Hugging Face's logo

EYEDOL
/
ZIN_N_V2

TensorBoard
Safetensors
whisper
Model card Files
xet
Metrics Community
EYEDOL commited on
Commit
5dcf80e
·
verified ·
1 Parent(s): 85f8a54

Upload folder using huggingface_hub

Browse files
Files changed (21) hide show
  1. checkpoint-1500/config.json +47 -0
  2. checkpoint-1500/generation_config.json +9 -0
  3. checkpoint-1500/model.safetensors +3 -0
  4. checkpoint-1500/optimizer.pt +3 -0
  5. checkpoint-1500/preprocessor_config.json +15 -0
  6. checkpoint-1500/rng_state.pth +3 -0
  7. checkpoint-1500/scaler.pt +3 -0
  8. checkpoint-1500/scheduler.pt +3 -0
  9. checkpoint-1500/trainer_state.json +110 -0
  10. checkpoint-1500/training_args.bin +3 -0
  11. checkpoint-1746/config.json +47 -0
  12. checkpoint-1746/generation_config.json +9 -0
  13. checkpoint-1746/model.safetensors +3 -0
  14. checkpoint-1746/optimizer.pt +3 -0
  15. checkpoint-1746/preprocessor_config.json +15 -0
  16. checkpoint-1746/rng_state.pth +3 -0
  17. checkpoint-1746/scaler.pt +3 -0
  18. checkpoint-1746/scheduler.pt +3 -0
  19. checkpoint-1746/trainer_state.json +117 -0
  20. checkpoint-1746/training_args.bin +3 -0
  21. runs/Aug01_10-03-43_53b9af734281/events.out.tfevents.1754042625.53b9af734281.20.0 +2 -2
checkpoint-1500/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "gelu",
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": null,
10
+ "bos_token_id": 50257,
11
+ "classifier_proj_size": 256,
12
+ "d_model": 768,
13
+ "decoder_attention_heads": 12,
14
+ "decoder_ffn_dim": 3072,
15
+ "decoder_layerdrop": 0.0,
16
+ "decoder_layers": 12,
17
+ "decoder_start_token_id": 50258,
18
+ "dropout": 0.0,
19
+ "encoder_attention_heads": 12,
20
+ "encoder_ffn_dim": 3072,
21
+ "encoder_layerdrop": 0.0,
22
+ "encoder_layers": 12,
23
+ "eos_token_id": 50257,
24
+ "forced_decoder_ids": null,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "mask_feature_length": 10,
28
+ "mask_feature_min_masks": 0,
29
+ "mask_feature_prob": 0.0,
30
+ "mask_time_length": 10,
31
+ "mask_time_min_masks": 2,
32
+ "mask_time_prob": 0.05,
33
+ "max_length": null,
34
+ "max_source_positions": 1500,
35
+ "max_target_positions": 448,
36
+ "median_filter_width": 7,
37
+ "model_type": "whisper",
38
+ "num_hidden_layers": 12,
39
+ "num_mel_bins": 80,
40
+ "pad_token_id": 50257,
41
+ "scale_embedding": false,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.52.4",
44
+ "use_cache": false,
45
+ "use_weighted_layer_sum": false,
46
+ "vocab_size": 51865
47
+ }
checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 50257,
3
+ "decoder_start_token_id": 50258,
4
+ "eos_token_id": 50257,
5
+ "pad_token_id": 50257,
6
+ "suppress_tokens": [],
7
+ "transformers_version": "4.52.4",
8
+ "use_cache": false
9
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a54e31456f61b6292bfb0942d95687dbd370c48685c90f360bc9897c2269e6f5
3
+ size 966995080
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcec7e4c4d2c7faa43333ce7e79a2cfffa340009154f9409f49141d4151be8a4
3
+ size 1925064044
checkpoint-1500/preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 80,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "right",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8719644d219fcadadc5aa5fd3b45dde0693f9c8b548056ca969d80d81c56e480
3
+ size 14244
checkpoint-1500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e542f0d1ed4ff8c1a75f8edda5a50cf49629c17901da6f9303bc5ac8650e1a4f
3
+ size 988
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56eedb4382f14e0dbb748943a6749fc9322bec46ad2828e0b82a37e90570a916
3
+ size 1064
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1500,
3
+ "best_metric": 29.051360086819074,
4
+ "best_model_checkpoint": "./ZINASR/checkpoint-1500",
5
+ "epoch": 2.5778159931212383,
6
+ "eval_steps": 500,
7
+ "global_step": 1500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.34393809114359414,
14
+ "grad_norm": 3.4750354290008545,
15
+ "learning_rate": 9.9e-06,
16
+ "loss": 0.1655,
17
+ "step": 200
18
+ },
19
+ {
20
+ "epoch": 0.6878761822871883,
21
+ "grad_norm": 3.458239793777466,
22
+ "learning_rate": 8.719275549805952e-06,
