Hugging Face's logo

noflm
/
whisper-base-ja-cv11

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Japanese
whisper
whisper-event
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
2
noflm commited on
Commit
7771c3e
•
1 Parent(s): d83f8ea

Training in progress, step 3000

Browse files
Files changed (30) hide show
  1. .gitignore +0 -1
  2. checkpoint-1000/config.json +54 -0
  3. checkpoint-1000/optimizer.pt +3 -0
  4. checkpoint-1000/preprocessor_config.json +0 -0
  5. checkpoint-1000/pytorch_model.bin +3 -0
  6. checkpoint-1000/rng_state.pth +3 -0
  7. checkpoint-1000/scaler.pt +3 -0
  8. checkpoint-1000/scheduler.pt +3 -0
  9. checkpoint-1000/trainer_state.json +37 -0
  10. checkpoint-1000/training_args.bin +3 -0
  11. checkpoint-2000/config.json +54 -0
  12. checkpoint-2000/optimizer.pt +3 -0
  13. checkpoint-2000/preprocessor_config.json +0 -0
  14. checkpoint-2000/pytorch_model.bin +3 -0
  15. checkpoint-2000/rng_state.pth +3 -0
  16. checkpoint-2000/scaler.pt +3 -0
  17. checkpoint-2000/scheduler.pt +3 -0
  18. checkpoint-2000/trainer_state.json +58 -0
  19. checkpoint-2000/training_args.bin +3 -0
  20. checkpoint-3000/config.json +54 -0
  21. checkpoint-3000/optimizer.pt +3 -0
  22. checkpoint-3000/preprocessor_config.json +0 -0
  23. checkpoint-3000/pytorch_model.bin +3 -0
  24. checkpoint-3000/rng_state.pth +3 -0
  25. checkpoint-3000/scaler.pt +3 -0
  26. checkpoint-3000/scheduler.pt +3 -0
  27. checkpoint-3000/trainer_state.json +79 -0
  28. checkpoint-3000/training_args.bin +3 -0
  29. pytorch_model.bin +1 -1
  30. runs/Dec27_09-55-14_srv/events.out.tfevents.1672134971.srv.1172111.0 +2 -2
.gitignore DELETED
@@ -1 +0,0 @@
1
- checkpoint-*/
 
