sengtha commited on
Commit
744c4be
·
verified ·
1 Parent(s): 2d9afcb

Training in progress, step 2500, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e2d78c4b7b01cdbe69b518e77ef413c1addc5a7be7dd0cef7c85281dc9615a6b
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfd776d53e573ac50b3cab211cedbe36f76e008afccf540bd65d642d7e00f35f
3
  size 290403936
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c6f065ff58cbda0182d812f0be127f7d97b9b728313c3bd6638b4a2a0304d992
3
  size 580960203
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc44d04546fddcfaa67d984dcec4c4fb2b04562806e221daa23cc3e536204c28
3
  size 580960203
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c6571af89e8b3a6b84957da68af45f6bbfb3fe95a466f1115b2e34506df566b
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0987affb1f2cf4f65c0e2061fb266d884cfb3c64f8abe1a36d6ece3e41d8456c
3
  size 14645
last-checkpoint/scaler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c170513fa2e39f9edc82ba4f7b5119ca1f241cb985bab799b9ac8cd7f8d297c9
3
  size 1383
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd1fd7616f7cd1ff00542dedd82b7891f230affa3f4a1e117e220b1112892950
3
  size 1383
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:33e315d7dff5da6fa2c0451708df97aea4b85242f6ab53444f9d2f871e96d726
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:429e74be3c16328a78d580d5d2d6dce9ef5926e6bfc78bc66543a7ea6ce73cf5
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -1,10 +1,10 @@
1
  {
2
- "best_global_step": 2000,
3
- "best_metric": 20.79825015287643,
4
- "best_model_checkpoint": "/workspace/whisper-base-khmer/checkpoint-2000",
5
- "epoch": 0.9583133684714902,
6
  "eval_steps": 500,
7
- "global_step": 2000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -604,6 +604,155 @@
604
  "eval_samples_per_second": 2.982,
605
  "eval_steps_per_second": 0.377,
606
  "step": 2000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
607
  }
608
  ],
609
  "logging_steps": 25,
@@ -623,7 +772,7 @@
623
  "attributes": {}
624
  }
625
  },
626
- "total_flos": 4.15103975424e+18,
627
  "train_batch_size": 16,
628
  "trial_name": null,
629
  "trial_params": null
 