23
+ "loss": 0.1352,
24
+ "step": 400
25
+ },
26
+ {
27
+ "epoch": 0.8598452278589854,
28
+ "eval_loss": 0.43551918864250183,
29
+ "eval_runtime": 3356.5685,
30
+ "eval_samples_per_second": 1.385,
31
+ "eval_steps_per_second": 0.173,
32
+ "eval_wer": 30.011560148158633,
33
+ "step": 500
34
+ },
35
+ {
36
+ "epoch": 1.0309544282029235,
37
+ "grad_norm": 2.6634328365325928,
38
+ "learning_rate": 7.425614489003882e-06,
39
+ "loss": 0.135,
40
+ "step": 600
41
+ },
42
+ {
43
+ "epoch": 1.3748925193465176,
44
+ "grad_norm": 3.2597954273223877,
45
+ "learning_rate": 6.1319534282018115e-06,
46
+ "loss": 0.0783,
47
+ "step": 800
48
+ },
49
+ {
50
+ "epoch": 1.7188306104901119,
51
+ "grad_norm": 3.3078627586364746,
52
+ "learning_rate": 4.838292367399742e-06,
53
+ "loss": 0.0944,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 1.7188306104901119,
58
+ "eval_loss": 0.43551915884017944,
59
+ "eval_runtime": 3363.7968,
60
+ "eval_samples_per_second": 1.382,
61
+ "eval_steps_per_second": 0.173,
62
+ "eval_wer": 29.334465756010097,
63
+ "step": 1000
64
+ },
65
+ {
66
+ "epoch": 2.061908856405847,
67
+ "grad_norm": 3.6673755645751953,
68
+ "learning_rate": 3.5446313065976717e-06,
69
+ "loss": 0.1003,
70
+ "step": 1200
71
+ },
72
+ {
73
+ "epoch": 2.405846947549441,
74
+ "grad_norm": 3.501398801803589,
75
+ "learning_rate": 2.2509702457956016e-06,
76
+ "loss": 0.0474,
77
+ "step": 1400
78
+ },
79
+ {
80
+ "epoch": 2.5778159931212383,
81
+ "eval_loss": 0.45564955472946167,
82
+ "eval_runtime": 3360.7459,
83
+ "eval_samples_per_second": 1.384,
84
+ "eval_steps_per_second": 0.173,
85
+ "eval_wer": 29.051360086819074,
86
+ "step": 1500
87
+ }
88
+ ],
89
+ "logging_steps": 200,
90
+ "max_steps": 1746,
91
+ "num_input_tokens_seen": 0,
92
+ "num_train_epochs": 3,
93
+ "save_steps": 1500,
94
+ "stateful_callbacks": {
95
+ "TrainerControl": {
96
+ "args": {
97
+ "should_epoch_stop": false,
98
+ "should_evaluate": false,
99
+ "should_log": false,
100
+ "should_save": true,
101
+ "should_training_stop": false
102
+ },
103
+ "attributes": {}
104
+ }
105
+ },
106
+ "total_flos": 1.383651560374272e+19,
107
+ "train_batch_size": 16,
108
+ "trial_name": null,
109
+ "trial_params": null
110
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df3d122a7d105e8decffe56aec61082b73d896eb022d98ef2e424e6cc2d226ec
3
+ size 5432
checkpoint-1746/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "gelu",
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": null,
10
+ "bos_token_id": 50257,
11
+ "classifier_proj_size": 256,
12
+ "d_model": 768,
13
+ "decoder_attention_heads": 12,
14
+ "decoder_ffn_dim": 3072,
15
+ "decoder_layerdrop": 0.0,
16
+ "decoder_layers": 12,
17
+ "decoder_start_token_id": 50258,
18
+ "dropout": 0.0,
19
+ "encoder_attention_heads": 12,
20
+ "encoder_ffn_dim": 3072,
21
+ "encoder_layerdrop": 0.0,
22
+ "encoder_layers": 12,
23
+ "eos_token_id": 50257,
24
+ "forced_decoder_ids": null,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "mask_feature_length": 10,
28
+ "mask_feature_min_masks": 0,
29
+ "mask_feature_prob": 0.0,
30
+ "mask_time_length": 10,
31
+ "mask_time_min_masks": 2,
32
+ "mask_time_prob": 0.05,
33
+ "max_length": null,
34
+ "max_source_positions": 1500,
35
+ "max_target_positions": 448,
36
+ "median_filter_width": 7,
37
+ "model_type": "whisper",
38
+ "num_hidden_layers": 12,
39
+ "num_mel_bins": 80,
40
+ "pad_token_id": 50257,
41
+ "scale_embedding": false,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.52.4",
44
+ "use_cache": false,
45
+ "use_weighted_layer_sum": false,
46
+ "vocab_size": 51865
47
+ }
checkpoint-1746/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 50257,
3
+ "decoder_start_token_id": 50258,
4
+ "eos_token_id": 50257,
5
+ "pad_token_id": 50257,
6
+ "suppress_tokens": [],
7
+ "transformers_version": "4.52.4",
8
+ "use_cache": false
9
+ }
checkpoint-1746/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b2e6f42b5b341f2131633fe0eb841a74d0cba5e287273600ff4c118db45061f
3
+ size 966995080
checkpoint-1746/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c548463381aea18b595e6842de9411ce1502f8abb7f78f2f454e7003034afe72