 
checkpoint-1000/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-base",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50257
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "d_model": 512,
15
+ "decoder_attention_heads": 8,
16
+ "decoder_ffn_dim": 2048,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 8,
22
+ "encoder_ffn_dim": 2048,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 6,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": [
27
+ [
28
+ 1,
29
+ 50266
30
+ ],
31
+ [
32
+ 2,
33
+ 50359
34
+ ],
35
+ [
36
+ 3,
37
+ 50363
38
+ ]
39
+ ],
40
+ "init_std": 0.02,
41
+ "is_encoder_decoder": true,
42
+ "max_length": 448,
43
+ "max_source_positions": 1500,
44
+ "max_target_positions": 448,
45
+ "model_type": "whisper",
46
+ "num_hidden_layers": 6,
47
+ "num_mel_bins": 80,
48
+ "pad_token_id": 50257,
49
+ "scale_embedding": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.26.0.dev0",
52
+ "use_cache": false,
53
+ "vocab_size": 51865
54
+ }
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f63f9bf49792ca9ae6edf0f39114a942a4e93d15e36308ed2618d11c0ec8a422
3
+ size 580895237
checkpoint-1000/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03b67dc263908466ddf56394bb55830c8275a3a5888cb54b0866d244049dcd5
3
+ size 290458721
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:505e2b82433c65c133ad28a1f8a434ffa52e8a586b6eb462b17c450e063a9112
3
+ size 14575
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26dbe439de3dd73741374b46df74ee981e9ad3b04a0c5c8a564c625a65b8aee5
3
+ size 557
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6228413127edc44d11ec9d1fb398bdf140ed68debc76f0505ae35c8406acb91b
3
+ size 627
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0157,
5
+ "global_step": 1000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.02,
12
+ "learning_rate": 9.960000000000001e-06,
13
+ "loss": 0.5615,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.02,
18
+ "learning_rate": 1e-05,
19
+ "loss": 0.3273,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 3.02,
24
+ "eval_loss": 0.4225325584411621,
25
+ "eval_runtime": 1489.9228,
26
+ "eval_samples_per_second": 3.09,
27
+ "eval_steps_per_second": 0.193,
28
+ "eval_wer": 20.82533081951615,
29
+ "step": 1000
30
+ }
31
+ ],
32
+ "max_steps": 10000,
33
+ "num_train_epochs": 9223372036854775807,
34
+ "total_flos": 2.07123911737344e+18,
35
+ "trial_name": null,
36
+ "trial_params": null
37
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63250d2e981e0cfad1a5508019d2b4e5ef4e64d9d02e786f5ee65312228c5eb2
3
+ size 3643
checkpoint-2000/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-base",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50257
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "d_model": 512,
15
+ "decoder_attention_heads": 8,
16
+ "decoder_ffn_dim": 2048,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 8,
22
+ "encoder_ffn_dim": 2048,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 6,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": [
27
+ [
28
+ 1,
29
+ 50266
30
+ ],
31
+ [
32
+ 2,
33
+ 50359
34
+ ],
35
+ [
36
+ 3,
37
+ 50363
38
+ ]
39
+ ],
40
+ "init_std": 0.02,
41
+ "is_encoder_decoder": true,
42
+ "max_length": 448,
43
+ "max_source_positions": 1500,
44
+ "max_target_positions": 448,
45
+ "model_type": "whisper",
46
+ "num_hidden_layers": 6,
47
+ "num_mel_bins": 80,
48
+ "pad_token_id": 50257,
49
+ "scale_embedding": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.26.0.dev0",
52
+ "use_cache": false,
53
+ "vocab_size": 51865
54
+ }
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70049a7047fe98285c7f12ad76b8493e13ac3b61a0aafaccd8b8043b3501ebd4
3
+ size 580895237
checkpoint-2000/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc104e38d151120f224615af7624d6c89f5220ec0f2e22ce887c55fe66b2091f
3
+ size 290458721
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93ac781da7c59e886748e649f8813a700f5c11a2bbf8688b1997b0ec93fa3817
3
+ size 14575
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93dcbb27405421069b5f31003e3f224646e375ff4c9d7ba4b0d265614f660d0f
3
+ size 557
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16736fdc76760699923db14f70321b0e15ba9013425b6a35e4c7a5d590ef5c0f
3
+ size 627
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 7.0033,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.02,
12
+ "learning_rate": 9.960000000000001e-06,
13
+ "loss": 0.5615,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.02,
18
+ "learning_rate": 1e-05,
19
+ "loss": 0.3273,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 3.02,
24
+ "eval_loss": 0.4225325584411621,
25
+ "eval_runtime": 1489.9228,
26
+ "eval_samples_per_second": 3.09,
27
+ "eval_steps_per_second": 0.193,
28
+ "eval_wer": 20.82533081951615,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 5.01,
