Commit
β’
3ff632d
1
Parent(s):
d7d6ee2
Training in progress, step 3100
Browse files- {checkpoint-1900 β checkpoint-3000}/config.json +0 -0
- {checkpoint-1900 β checkpoint-3000}/optimizer.pt +1 -1
- {checkpoint-1900 β checkpoint-3000}/preprocessor_config.json +0 -0
- {checkpoint-2000 β checkpoint-3000}/pytorch_model.bin +1 -1
- {checkpoint-2000 β checkpoint-3000}/rng_state.pth +1 -1
- {checkpoint-2000 β checkpoint-3000}/scaler.pt +1 -1
- {checkpoint-1900 β checkpoint-3000}/scheduler.pt +1 -1
- {checkpoint-2000 β checkpoint-3000}/trainer_state.json +213 -3
- {checkpoint-1900 β checkpoint-3000}/training_args.bin +0 -0
- {checkpoint-2000 β checkpoint-3100}/config.json +0 -0
- {checkpoint-2000 β checkpoint-3100}/optimizer.pt +1 -1
- {checkpoint-2000 β checkpoint-3100}/preprocessor_config.json +0 -0
- {checkpoint-1900 β checkpoint-3100}/pytorch_model.bin +1 -1
- {checkpoint-1900 β checkpoint-3100}/rng_state.pth +2 -2
- {checkpoint-1900 β checkpoint-3100}/scaler.pt +1 -1
- {checkpoint-2000 β checkpoint-3100}/scheduler.pt +1 -1
- {checkpoint-1900 β checkpoint-3100}/trainer_state.json +255 -3
- {checkpoint-2000 β checkpoint-3100}/training_args.bin +0 -0
- pytorch_model.bin +1 -1
{checkpoint-1900 β checkpoint-3000}/config.json
RENAMED
File without changes
|
{checkpoint-1900 β checkpoint-3000}/optimizer.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2490946501
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:add5027d5fd3913c88e3ce4c98bba2cacc80d01308c6d7e2ef915041867428fd
|
3 |
size 2490946501
|
{checkpoint-1900 β checkpoint-3000}/preprocessor_config.json
RENAMED
File without changes
|
{checkpoint-2000 β checkpoint-3000}/pytorch_model.bin
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262344621
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17b0809c17d9dbb31d634626a048a7d434c078f79477bfe8f7a8cf3593908c7c
|
3 |
size 1262344621
|
{checkpoint-2000 β checkpoint-3000}/rng_state.pth
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14639
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:54e994db3699b40bee8c6677e4890450c771f5c163635dc7865e66b42965e184
|
3 |
size 14639
|
{checkpoint-2000 β checkpoint-3000}/scaler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 557
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e09035270122b5f507afb3c91029317a19c9951c55151aa7d9fc1f5691d8ea29
|
3 |
size 557
|
{checkpoint-1900 β checkpoint-3000}/scheduler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 627
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9e3cf2d9b187c53d4ab17e26aaa543c99e77f6fa20e6dcea24354d1208058809
|
3 |
size 627
|
{checkpoint-2000 β checkpoint-3000}/trainer_state.json
RENAMED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -426,11 +426,221 @@
|
|
426 |
"eval_samples_per_second": 23.518,
|
427 |
"eval_steps_per_second": 2.96,
|
428 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
429 |
}
|
430 |
],
|
431 |
"max_steps": 3675,
|
432 |
"num_train_epochs": 25,
|
433 |
-
"total_flos":
|
434 |
"trial_name": null,
|
435 |
"trial_params": null
|
436 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 20.408163265306122,
|
5 |
+
"global_step": 3000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
426 |
"eval_samples_per_second": 23.518,
|
427 |
"eval_steps_per_second": 2.96,
|
428 |
"step": 2000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 13.95,
|
432 |
+
"learning_rate": 0.0001538267716535433,
|
433 |
+
"loss": 0.279,
|
434 |
+
"step": 2050
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 14.29,
|
438 |
+
"learning_rate": 0.0001491023622047244,
|
439 |
+
"loss": 0.2674,
|
440 |
+
"step": 2100
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 14.29,
|
444 |
+
"eval_cer": 0.15731289744007826,
|
445 |
+
"eval_loss": 0.3531067371368408,
|
446 |
+
"eval_runtime": 24.8381,
|
447 |
+
"eval_samples_per_second": 23.673,
|
448 |
+
"eval_steps_per_second": 2.979,
|
449 |
+
"step": 2100
