Commit
•
b09b7d8
1
Parent(s):
3ff632d
Training in progress, step 3300
Browse files- {checkpoint-2100 → checkpoint-3200}/config.json +0 -0
- {checkpoint-2100 → checkpoint-3200}/optimizer.pt +1 -1
- {checkpoint-2100 → checkpoint-3200}/preprocessor_config.json +0 -0
- {checkpoint-2200 → checkpoint-3200}/pytorch_model.bin +1 -1
- {checkpoint-2200 → checkpoint-3200}/rng_state.pth +1 -1
- {checkpoint-2200 → checkpoint-3200}/scaler.pt +1 -1
- {checkpoint-2100 → checkpoint-3200}/scheduler.pt +1 -1
- {checkpoint-2200 → checkpoint-3200}/trainer_state.json +213 -3
- {checkpoint-2100 → checkpoint-3200}/training_args.bin +0 -0
- {checkpoint-2200 → checkpoint-3300}/config.json +0 -0
- {checkpoint-2200 → checkpoint-3300}/optimizer.pt +1 -1
- {checkpoint-2200 → checkpoint-3300}/preprocessor_config.json +0 -0
- {checkpoint-2100 → checkpoint-3300}/pytorch_model.bin +1 -1
- {checkpoint-2100 → checkpoint-3300}/rng_state.pth +1 -1
- {checkpoint-2100 → checkpoint-3300}/scaler.pt +1 -1
- {checkpoint-2200 → checkpoint-3300}/scheduler.pt +1 -1
- {checkpoint-2100 → checkpoint-3300}/trainer_state.json +255 -3
- {checkpoint-2200 → checkpoint-3300}/training_args.bin +0 -0
- pytorch_model.bin +1 -1
{checkpoint-2100 → checkpoint-3200}/config.json
RENAMED
|
File without changes
|
{checkpoint-2100 → checkpoint-3200}/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 sha256:3b096ae7b366c7c999b92a9c0d6ce1c8f70bba3cd30c87556eab078c149c9e3d
|
| 3 |
size 2490946501
|
{checkpoint-2100 → checkpoint-3200}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-2200 → checkpoint-3200}/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:f1aab560189c775afd4d52f128e4bd0cfd72f760ad25aab0de2f45c34b0b45e2
|
| 3 |
size 1262344621
|
{checkpoint-2200 → checkpoint-3200}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86fe28654777ce9becd5efc4b0839519aac223aa554449121cb9e9b1d6391db8
|
| 3 |
size 14575
|
{checkpoint-2200 → checkpoint-3200}/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:bd7c34e3c6e4a21fa45cdb54a89b626e4c29dfd67a2bac43d63b8bce896a050f
|
| 3 |
size 557
|
{checkpoint-2100 → checkpoint-3200}/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:d9eae38967de4c32ee3782c111dc2a1911655d360e132f06fbb454136bb7010c
|
| 3 |
size 627
|
{checkpoint-2200 → checkpoint-3200}/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,
|
|
@@ -468,11 +468,221 @@
|
|
| 468 |
"eval_samples_per_second": 23.695,
|
| 469 |
"eval_steps_per_second": 2.982,
|
| 470 |
"step": 2200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 471 |
}
|
| 472 |
],
|
| 473 |
"max_steps": 3675,
|
| 474 |
"num_train_epochs": 25,
|
| 475 |
-
"total_flos":
|
| 476 |
"trial_name": null,
|
| 477 |
"trial_params": null
|
| 478 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 21.768707482993197,
|
| 5 |
+
"global_step": 3200,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 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 |
+
"epoch": 21.43,
|
| 663 |
+
"learning_rate": 4.9889763779527555e-05,
|
| 664 |
+
"loss": 0.1578,
|
| 665 |
+
"step": 3150
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 21.77,
|
| 669 |
+
"learning_rate": 4.516535433070866e-05,
|
| 670 |
+
"loss": 0.1742,
|
| 671 |
+
"step": 3200
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 21.77,
|
| 675 |
+
"eval_cer": 0.148638512962661,
|
| 676 |
+
"eval_loss": 0.3685232102870941,
|
| 677 |
+
"eval_runtime": 25.1361,
|
| 678 |
+
"eval_samples_per_second": 23.393,
|
| 679 |
+
"eval_steps_per_second": 2.944,
|
| 680 |
+
"step": 3200
|
| 681 |
}
|
| 682 |
],
|
| 683 |
"max_steps": 3675,
|
| 684 |
"num_train_epochs": 25,