33
+ "learning_rate": 1e-05,
34
+ "loss": 0.1859,
35
+ "step": 1500
36
+ },
37
+ {
38
+ "epoch": 7.0,
39
+ "learning_rate": 1e-05,
40
+ "loss": 0.0923,
41
+ "step": 2000
42
+ },
43
+ {
44
+ "epoch": 7.0,
45
+ "eval_loss": 0.46434369683265686,
46
+ "eval_runtime": 1504.5618,
47
+ "eval_samples_per_second": 3.06,
48
+ "eval_steps_per_second": 0.191,
49
+ "eval_wer": 21.22002361108664,
50
+ "step": 2000
51
+ }
52
+ ],
53
+ "max_steps": 10000,
54
+ "num_train_epochs": 9223372036854775807,
55
+ "total_flos": 4.14105131483136e+18,
56
+ "trial_name": null,
57
+ "trial_params": null
58
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63250d2e981e0cfad1a5508019d2b4e5ef4e64d9d02e786f5ee65312228c5eb2
3
+ size 3643
checkpoint-3000/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-base",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50257
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "d_model": 512,
15
+ "decoder_attention_heads": 8,
16
+ "decoder_ffn_dim": 2048,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 8,
22
+ "encoder_ffn_dim": 2048,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 6,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": [
27
+ [
28
+ 1,
29
+ 50266
30
+ ],
31
+ [
32
+ 2,
33
+ 50359
34
+ ],
35
+ [
36
+ 3,
37
+ 50363
38
+ ]
39
+ ],
40
+ "init_std": 0.02,
41
+ "is_encoder_decoder": true,
42
+ "max_length": 448,
43
+ "max_source_positions": 1500,
44
+ "max_target_positions": 448,
45
+ "model_type": "whisper",
46
+ "num_hidden_layers": 6,
47
+ "num_mel_bins": 80,
48
+ "pad_token_id": 50257,
49
+ "scale_embedding": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.26.0.dev0",
52
+ "use_cache": false,
53
+ "vocab_size": 51865
54
+ }
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed8f87e446571f3c8e9bbbaf08eb695b8ab1032ff5ed7fe2450051dec24f01c1
3
+ size 580895237
checkpoint-3000/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be375a567856da345d07fddafbc867a1e8a9b3222b1c2aa85cabc598641d2c46
3
+ size 290458721
checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffa72e9bdabc679f172d53cc335a354dd99e8511cac8305c188dfc63d7155363
3
+ size 14575
checkpoint-3000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d2a59dd750539c973981813825488c88f87133aca172b473f2a8e03b35e2802
3
+ size 557
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e3cd6cdaa5da580e18c0e9ab9c2eb7a710bb6f37577950af7d770bb2b23fd69
3
+ size 627
checkpoint-3000/trainer_state.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.019,
5
+ "global_step": 3000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.02,
12
+ "learning_rate": 9.960000000000001e-06,
13
+ "loss": 0.5615,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.02,
18
+ "learning_rate": 1e-05,
19
+ "loss": 0.3273,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 3.02,
24
+ "eval_loss": 0.4225325584411621,
25
+ "eval_runtime": 1489.9228,
26
+ "eval_samples_per_second": 3.09,
27
+ "eval_steps_per_second": 0.193,
28
+ "eval_wer": 20.82533081951615,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 5.01,
33
+ "learning_rate": 1e-05,
34
+ "loss": 0.1859,
35
+ "step": 1500
36
+ },
37
+ {
38
+ "epoch": 7.0,
39
+ "learning_rate": 1e-05,
40
+ "loss": 0.0923,
41
+ "step": 2000
42
+ },
43
+ {
44
+ "epoch": 7.0,
45
+ "eval_loss": 0.46434369683265686,
46
+ "eval_runtime": 1504.5618,
47
+ "eval_samples_per_second": 3.06,
48
+ "eval_steps_per_second": 0.191,
49
+ "eval_wer": 21.22002361108664,
50
+ "step": 2000
51
+ },
52
+ {
53
+ "epoch": 8.03,
54
+ "learning_rate": 1e-05,
55
+ "loss": 0.0369,
56
+ "step": 2500
57
+ },
58
+ {
59
+ "epoch": 10.02,
60
+ "learning_rate": 1e-05,
61
+ "loss": 0.0164,
62
+ "step": 3000
63
+ },
64
+ {
65
+ "epoch": 10.02,
66
+ "eval_loss": 0.5403426885604858,
67
+ "eval_runtime": 1504.7378,
68
+ "eval_samples_per_second": 3.06,
69
+ "eval_steps_per_second": 0.191,
70
+ "eval_wer": 22.962662766725987,
71
+ "step": 3000
72
+ }
73
+ ],
74
+ "max_steps": 10000,
75
+ "num_train_epochs": 9223372036854775807,
76
+ "total_flos": 6.2122904322048e+18,
77
+ "trial_name": null,
78
+ "trial_params": null
79
+ }
checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63250d2e981e0cfad1a5508019d2b4e5ef4e64d9d02e786f5ee65312228c5eb2
3
+ size 3643
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc104e38d151120f224615af7624d6c89f5220ec0f2e22ce887c55fe66b2091f
3
  size 290458721
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be375a567856da345d07fddafbc867a1e8a9b3222b1c2aa85cabc598641d2c46
3
  size 290458721
runs/Dec27_09-55-14_srv/events.out.tfevents.1672134971.srv.1172111.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:17cc6a0bee6d09ad1b194842563140a0c4fd9ac3cbc9c69ac8603720e542cbc4
3
- size 5692
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47abf8eeb54dda019a809756fc2c10eb9dc73ef8ae80dfef528cba96123d3874
3
+ size 6324