Commit
•
d7d6ee2
1
Parent(s):
2f23d51
Training in progress, step 2900
Browse files- {checkpoint-1700 → checkpoint-2800}/config.json +0 -0
- {checkpoint-1700 → checkpoint-2800}/optimizer.pt +1 -1
- {checkpoint-1700 → checkpoint-2800}/preprocessor_config.json +0 -0
- {checkpoint-1800 → checkpoint-2800}/pytorch_model.bin +1 -1
- {checkpoint-1800 → checkpoint-2800}/rng_state.pth +1 -1
- {checkpoint-1800 → checkpoint-2800}/scaler.pt +1 -1
- {checkpoint-1700 → checkpoint-2800}/scheduler.pt +1 -1
- {checkpoint-1800 → checkpoint-2800}/trainer_state.json +213 -3
- {checkpoint-1700 → checkpoint-2800}/training_args.bin +0 -0
- {checkpoint-1800 → checkpoint-2900}/config.json +0 -0
- {checkpoint-1800 → checkpoint-2900}/optimizer.pt +1 -1
- {checkpoint-1800 → checkpoint-2900}/preprocessor_config.json +0 -0
- {checkpoint-1700 → checkpoint-2900}/pytorch_model.bin +1 -1
- {checkpoint-1700 → checkpoint-2900}/rng_state.pth +1 -1
- {checkpoint-1700 → checkpoint-2900}/scaler.pt +1 -1
- {checkpoint-1800 → checkpoint-2900}/scheduler.pt +1 -1
- {checkpoint-1700 → checkpoint-2900}/trainer_state.json +255 -3
- {checkpoint-1800 → checkpoint-2900}/training_args.bin +0 -0
- pytorch_model.bin +1 -1
{checkpoint-1700 → checkpoint-2800}/config.json
RENAMED
|
File without changes
|
{checkpoint-1700 → checkpoint-2800}/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 sha256:284fe9fd3efe2fa6bfac3dde889d7df4a894fadafdf72a079a9aa70dae570bb6
|
| 3 |
size 2490946501
|
{checkpoint-1700 → checkpoint-2800}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-1800 → checkpoint-2800}/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:d79086abe43be08504757ae268247e875b4ca0ee40d8541768b0d1afc44dcff9
|
| 3 |
size 1262344621
|
{checkpoint-1800 → checkpoint-2800}/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 sha256:27687fa73261753e3f3ea1442dcfbb7ebac6e911817034879446db31b67ab26f
|
| 3 |
size 14575
|
{checkpoint-1800 → checkpoint-2800}/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 sha256:05b468846eb222417f6778288d55e188312cf2d6a8edd4f708c9744ca1711621
|
| 3 |
size 557
|
{checkpoint-1700 → checkpoint-2800}/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 sha256:2b188442f4189d4023ae493e876a79dd05e958bbee2b254952060c03892fff41
|
| 3 |
size 627
|
{checkpoint-1800 → checkpoint-2800}/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,
|
|
@@ -384,11 +384,221 @@
|
|
| 384 |
"eval_samples_per_second": 23.79,
|
| 385 |
"eval_steps_per_second": 2.994,
|
| 386 |
"step": 1800
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 387 |
}
|
| 388 |
],
|
| 389 |
"max_steps": 3675,
|
| 390 |
"num_train_epochs": 25,
|
| 391 |
-
"total_flos":
|
| 392 |
"trial_name": null,
|
| 393 |
"trial_params": null
|
| 394 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.047619047619047,
|
| 5 |
+
"global_step": 2800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 384 |
"eval_samples_per_second": 23.79,
|
| 385 |
"eval_steps_per_second": 2.994,
|
| 386 |
"step": 1800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 12.59,
|
| 390 |
+
"learning_rate": 0.0001727244094488189,
|
| 391 |
+
"loss": 0.3208,
|
| 392 |
+
"step": 1850
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 12.93,
|
| 396 |
+
"learning_rate": 0.000168,
|
| 397 |
+
"loss": 0.3063,
|
| 398 |
+
"step": 1900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 12.93,
|
| 402 |
+
"eval_cer": 0.15904125224196966,
|
| 403 |
+
"eval_loss": 0.3622555434703827,
|
| 404 |
+
"eval_runtime": 24.4729,
|
| 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 |
"max_steps": 3675,
|
| 600 |
