Training in progress, step 3000, 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:cdc164423f2a3f84efc5dfbeba06f3615eef5f9d37c1db29ef7adf3cb00ef228
|
| 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 sha256:45427cbf6396a019fcab45f20b6872b19cfd38c2afec6948c7f84225ab41122e
|
| 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:a9e77c3664e6c2303c974515f610c095940e1b9f1a09380dcd8d25d4c4eb1d05
|
| 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 sha256:9965564127cf0748ae57ecf2b02aba0f15495da8346241873c057a0c14f61d6d
|
| 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": 1.
|
| 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,
|
|
@@ -585,6 +585,295 @@
|
|
| 585 |
"eval_samples_per_second": 2.398,
|
| 586 |
"eval_steps_per_second": 0.301,
|
| 587 |
"step": 2000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 588 |
}
|
| 589 |
],
|
| 590 |
"logging_steps": 25,
|
|
@@ -592,7 +881,7 @@
|
|
| 592 |
"num_input_tokens_seen": 0,
|
| 593 |
"num_train_epochs": 3,
|
| 594 |
"save_steps": 1000,
|
| 595 |
-
"total_flos":
|
| 596 |
"train_batch_size": 16,
|
| 597 |
"trial_name": null,
|
| 598 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 51.336001032657805,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-3000",
|
| 4 |
+
"epoch": 1.935483870967742,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 3000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 585 |
"eval_samples_per_second": 2.398,
|
| 586 |
"eval_steps_per_second": 0.301,
|
| 587 |
"step": 2000
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.3064516129032258,
|
| 591 |
+
"grad_norm": 9.99063777923584,
|
| 592 |
+
"learning_rate": 6.272401433691757e-06,
|
| 593 |
+
"loss": 0.588,
|
| 594 |
+
"step": 2025
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.3225806451612903,
|
| 598 |
+
"grad_norm": 13.123091697692871,
|
| 599 |
+
"learning_rate": 6.212664277180407e-06,
|
| 600 |
+
"loss": 0.5886,
|
| 601 |
+
"step": 2050
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.3387096774193548,
|
| 605 |
+
"grad_norm": 10.930394172668457,
|
| 606 |
+
"learning_rate": 6.152927120669057e-06,
|
| 607 |
+
"loss": 0.6117,
|
| 608 |
+
"step": 2075
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 1.3548387096774195,
|
| 612 |
+
"grad_norm": 12.531543731689453,
|
| 613 |
+
"learning_rate": 6.0931899641577065e-06,
|
| 614 |
+
"loss": 0.5931,
|
| 615 |
+
"step": 2100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.370967741935484,
|
| 619 |
+
"grad_norm": 13.16308307647705,
|
| 620 |
+
"learning_rate": 6.033452807646356e-06,
|
| 621 |
+
"loss": 0.598,
|
| 622 |
+
"step": 2125
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 1.3870967741935485,
|
| 626 |
+
"grad_norm": 11.17799186706543,
|
| 627 |
+
"learning_rate": 5.973715651135007e-06,
|
| 628 |
+
"loss": 0.6141,
|
| 629 |
+
"step": 2150
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.403225806451613,
|
| 633 |
+
"grad_norm": 10.640506744384766,
|
| 634 |
+
"learning_rate": 5.9139784946236566e-06,
|
| 635 |
+
"loss": 0.5682,
|
| 636 |
+
"step": 2175
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.4193548387096775,
|
| 640 |
+
"grad_norm": 11.789594650268555,
|
| 641 |
+
"learning_rate": 5.854241338112307e-06,
|
| 642 |
+
"loss": 0.5598,
|
| 643 |
+
"step": 2200
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.435483870967742,
|
| 647 |
+
"grad_norm": 11.937474250793457,
|
| 648 |
+
"learning_rate": 5.794504181600956e-06,
|
| 649 |
+
"loss": 0.6344,
|
| 650 |
+
"step": 2225
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.4516129032258065,
|
| 654 |
+
"grad_norm": 14.106030464172363,
|
| 655 |
+
"learning_rate": 5.734767025089606e-06,