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 14.63,
|
453 |
+
"learning_rate": 0.0001443779527559055,
|
454 |
+
"loss": 0.2584,
|
455 |
+
"step": 2150
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 14.97,
|
459 |
+
"learning_rate": 0.0001396535433070866,
|
460 |
+
"loss": 0.2796,
|
461 |
+
"step": 2200
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 14.97,
|
465 |
+
"eval_cer": 0.14808413500733736,
|
466 |
+
"eval_loss": 0.3606802523136139,
|
467 |
+
"eval_runtime": 24.8151,
|
468 |
+
"eval_samples_per_second": 23.695,
|
469 |
+
"eval_steps_per_second": 2.982,
|
470 |
+
"step": 2200
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 15.31,
|
474 |
+
"learning_rate": 0.0001349291338582677,
|
475 |
+
"loss": 0.2462,
|
476 |
+
"step": 2250
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 15.65,
|
480 |
+
"learning_rate": 0.0001302047244094488,
|
481 |
+
"loss": 0.256,
|
482 |
+
"step": 2300
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 15.65,
|
486 |
+
"eval_cer": 0.15819338007500408,
|
487 |
+
"eval_loss": 0.3580550253391266,
|
488 |
+
"eval_runtime": 24.5695,
|
489 |
+
"eval_samples_per_second": 23.932,
|
490 |
+
"eval_steps_per_second": 3.012,
|
491 |
+
"step": 2300
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 15.99,
|
495 |
+
"learning_rate": 0.00012548031496062992,
|
496 |
+
"loss": 0.2524,
|
497 |
+
"step": 2350
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 16.33,
|
501 |
+
"learning_rate": 0.00012075590551181102,
|
502 |
+
"loss": 0.2219,
|
503 |
+
"step": 2400
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 16.33,
|
507 |
+
"eval_cer": 0.14801891407141693,
|
508 |
+
"eval_loss": 0.35925593972206116,
|
509 |
+
"eval_runtime": 24.982,
|
510 |
+
"eval_samples_per_second": 23.537,
|
511 |
+
"eval_steps_per_second": 2.962,
|
512 |
+
"step": 2400
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 16.67,
|
516 |
+
"learning_rate": 0.0001160314960629921,
|
517 |
+
"loss": 0.2364,
|
518 |
+
"step": 2450
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 17.01,
|
522 |
+
"learning_rate": 0.00011130708661417321,
|
523 |
+
"loss": 0.2291,
|
524 |
+
"step": 2500
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 17.01,
|
528 |
+
"eval_cer": 0.1471058209685309,
|
529 |
+
"eval_loss": 0.35567909479141235,
|
530 |
+
"eval_runtime": 24.4749,
|
531 |
+
"eval_samples_per_second": 24.025,
|
532 |
+
"eval_steps_per_second": 3.024,
|
533 |
+
"step": 2500
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 17.35,
|
537 |
+
"learning_rate": 0.00010658267716535431,
|
538 |
+
"loss": 0.2045,
|
539 |
+
"step": 2550
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 17.69,
|
543 |
+
"learning_rate": 0.00010185826771653542,
|
544 |
+
"loss": 0.2172,
|
545 |
+
"step": 2600
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 17.69,
|
549 |
+
"eval_cer": 0.14792108266753629,
|
550 |
+
"eval_loss": 0.3606509566307068,
|
551 |
+
"eval_runtime": 25.1105,
|
552 |
+
"eval_samples_per_second": 23.416,
|
553 |
+
"eval_steps_per_second": 2.947,
|
554 |
+
"step": 2600
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 18.03,
|
558 |
+
"learning_rate": 9.713385826771652e-05,
|
559 |
+
"loss": 0.2271,
|
560 |
+
"step": 2650
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 18.37,
|
564 |
+
"learning_rate": 9.240944881889763e-05,
|
565 |
+
"loss": 0.1858,
|
566 |
+
"step": 2700
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 18.37,
|
570 |
+
"eval_cer": 0.15144301320723952,
|
571 |
+
"eval_loss": 0.3589307963848114,
|
572 |
+
"eval_runtime": 24.5005,
|
573 |
+
"eval_samples_per_second": 24.0,
|
574 |
+
"eval_steps_per_second": 3.02,
|
575 |
+
"step": 2700
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 18.71,
|
579 |
+
"learning_rate": 8.768503937007873e-05,
|
580 |
+
"loss": 0.1995,
|
581 |
+
"step": 2750
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 19.05,
|
585 |
+
"learning_rate": 8.296062992125984e-05,