"num_train_epochs": 25,
|
| 601 |
+
"total_flos": 1.0878260136221063e+19,
|
| 602 |
"trial_name": null,
|
| 603 |
"trial_params": null
|
| 604 |
}
|
{checkpoint-1700 → checkpoint-2800}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-1800 → checkpoint-2900}/config.json
RENAMED
|
File without changes
|
{checkpoint-1800 → checkpoint-2900}/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:da92116ac7249a9c2f78c6d915c53ed280d77d3eee1ec0daa3648db76ed65088
|
| 3 |
size 2490946501
|
{checkpoint-1800 → checkpoint-2900}/preprocessor_config.json
RENAMED
|
File without changes
|
{checkpoint-1700 → checkpoint-2900}/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:a6ffc772554604c775a56582be10709d4c3e456a09beff5590f70c6d6164bae7
|
| 3 |
size 1262344621
|
{checkpoint-1700 → checkpoint-2900}/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 sha256:2f89b2092406b7051521efc60918e39ddbc8577addacf224a167b0840e13a8b1
|
| 3 |
size 14639
|
{checkpoint-1700 → checkpoint-2900}/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:e9d0e0ea2050687876737e0313597d903728681aadc2ef7b13381f1a340209c4
|
| 3 |
size 557
|
{checkpoint-1800 → checkpoint-2900}/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 sha256:2a694febd2309e1fb1cb50db9210a98ee65b7677fbb4514ceb39bb98bbdadc9d
|
| 3 |
size 627
|
{checkpoint-1700 → checkpoint-2900}/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,
|
|
@@ -363,11 +363,263 @@
|
|
| 363 |
"eval_samples_per_second": 24.123,
|
| 364 |
"eval_steps_per_second": 3.036,
|
| 365 |
"step": 1700
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 366 |
}
|
| 367 |
],
|
| 368 |
"max_steps": 3675,
|
| 369 |
"num_train_epochs": 25,
|
| 370 |
-
"total_flos":
|
| 371 |
"trial_name": null,
|
| 372 |
"trial_params": null
|
| 373 |
}
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.727891156462587,
|
| 5 |
+
"global_step": 2900,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 363 |
"eval_samples_per_second": 24.123,
|
| 364 |
"eval_steps_per_second": 3.036,
|
| 365 |
"step": 1700
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 11.9,
|
| 369 |
+
"learning_rate": 0.00018217322834645667,
|
| 370 |
+
"loss": 0.3455,
|
| 371 |
+
"step": 1750
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 12.24,
|
| 375 |
+
"learning_rate": 0.00017744881889763777,
|
| 376 |
+
"loss": 0.3245,
|
| 377 |
+
"step": 1800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.24,
|
| 381 |
+
"eval_cer": 0.15816076960704387,
|
| 382 |
+
"eval_loss": 0.3740461468696594,
|
| 383 |
+
"eval_runtime": 24.7162,
|
| 384 |
+
"eval_samples_per_second": 23.79,
|
| 385 |
+
"eval_steps_per_second": 2.994,
|
| 386 |
+
"step": 1800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 12.59,
|
| 390 |
+
"learning_rate": 0.0001727244094488189,
|
| 391 |
+
"loss": 0.3208,
|
| 392 |
+
"step": 1850
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 12.93,
|
| 396 |
+
"learning_rate": 0.000168,
|
| 397 |
+
"loss": 0.3063,
|
| 398 |
+
"step": 1900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 12.93,
|
| 402 |
+
"eval_cer": 0.15904125224196966,
|
| 403 |
+
"eval_loss": 0.3622555434703827,
|
| 404 |
+
"eval_runtime": 24.4729,
|
| 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 |
"max_steps": 3675,
|
| 621 |
"num_train_epochs": 25,
|
| 622 |
+
"total_flos": 1.1249154319313052e+19,
|
| 623 |
"trial_name": null,
|
| 624 |
"trial_params": null
|
| 625 |
}
|
{checkpoint-1800 → checkpoint-2900}/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:a6ffc772554604c775a56582be10709d4c3e456a09beff5590f70c6d6164bae7
|
| 3 |
size 1262344621
|