|
| 685 |
+
"total_flos": 1.242459434699676e+19,
|
| 686 |
"trial_name": null,
|
| 687 |
"trial_params": null
|
| 688 |
}
|
{checkpoint-2100 → checkpoint-3200}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-2200 → checkpoint-3300}/config.json
RENAMED
|
File without changes
|
{checkpoint-2200 → checkpoint-3300}/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 sha256:8b656fc491d0ed8f6ebfccb7267d33dde8cfdd3fc42b0f12ed1ea145c2467b56
|
| 3 |
size 2490946501
|
{checkpoint-2200 → checkpoint-3300}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-2100 → checkpoint-3300}/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 sha256:71b3268161afb2bcdf8b28a1a1d911f835d6d1a33dd0c929b5d6cf7248b065dc
|
| 3 |
size 1262344621
|
{checkpoint-2100 → checkpoint-3300}/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:e8d02f162a600f1e16123bc1f3de13d710b69a81bc405e4671daab4d2f276389
|
| 3 |
size 14639
|
{checkpoint-2100 → checkpoint-3300}/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:f5d87cd8a2522fd3677eac24590995de677069aa54288fb9a97a9c0b1ce60c09
|
| 3 |
size 557
|
{checkpoint-2200 → checkpoint-3300}/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 sha256:96a17614a18bd66b442195d1c5c4d8d5494ab5df6a2f462116d009a1efdf394c
|
| 3 |
size 627
|
{checkpoint-2100 → checkpoint-3300}/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,
|
|
@@ -447,11 +447,263 @@
|
|
| 447 |
"eval_samples_per_second": 23.673,
|
| 448 |
"eval_steps_per_second": 2.979,
|
| 449 |
"step": 2100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 450 |
}
|
| 451 |
],
|
| 452 |
"max_steps": 3675,
|
| 453 |
"num_train_epochs": 25,
|
| 454 |
-
"total_flos":
|
| 455 |
"trial_name": null,
|
| 456 |
"trial_params": null
|
| 457 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 22.448979591836736,
|
| 5 |
+
"global_step": 3300,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 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 |
+
"epoch": 21.43,
|
| 663 |
+
"learning_rate": 4.9889763779527555e-05,
|
| 664 |
+
"loss": 0.1578,
|
| 665 |
+
"step": 3150
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 21.77,
|
| 669 |
+
"learning_rate": 4.516535433070866e-05,
|
| 670 |
+
"loss": 0.1742,
|
| 671 |
+
"step": 3200
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 21.77,
|
| 675 |
+
"eval_cer": 0.148638512962661,
|
| 676 |
+
"eval_loss": 0.3685232102870941,
|
| 677 |
+
"eval_runtime": 25.1361,
|
| 678 |
+
"eval_samples_per_second": 23.393,
|
| 679 |
+
"eval_steps_per_second": 2.944,
|
| 680 |
+
"step": 3200
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 22.11,
|
| 684 |
+
"learning_rate": 4.0440944881889764e-05,
|
| 685 |
+
"loss": 0.1666,
|
| 686 |
+
"step": 3250
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 22.45,
|
| 690 |
+
"learning_rate": 3.571653543307086e-05,
|
| 691 |
+
"loss": 0.1605,
|
| 692 |
+
"step": 3300
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 22.45,
|
| 696 |
+
"eval_cer": 0.14746453611609325,
|
| 697 |
+
"eval_loss": 0.3691498637199402,
|
| 698 |
+
"eval_runtime": 24.7916,
|
| 699 |
+
"eval_samples_per_second": 23.718,
|
| 700 |
+
"eval_steps_per_second": 2.985,
|
| 701 |
+
"step": 3300
|
| 702 |
}
|
| 703 |
],
|
| 704 |
"max_steps": 3675,
|
| 705 |
"num_train_epochs": 25,
|
| 706 |
+
"total_flos": 1.2812925033719341e+19,
|
| 707 |
"trial_name": null,
|
| 708 |
"trial_params": null
|
| 709 |
}
|
{checkpoint-2200 → checkpoint-3300}/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 sha256:71b3268161afb2bcdf8b28a1a1d911f835d6d1a33dd0c929b5d6cf7248b065dc
|
| 3 |
size 1262344621
|