|
586 |
+
"loss": 0.1872,
|
587 |
+
"step": 2800
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 19.05,
|
591 |
+
"eval_cer": 0.14766019892385457,
|
592 |
+
"eval_loss": 0.36663514375686646,
|
593 |
+
"eval_runtime": 24.4383,
|
594 |
+
"eval_samples_per_second": 24.061,
|
595 |
+
"eval_steps_per_second": 3.028,
|
596 |
+
"step": 2800
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 19.39,
|
600 |
+
"learning_rate": 7.823622047244094e-05,
|
601 |
+
"loss": 0.177,
|
602 |
+
"step": 2850
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 19.73,
|
606 |
+
"learning_rate": 7.351181102362205e-05,
|
607 |
+
"loss": 0.1855,
|
608 |
+
"step": 2900
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 19.73,
|
612 |
+
"eval_cer": 0.1432577857492255,
|
613 |
+
"eval_loss": 0.3651330769062042,
|
614 |
+
"eval_runtime": 24.7826,
|
615 |
+
"eval_samples_per_second": 23.726,
|
616 |
+
"eval_steps_per_second": 2.986,
|
617 |
+
"step": 2900
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 20.07,
|
621 |
+
"learning_rate": 6.878740157480315e-05,
|
622 |
+
"loss": 0.1773,
|
623 |
+
"step": 2950
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 20.41,
|
627 |
+
"learning_rate": 6.406299212598424e-05,
|
628 |
+
"loss": 0.185,
|
629 |
+
"step": 3000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 20.41,
|
633 |
+
"eval_cer": 0.1447578672753954,
|
634 |
+
"eval_loss": 0.36554473638534546,
|
635 |
+
"eval_runtime": 24.7061,
|
636 |
+
"eval_samples_per_second": 23.8,
|
637 |
+
"eval_steps_per_second": 2.995,
|
638 |
+
"step": 3000
|
639 |
}
|
640 |
],
|
641 |
"max_steps": 3675,
|
642 |
"num_train_epochs": 25,
|
643 |
+
"total_flos": 1.1654661840261913e+19,
|
644 |
"trial_name": null,
|
645 |
"trial_params": null
|
646 |
}
|
{checkpoint-1900 β checkpoint-3000}/training_args.bin
RENAMED
File without changes
|
{checkpoint-2000 β checkpoint-3100}/config.json
RENAMED
File without changes
|
{checkpoint-2000 β checkpoint-3100}/optimizer.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2490946501
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e0547651e9fa5e5a88a9166ef0d5f076cee95faf27589392ad547f526db7c4fe
|
3 |
size 2490946501
|
{checkpoint-2000 β checkpoint-3100}/preprocessor_config.json
RENAMED
File without changes
|
{checkpoint-1900 β checkpoint-3100}/pytorch_model.bin
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262344621
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aa48ab3bfa27ac8b74f16b85b0b0e9423a7006947f26bc79b3a134b3a28201d9
|
3 |
size 1262344621
|
{checkpoint-1900 β checkpoint-3100}/rng_state.pth
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e6882a736f444e030f1e63185cd593b64dcbe537747a536c06d91a9ff822878
|
3 |
+
size 14575
|
{checkpoint-1900 β checkpoint-3100}/scaler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 557
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2a5a1c934e0cb72b172df77c7f977d3f9bcedc7c9991dd7cf5ab0d4a3a9e556e
|
3 |
size 557
|
{checkpoint-2000 β checkpoint-3100}/scheduler.pt
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 627
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4f224e654f13b383690d5d87143e399102b541cd0bfe5fa57c4e0ad3f1abceb1
|
3 |
size 627
|
{checkpoint-1900 β checkpoint-3100}/trainer_state.json
RENAMED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -405,11 +405,263 @@
|
|
405 |
"eval_samples_per_second": 24.027,
|
406 |
"eval_steps_per_second": 3.024,
|
407 |
"step": 1900
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
408 |
}
|
409 |
],
|
410 |
"max_steps": 3675,
|
411 |
"num_train_epochs": 25,
|
412 |
-
"total_flos":
|
413 |
"trial_name": null,
|
414 |
"trial_params": null
|
415 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 21.08843537414966,
|
5 |
+
"global_step": 3100,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
405 |
"eval_samples_per_second": 24.027,
|
406 |
"eval_steps_per_second": 3.024,
|
407 |
