| 568 |
+
"step": 850
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 10.89,
|
| 572 |
+
"learning_rate": 6.4275e-07,
|
| 573 |
+
"loss": 3.7,
|
| 574 |
+
"step": 860
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 11.01,
|
| 578 |
+
"learning_rate": 6.5025e-07,
|
| 579 |
+
"loss": 3.8526,
|
| 580 |
+
"step": 870
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 11.14,
|
| 584 |
+
"learning_rate": 6.5775e-07,
|
| 585 |
+
"loss": 3.7276,
|
| 586 |
+
"step": 880
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 11.27,
|
| 590 |
+
"learning_rate": 6.6525e-07,
|
| 591 |
+
"loss": 3.989,
|
| 592 |
+
"step": 890
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 11.39,
|
| 596 |
+
"learning_rate": 6.7275e-07,
|
| 597 |
+
"loss": 3.7838,
|
| 598 |
+
"step": 900
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 11.52,
|
| 602 |
+
"learning_rate": 6.802500000000001e-07,
|
| 603 |
+
"loss": 3.698,
|
| 604 |
+
"step": 910
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 11.65,
|
| 608 |
+
"learning_rate": 6.8775e-07,
|
| 609 |
+
"loss": 3.4391,
|
| 610 |
+
"step": 920
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 11.77,
|
| 614 |
+
"learning_rate": 6.9525e-07,
|
| 615 |
+
"loss": 3.7371,
|
| 616 |
+
"step": 930
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 11.9,
|
| 620 |
+
"learning_rate": 7.027500000000001e-07,
|
| 621 |
+
"loss": 3.4201,
|
| 622 |
+
"step": 940
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 12.03,
|
| 626 |
+
"learning_rate": 7.1025e-07,
|
| 627 |
+
"loss": 3.7721,
|
| 628 |
+
"step": 950
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 12.15,
|
| 632 |
+
"learning_rate": 7.1775e-07,
|
| 633 |
+
"loss": 3.7842,
|
| 634 |
+
"step": 960
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 12.28,
|
| 638 |
+
"learning_rate": 7.2525e-07,
|
| 639 |
+
"loss": 3.8125,
|
| 640 |
+
"step": 970
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 12.41,
|
| 644 |
+
"learning_rate": 7.3275e-07,
|
| 645 |
+
"loss": 3.7673,
|
| 646 |
+
"step": 980
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 12.53,
|
| 650 |
+
"learning_rate": 7.395e-07,
|
| 651 |
+
"loss": 3.7048,
|
| 652 |
+
"step": 990
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 12.66,
|
| 656 |
+
"learning_rate": 7.47e-07,
|
| 657 |
+
"loss": 3.3986,
|
| 658 |
+
"step": 1000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 12.66,
|
| 662 |
+
"eval_cer": 0.9608244326394233,
|
| 663 |
+
"eval_loss": 4.192136287689209,
|
| 664 |
+
"eval_runtime": 52.6187,
|
| 665 |
+
"eval_samples_per_second": 9.141,
|
| 666 |
+
"eval_steps_per_second": 1.159,
|
| 667 |
+
"eval_wer": 0.9294849931787176,
|
| 668 |
+
"step": 1000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 12.66,
|
| 672 |
+
"step": 1000,
|
| 673 |
+
"total_flos": 3.5131981783950033e+19,
|
| 674 |
+
"train_loss": 1.9340453338623047,
|
| 675 |
+
"train_runtime": 2875.5301,
|
| 676 |
+
"train_samples_per_second": 11.128,
|
| 677 |
+
"train_steps_per_second": 0.348
|
| 678 |
}
|
| 679 |
],
|
| 680 |
+
"max_steps": 1000,
|
| 681 |
+
"num_train_epochs": 13,
|
| 682 |
+
"total_flos": 3.5131981783950033e+19,
|
| 683 |
"trial_name": null,
|
| 684 |
"trial_params": null
|
| 685 |
}
|