|
| 656 |
+
"loss": 0.5783,
|
| 657 |
+
"step": 2250
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 1.467741935483871,
|
| 661 |
+
"grad_norm": 12.365781784057617,
|
| 662 |
+
"learning_rate": 5.675029868578256e-06,
|
| 663 |
+
"loss": 0.6335,
|
| 664 |
+
"step": 2275
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 1.4838709677419355,
|
| 668 |
+
"grad_norm": 14.670917510986328,
|
| 669 |
+
"learning_rate": 5.615292712066906e-06,
|
| 670 |
+
"loss": 0.5988,
|
| 671 |
+
"step": 2300
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 1.5,
|
| 675 |
+
"grad_norm": 10.45535659790039,
|
| 676 |
+
"learning_rate": 5.555555555555557e-06,
|
| 677 |
+
"loss": 0.5912,
|
| 678 |
+
"step": 2325
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.5161290322580645,
|
| 682 |
+
"grad_norm": 15.059216499328613,
|
| 683 |
+
"learning_rate": 5.495818399044206e-06,
|
| 684 |
+
"loss": 0.5405,
|
| 685 |
+
"step": 2350
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 1.532258064516129,
|
| 689 |
+
"grad_norm": 12.705628395080566,
|
| 690 |
+
"learning_rate": 5.436081242532856e-06,
|
| 691 |
+
"loss": 0.5816,
|
| 692 |
+
"step": 2375
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.5483870967741935,
|
| 696 |
+
"grad_norm": 14.382452964782715,
|
| 697 |
+
"learning_rate": 5.376344086021506e-06,
|
| 698 |
+
"loss": 0.5437,
|
| 699 |
+
"step": 2400
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.564516129032258,
|
| 703 |
+
"grad_norm": 10.80752944946289,
|
| 704 |
+
"learning_rate": 5.316606929510155e-06,
|
| 705 |
+
"loss": 0.5975,
|
| 706 |
+
"step": 2425
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.5806451612903225,
|
| 710 |
+
"grad_norm": 12.146509170532227,
|
| 711 |
+
"learning_rate": 5.2568697729988065e-06,
|
| 712 |
+
"loss": 0.599,
|
| 713 |
+
"step": 2450
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.596774193548387,
|
| 717 |
+
"grad_norm": 12.145088195800781,
|
| 718 |
+
"learning_rate": 5.197132616487456e-06,
|
| 719 |
+
"loss": 0.6506,
|
| 720 |
+
"step": 2475
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.6129032258064515,
|
| 724 |
+
"grad_norm": 13.103174209594727,
|
| 725 |
+
"learning_rate": 5.137395459976105e-06,
|
| 726 |
+
"loss": 0.5649,
|
| 727 |
+
"step": 2500
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.629032258064516,
|
| 731 |
+
"grad_norm": 13.602423667907715,
|
| 732 |
+
"learning_rate": 5.077658303464756e-06,
|
| 733 |
+
"loss": 0.5424,
|
| 734 |
+
"step": 2525
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.6451612903225805,
|
| 738 |
+
"grad_norm": 14.787790298461914,
|
| 739 |
+
"learning_rate": 5.017921146953405e-06,
|
| 740 |
+
"loss": 0.5628,
|
| 741 |
+
"step": 2550
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.661290322580645,
|
| 745 |
+
"grad_norm": 11.559283256530762,
|
| 746 |
+
"learning_rate": 4.9581839904420555e-06,
|
| 747 |
+
"loss": 0.6216,
|
| 748 |
+
"step": 2575
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.6774193548387095,
|
| 752 |
+
"grad_norm": 13.20376968383789,
|
| 753 |
+
"learning_rate": 4.898446833930705e-06,
|
| 754 |
+
"loss": 0.5694,
|
| 755 |
+
"step": 2600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.6935483870967742,
|
| 759 |
+
"grad_norm": 9.632781982421875,
|
| 760 |
+
"learning_rate": 4.838709677419355e-06,
|
| 761 |
+
"loss": 0.5808,
|
| 762 |
+
"step": 2625
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.7096774193548387,
|
| 766 |
+
"grad_norm": 12.304398536682129,
|
| 767 |
+
"learning_rate": 4.7789725209080055e-06,
|
| 768 |
+
"loss": 0.5777,
|
| 769 |
+
"step": 2650
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.7258064516129032,