"step": 1900
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 13.27,
|
411 |
+
"learning_rate": 0.0001632755905511811,
|
412 |
+
"loss": 0.3019,
|
413 |
+
"step": 1950
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 13.61,
|
417 |
+
"learning_rate": 0.00015855118110236219,
|
418 |
+
"loss": 0.2945,
|
419 |
+
"step": 2000
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 13.61,
|
423 |
+
"eval_cer": 0.16634599706505787,
|
424 |
+
"eval_loss": 0.3725011348724365,
|
425 |
+
"eval_runtime": 25.0023,
|
426 |
+
"eval_samples_per_second": 23.518,
|
427 |
+
"eval_steps_per_second": 2.96,
|
428 |
+
"step": 2000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 13.95,
|
432 |
+
"learning_rate": 0.0001538267716535433,
|
433 |
+
"loss": 0.279,
|
434 |
+
"step": 2050
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 14.29,
|
438 |
+
"learning_rate": 0.0001491023622047244,
|
439 |
+
"loss": 0.2674,
|
440 |
+
"step": 2100
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 14.29,
|
444 |
+
"eval_cer": 0.15731289744007826,
|
445 |
+
"eval_loss": 0.3531067371368408,
|
446 |
+
"eval_runtime": 24.8381,
|
447 |
+
"eval_samples_per_second": 23.673,
|
448 |
+
"eval_steps_per_second": 2.979,
|
449 |
+
"step": 2100
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 14.63,
|
453 |
+
"learning_rate": 0.0001443779527559055,
|
454 |
+
"loss": 0.2584,
|
455 |
+
"step": 2150
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 14.97,
|
459 |
+
"learning_rate": 0.0001396535433070866,
|
460 |
+
"loss": 0.2796,
|
461 |
+
"step": 2200
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 14.97,
|
465 |
+
"eval_cer": 0.14808413500733736,
|
466 |
+
"eval_loss": 0.3606802523136139,
|
467 |
+
"eval_runtime": 24.8151,
|
468 |
+
"eval_samples_per_second": 23.695,
|
469 |
+
"eval_steps_per_second": 2.982,
|
470 |
+
"step": 2200
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 15.31,
|
474 |
+
"learning_rate": 0.0001349291338582677,
|
475 |
+
"loss": 0.2462,
|
476 |
+
"step": 2250
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 15.65,
|
480 |
+
"learning_rate": 0.0001302047244094488,
|
481 |
+
"loss": 0.256,
|
482 |
+
"step": 2300
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 15.65,
|
486 |
+
"eval_cer": 0.15819338007500408,
|
487 |
+
"eval_loss": 0.3580550253391266,
|
488 |
+
"eval_runtime": 24.5695,
|
489 |
+
"eval_samples_per_second": 23.932,
|
490 |
+
"eval_steps_per_second": 3.012,
|
491 |
+
"step": 2300
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 15.99,
|
495 |
+
"learning_rate": 0.00012548031496062992,
|
496 |
+
"loss": 0.2524,
|
497 |
+
"step": 2350
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 16.33,
|
501 |
+
"learning_rate": 0.00012075590551181102,
|
502 |
+
"loss": 0.2219,
|
503 |
+
"step": 2400
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 16.33,
|
507 |
+
"eval_cer": 0.14801891407141693,
|
508 |
+
"eval_loss": 0.35925593972206116,
|
509 |
+
"eval_runtime": 24.982,
|
510 |
+
"eval_samples_per_second": 23.537,
|
511 |
+
"eval_steps_per_second": 2.962,
|
512 |
+
"step": 2400
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 16.67,
|
516 |
+
"learning_rate": 0.0001160314960629921,
|
517 |
+
"loss": 0.2364,
|
518 |
+
"step": 2450
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 17.01,
|
522 |
+
"learning_rate": 0.00011130708661417321,
|
523 |
+
"loss": 0.2291,
|
524 |
+
"step": 2500
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 17.01,
|
528 |
+
"eval_cer": 0.1471058209685309,
|
529 |
+
"eval_loss": 0.35567909479141235,
|
530 |
+
"eval_runtime": 24.4749,
|
531 |
+
"eval_samples_per_second": 24.025,
|
532 |
+
"eval_steps_per_second": 3.024,
|
533 |
+
"step": 2500
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 17.35,
|
537 |
+
"learning_rate": 0.00010658267716535431,
|
538 |
+
"loss": 0.2045,
|
539 |
+
"step": 2550
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 17.69,
|
543 |
+