1
  {
2
+ "best_global_step": 2500,
3
+ "best_metric": 19.40825062326544,
4
+ "best_model_checkpoint": "/workspace/whisper-base-khmer/checkpoint-2500",
5
+ "epoch": 1.1978917105893627,
6
  "eval_steps": 500,
7
+ "global_step": 2500,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
604
  "eval_samples_per_second": 2.982,
605
  "eval_steps_per_second": 0.377,
606
  "step": 2000
607
+ },
608
+ {
609
+ "epoch": 0.9702922855773838,
610
+ "grad_norm": 2.794386625289917,
611
+ "learning_rate": 1.307142857142857e-05,
612
+ "loss": 0.157889986038208,
613
+ "step": 2025
614
+ },
615
+ {
616
+ "epoch": 0.9822712026832774,
617
+ "grad_norm": 2.5509254932403564,
618
+ "learning_rate": 1.2736607142857141e-05,
619
+ "loss": 0.1639389991760254,
620
+ "step": 2050
621
+ },
622
+ {
623
+ "epoch": 0.9942501197891711,
624
+ "grad_norm": 2.380795478820801,
625
+ "learning_rate": 1.2401785714285714e-05,
626
+ "loss": 0.16085710525512695,
627
+ "step": 2075
628
+ },
629
+ {
630
+ "epoch": 1.0062290368950646,
631
+ "grad_norm": 2.5782344341278076,
632
+ "learning_rate": 1.2066964285714284e-05,
633
+ "loss": 0.1446963882446289,
634
+ "step": 2100
635
+ },
636
+ {
637
+ "epoch": 1.0182079540009583,
638
+ "grad_norm": 2.2326574325561523,
639
+ "learning_rate": 1.1732142857142855e-05,
640
+ "loss": 0.1341816806793213,
641
+ "step": 2125
642
+ },
643
+ {
644
+ "epoch": 1.030186871106852,
645
+ "grad_norm": 2.299788236618042,
646
+ "learning_rate": 1.1397321428571428e-05,
647
+ "loss": 0.12288969039916992,
648
+ "step": 2150
649
+ },
650
+ {
651
+ "epoch": 1.0421657882127455,
652
+ "grad_norm": 3.196174144744873,
653
+ "learning_rate": 1.1062499999999998e-05,
654
+ "loss": 0.12626271247863768,
655
+ "step": 2175
656
+ },
657
+ {
658
+ "epoch": 1.0541447053186392,
659
+ "grad_norm": 2.4003419876098633,
660
+ "learning_rate": 1.072767857142857e-05,
661
+ "loss": 0.13322246551513672,
662
+ "step": 2200
663
+ },
664
+ {
665
+ "epoch": 1.0661236224245327,
666
+ "grad_norm": 1.873734951019287,
667
+ "learning_rate": 1.0392857142857142e-05,
668
+ "loss": 0.12326155662536621,
669
+ "step": 2225
670
+ },
671
+ {
672
+ "epoch": 1.0781025395304265,
673
+ "grad_norm": 2.372314691543579,
674
+ "learning_rate": 1.0058035714285715e-05,
675
+ "loss": 0.12436847686767578,
676
+ "step": 2250
677
+ },
678
+ {
679
+ "epoch": 1.0900814566363202,
680
+ "grad_norm": 1.8567376136779785,
681
+ "learning_rate": 9.723214285714285e-06,
682
+ "loss": 0.12712265014648438,
683
+ "step": 2275
684
+ },
685
+ {
686
+ "epoch": 1.1020603737422137,
687
+ "grad_norm": 2.294780731201172,
688
+ "learning_rate": 9.388392857142856e-06,
689
+ "loss": 0.12966078758239746,
690
+ "step": 2300
691
+ },
692
+ {
693
+ "epoch": 1.1140392908481074,
694
+ "grad_norm": 2.4286816120147705,
695
+ "learning_rate": 9.053571428571429e-06,
696
+ "loss": 0.1323995018005371,
697
+ "step": 2325
698
+ },
699
+ {
700
+ "epoch": 1.1260182079540009,
701
+ "grad_norm": 2.349473714828491,
702
+ "learning_rate": 8.71875e-06,
703
+ "loss": 0.1263618278503418,
704
+ "step": 2350
705
+ },
706
+ {
707
+ "epoch": 1.1379971250598946,
708
+ "grad_norm": 2.5289194583892822,
709
+ "learning_rate": 8.38392857142857e-06,
710
+ "loss": 0.1253653907775879,
711
+ "step": 2375
712
+ },
713
+ {
714
+ "epoch": 1.1499760421657883,
715
+ "grad_norm": 2.636150598526001,
716
+ "learning_rate": 8.049107142857141e-06,
717
+ "loss": 0.12175737380981445,
718
+ "step": 2400
719
+ },
720
+ {
721
+ "epoch": 1.1619549592716818,
722
+ "grad_norm": 2.0082380771636963,
723
+ "learning_rate": 7.714285714285714e-06,
724
+ "loss": 0.12090700149536132,
725
+ "step": 2425
726
+ },
727
+ {
728
+ "epoch": 1.1739338763775755,
729
+ "grad_norm": 2.1470961570739746,
730
+ "learning_rate": 7.379464285714285e-06,
731
+ "loss": 0.11910880088806153,
732
+ "step": 2450
733
+ },
734
+ {
735
+ "epoch": 1.185912793483469,
736
+ "grad_norm": 2.586113691329956,
737
+ "learning_rate": 7.0446428571428565e-06,
738
+ "loss": 0.11959208488464355,
739
+ "step": 2475
740
+ },
741
+ {
742
+ "epoch": 1.1978917105893627,
743
+ "grad_norm": 2.3216049671173096,
744
+ "learning_rate": 6.709821428571429e-06,
745
+ "loss": 0.11850960731506348,
746
+ "step": 2500
747
+ },
748
+ {
749
+ "epoch": 1.1978917105893627,
750
+ "eval_cer": 19.40825062326544,
751
+ "eval_loss": 0.08071762323379517,
752
+ "eval_runtime": 167.7275,
753
+ "eval_samples_per_second": 3.017,
754
+ "eval_steps_per_second": 0.382,
755
+ "step": 2500
756
  }
757
  ],
758
  "logging_steps": 25,
 
772
  "attributes": {}
773
  }
774
  },
775
+ "total_flos": 5.18828081283072e+18,
776
  "train_batch_size": 16,
777
  "trial_name": null,
778
  "trial_params": null