|
| 773 |
+
"grad_norm": 11.025238990783691,
|
| 774 |
+
"learning_rate": 4.719235364396655e-06,
|
| 775 |
+
"loss": 0.5964,
|
| 776 |
+
"step": 2675
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.7419354838709677,
|
| 780 |
+
"grad_norm": 13.640275955200195,
|
| 781 |
+
"learning_rate": 4.659498207885305e-06,
|
| 782 |
+
"loss": 0.5936,
|
| 783 |
+
"step": 2700
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.7580645161290323,
|
| 787 |
+
"grad_norm": 14.28750991821289,
|
| 788 |
+
"learning_rate": 4.599761051373955e-06,
|
| 789 |
+
"loss": 0.5814,
|
| 790 |
+
"step": 2725
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.7741935483870968,
|
| 794 |
+
"grad_norm": 14.228248596191406,
|
| 795 |
+
"learning_rate": 4.540023894862605e-06,
|
| 796 |
+
"loss": 0.5881,
|
| 797 |
+
"step": 2750
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.7903225806451613,
|
| 801 |
+
"grad_norm": 12.126937866210938,
|
| 802 |
+
"learning_rate": 4.480286738351255e-06,
|
| 803 |
+
"loss": 0.5568,
|
| 804 |
+
"step": 2775
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.8064516129032258,
|
| 808 |
+
"grad_norm": 12.653525352478027,
|
| 809 |
+
"learning_rate": 4.420549581839905e-06,
|
| 810 |
+
"loss": 0.5988,
|
| 811 |
+
"step": 2800
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.8225806451612905,
|
| 815 |
+
"grad_norm": 10.851930618286133,
|
| 816 |
+
"learning_rate": 4.360812425328555e-06,
|
| 817 |
+
"loss": 0.6073,
|
| 818 |
+
"step": 2825
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.838709677419355,
|
| 822 |
+
"grad_norm": 12.00724983215332,
|
| 823 |
+
"learning_rate": 4.3010752688172045e-06,
|
| 824 |
+
"loss": 0.5739,
|
| 825 |
+
"step": 2850
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.8548387096774195,
|
| 829 |
+
"grad_norm": 10.997614860534668,
|
| 830 |
+
"learning_rate": 4.241338112305855e-06,
|
| 831 |
+
"loss": 0.5663,
|
| 832 |
+
"step": 2875
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.870967741935484,
|
| 836 |
+
"grad_norm": 12.384391784667969,
|
| 837 |
+
"learning_rate": 4.181600955794505e-06,
|
| 838 |
+
"loss": 0.5325,
|
| 839 |
+
"step": 2900
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 1.8870967741935485,
|
| 843 |
+
"grad_norm": 10.200772285461426,
|
| 844 |
+
"learning_rate": 4.121863799283155e-06,
|
| 845 |
+
"loss": 0.5918,
|
| 846 |
+
"step": 2925
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.903225806451613,
|
| 850 |
+
"grad_norm": 13.224651336669922,
|
| 851 |
+
"learning_rate": 4.062126642771804e-06,
|
| 852 |
+
"loss": 0.5399,
|
| 853 |
+
"step": 2950
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.9193548387096775,
|
| 857 |
+
"grad_norm": 10.611023902893066,
|
| 858 |
+
"learning_rate": 4.002389486260454e-06,
|
| 859 |
+
"loss": 0.5593,
|
| 860 |
+
"step": 2975
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.935483870967742,
|
| 864 |
+
"grad_norm": 10.110644340515137,
|
| 865 |
+
"learning_rate": 3.942652329749105e-06,
|
| 866 |
+
"loss": 0.5611,
|
| 867 |
+
"step": 3000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.935483870967742,
|
| 871 |
+
"eval_cer": 51.336001032657805,
|
| 872 |
+
"eval_loss": 0.6702780723571777,
|
| 873 |
+
"eval_runtime": 963.0475,
|
| 874 |
+
"eval_samples_per_second": 2.369,
|
| 875 |
+
"eval_steps_per_second": 0.297,
|
| 876 |
+
"step": 3000
|
| 877 |
}
|
| 878 |
],
|
| 879 |
"logging_steps": 25,
|
|
|
|
| 881 |
"num_input_tokens_seen": 0,
|
| 882 |
"num_train_epochs": 3,
|
| 883 |
"save_steps": 1000,
|
| 884 |
+
"total_flos": 1.38506562883584e+19,
|
| 885 |
"train_batch_size": 16,
|
| 886 |
"trial_name": null,
|
| 887 |
"trial_params": null
|