"learning_rate": 0.00010185826771653542,
|
544 |
+
"loss": 0.2172,
|
545 |
+
"step": 2600
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 17.69,
|
549 |
+
"eval_cer": 0.14792108266753629,
|
550 |
+
"eval_loss": 0.3606509566307068,
|
551 |
+
"eval_runtime": 25.1105,
|
552 |
+
"eval_samples_per_second": 23.416,
|
553 |
+
"eval_steps_per_second": 2.947,
|
554 |
+
"step": 2600
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 18.03,
|
558 |
+
"learning_rate": 9.713385826771652e-05,
|
559 |
+
"loss": 0.2271,
|
560 |
+
"step": 2650
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 18.37,
|
564 |
+
"learning_rate": 9.240944881889763e-05,
|
565 |
+
"loss": 0.1858,
|
566 |
+
"step": 2700
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 18.37,
|
570 |
+
"eval_cer": 0.15144301320723952,
|
571 |
+
"eval_loss": 0.3589307963848114,
|
572 |
+
"eval_runtime": 24.5005,
|
573 |
+
"eval_samples_per_second": 24.0,
|
574 |
+
"eval_steps_per_second": 3.02,
|
575 |
+
"step": 2700
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 18.71,
|
579 |
+
"learning_rate": 8.768503937007873e-05,
|
580 |
+
"loss": 0.1995,
|
581 |
+
"step": 2750
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 19.05,
|
585 |
+
"learning_rate": 8.296062992125984e-05,
|
586 |
+
"loss": 0.1872,
|
587 |
+
"step": 2800
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 19.05,
|
591 |
+
"eval_cer": 0.14766019892385457,
|
592 |
+
"eval_loss": 0.36663514375686646,
|
593 |
+
"eval_runtime": 24.4383,
|
594 |
+
"eval_samples_per_second": 24.061,
|
595 |
+
"eval_steps_per_second": 3.028,
|
596 |
+
"step": 2800
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 19.39,
|
600 |
+
"learning_rate": 7.823622047244094e-05,
|
601 |
+
"loss": 0.177,
|
602 |
+
"step": 2850
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 19.73,
|
606 |
+
"learning_rate": 7.351181102362205e-05,
|
607 |
+
"loss": 0.1855,
|
608 |
+
"step": 2900
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 19.73,
|
612 |
+
"eval_cer": 0.1432577857492255,
|
613 |
+
"eval_loss": 0.3651330769062042,
|
614 |
+
"eval_runtime": 24.7826,
|
615 |
+
"eval_samples_per_second": 23.726,
|
616 |
+
"eval_steps_per_second": 2.986,
|
617 |
+
"step": 2900
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 20.07,
|
621 |
+
"learning_rate": 6.878740157480315e-05,
|
622 |
+
"loss": 0.1773,
|
623 |
+
"step": 2950
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 20.41,
|
627 |
+
"learning_rate": 6.406299212598424e-05,
|
628 |
+
"loss": 0.185,
|
629 |
+
"step": 3000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 20.41,
|
633 |
+
"eval_cer": 0.1447578672753954,
|
634 |
+
"eval_loss": 0.36554473638534546,
|
635 |
+
"eval_runtime": 24.7061,
|
636 |
+
"eval_samples_per_second": 23.8,
|
637 |
+
"eval_steps_per_second": 2.995,
|
638 |
+
"step": 3000
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 20.75,
|
642 |
+
"learning_rate": 5.933858267716535e-05,
|
643 |
+
"loss": 0.1735,
|
644 |
+
"step": 3050
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 21.09,
|
648 |
+
"learning_rate": 5.461417322834645e-05,
|
649 |
+
"loss": 0.1599,
|
650 |
+
"step": 3100
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 21.09,
|
654 |
+
"eval_cer": 0.14867112343062122,
|
655 |
+
"eval_loss": 0.37344449758529663,
|
656 |
+
"eval_runtime": 24.8752,
|
657 |
+
"eval_samples_per_second": 23.638,
|
658 |
+
"eval_steps_per_second": 2.975,
|
659 |
+
"step": 3100
|
660 |
}
|
661 |
],
|
662 |
"max_steps": 3675,
|
663 |
"num_train_epochs": 25,
|
664 |
+
"total_flos": 1.2047450649580579e+19,
|
665 |
"trial_name": null,
|
666 |
"trial_params": null
|
667 |
}
|
{checkpoint-2000 β checkpoint-3100}/training_args.bin
RENAMED
File without changes
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262344621
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aa48ab3bfa27ac8b74f16b85b0b0e9423a7006947f26bc79b3a134b3a28201d9
|
3 |
size 1262344621
|