3
+ size 1925064044
checkpoint-1746/preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 80,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "right",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
checkpoint-1746/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd89a1dbc0daeb96a71b180dc57f7d45972f67de1dbdc3b3edafe77ceb2230f4
3
+ size 14244
checkpoint-1746/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cac11f18e1efb9c17a7a84027418f195d0f1f5b7f5dd33c3b664f8927bc78cd
3
+ size 988
checkpoint-1746/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f074e0561b5a3e2acbe6a999f48f2ce761e5f877bed9efc523cc8df77741de15
3
+ size 1064
checkpoint-1746/trainer_state.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1500,
3
+ "best_metric": 29.051360086819074,
4
+ "best_model_checkpoint": "./ZINASR/checkpoint-1500",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1746,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.34393809114359414,
14
+ "grad_norm": 3.4750354290008545,
15
+ "learning_rate": 9.9e-06,
16
+ "loss": 0.1655,
17
+ "step": 200
18
+ },
19
+ {
20
+ "epoch": 0.6878761822871883,
21
+ "grad_norm": 3.458239793777466,
22
+ "learning_rate": 8.719275549805952e-06,
23
+ "loss": 0.1352,
24
+ "step": 400
25
+ },
26
+ {
27
+ "epoch": 0.8598452278589854,
28
+ "eval_loss": 0.43551918864250183,
29
+ "eval_runtime": 3356.5685,
30
+ "eval_samples_per_second": 1.385,
31
+ "eval_steps_per_second": 0.173,
32
+ "eval_wer": 30.011560148158633,
33
+ "step": 500
34
+ },
35
+ {
36
+ "epoch": 1.0309544282029235,
37
+ "grad_norm": 2.6634328365325928,
38
+ "learning_rate": 7.425614489003882e-06,
39
+ "loss": 0.135,
40
+ "step": 600
41
+ },
42
+ {
43
+ "epoch": 1.3748925193465176,
44
+ "grad_norm": 3.2597954273223877,
45
+ "learning_rate": 6.1319534282018115e-06,
46
+ "loss": 0.0783,
47
+ "step": 800
48
+ },
49
+ {
50
+ "epoch": 1.7188306104901119,
51
+ "grad_norm": 3.3078627586364746,
52
+ "learning_rate": 4.838292367399742e-06,
53
+ "loss": 0.0944,
54
+ "step": 1000
55
+ },
56
+ {
57
+ "epoch": 1.7188306104901119,
58
+ "eval_loss": 0.43551915884017944,
59
+ "eval_runtime": 3363.7968,
60
+ "eval_samples_per_second": 1.382,
61
+ "eval_steps_per_second": 0.173,
62
+ "eval_wer": 29.334465756010097,
63
+ "step": 1000
64
+ },
65
+ {
66
+ "epoch": 2.061908856405847,
67
+ "grad_norm": 3.6673755645751953,
68
+ "learning_rate": 3.5446313065976717e-06,
69
+ "loss": 0.1003,
70
+ "step": 1200
71
+ },
72
+ {
73
+ "epoch": 2.405846947549441,
74
+ "grad_norm": 3.501398801803589,
75
+ "learning_rate": 2.2509702457956016e-06,
76
+ "loss": 0.0474,
77
+ "step": 1400
78
+ },
79
+ {
80
+ "epoch": 2.5778159931212383,
81
+ "eval_loss": 0.45564955472946167,
82
+ "eval_runtime": 3360.7459,
83
+ "eval_samples_per_second": 1.384,
84
+ "eval_steps_per_second": 0.173,
85
+ "eval_wer": 29.051360086819074,
86
+ "step": 1500
87
+ },
88
+ {
89
+ "epoch": 2.749785038693035,
90
+ "grad_norm": 2.335383892059326,
91
+ "learning_rate": 9.573091849935317e-07,
92
+ "loss": 0.0456,
93
+ "step": 1600
94
+ }
95
+ ],
96
+ "logging_steps": 200,
97
+ "max_steps": 1746,
98
+ "num_input_tokens_seen": 0,
99
+ "num_train_epochs": 3,
100
+ "save_steps": 1500,
101
+ "stateful_callbacks": {
102
+ "TrainerControl": {
103
+ "args": {
104
+ "should_epoch_stop": false,
105
+ "should_evaluate": false,
106
+ "should_log": false,
107
+ "should_save": true,
108
+ "should_training_stop": true
109
+ },
110
+ "attributes": {}
111
+ }
112
+ },
113
+ "total_flos": 1.610046806925312e+19,
114
+ "train_batch_size": 16,
115
+ "trial_name": null,
116
+ "trial_params": null
117
+ }
checkpoint-1746/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df3d122a7d105e8decffe56aec61082b73d896eb022d98ef2e424e6cc2d226ec
3
+ size 5432
runs/Aug01_10-03-43_53b9af734281/events.out.tfevents.1754042625.53b9af734281.20.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8237bbcbff588229db6bb4408a038b8ec3e9d2cdebc8c2b4d4d470eb098d501c
3
- size 8326
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8678fe6f7f033834bede4266899bdb443fc1cecdd5a5aed80cf7a74567ac74c1
3
+ size 8680