Hugging Face's logo

SamagraDataGov
/
whisper-tiny-hindi2_test

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
whisper
Generated from Trainer
Model card Files
xet
Metrics Community
1
SamagraDataGov commited on
Commit
18adcb6
·
verified ·
1 Parent(s): fb2d63d

Training in progress, step 40

Browse files
Files changed (3) hide show
  1. pytorch_model.bin +1 -1
  2. runs/Jun29_09-39-32_bharatsahaiyak-test/events.out.tfevents.1719655172.bharatsahaiyak-test.1854617.1 +3 -0
  3. trainer_state.json +157 -0
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e0e9a8a6a2405f1ef747c57eb67d6430089edf114f56b97ad0d448acd182f8f2
3
  size 151099494
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5521753c8d6998f9c0d7662cb2d4fd33116986428657e8562519057a6cb32e44
3
  size 151099494
runs/Jun29_09-39-32_bharatsahaiyak-test/events.out.tfevents.1719655172.bharatsahaiyak-test.1854617.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d1875581f99b3834dc2d0568acefc7f292cddf1fd069d2b64dee4724d00dcb4
3
+ size 6396
trainer_state.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 50.0,
3
+ "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-120",
4
+ "epoch": 6.349206349206349,
5
+ "eval_steps": 40,
6
+ "global_step": 200,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.6349206349206349,
13
+ "grad_norm": 8.679319381713867,
14
+ "learning_rate": 3.75e-05,
15
+ "loss": 1.6908,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 1.2698412698412698,
20
+ "grad_norm": 7.228819370269775,
21
+ "learning_rate": 3.75e-05,
22
+ "loss": 0.6584,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 1.2698412698412698,
27
+ "eval_loss": 0.593582272529602,
28
+ "eval_runtime": 54.5955,
29
+ "eval_samples_per_second": 1.832,
30
+ "eval_steps_per_second": 1.832,
31
+ "eval_wer": 65.00843170320405,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 1.9047619047619047,
36
+ "grad_norm": 6.050016403198242,
37
+ "learning_rate": 3.75e-05,
38
+ "loss": 0.4786,
39
+ "step": 60
40
+ },
41
+ {
42
+ "epoch": 2.5396825396825395,
43
+ "grad_norm": 5.063595294952393,
44
+ "learning_rate": 3.75e-05,
45
+ "loss": 0.3575,
46
+ "step": 80
47
+ },
48
+ {
49
+ "epoch": 2.5396825396825395,
50
+ "eval_loss": 0.46842461824417114,
51
+ "eval_runtime": 55.6599,
52
+ "eval_samples_per_second": 1.797,
53
+ "eval_steps_per_second": 1.797,
54
+ "eval_wer": 55.31197301854974,
55
+ "step": 80
56
+ },
57
+ {
58
+ "epoch": 3.1746031746031744,
59
+ "grad_norm": 4.799503326416016,
60
+ "learning_rate": 3.75e-05,
61
+ "loss": 0.2994,
62
+ "step": 100
63
+ },
64
+ {
65
+ "epoch": 3.8095238095238093,
66
+ "grad_norm": 5.387985706329346,
67
+ "learning_rate": 3.75e-05,
68
+ "loss": 0.2254,
69
+ "step": 120
70
+ },
71
+ {
72
+ "epoch": 3.8095238095238093,
73
+ "eval_loss": 0.4344378709793091,
74
+ "eval_runtime": 54.7109,
75
+ "eval_samples_per_second": 1.828,
76
+ "eval_steps_per_second": 1.828,
77
+ "eval_wer": 50.0,
78
+ "step": 120
79
+ },
80
+ {
81
+ "epoch": 4.444444444444445,
82
+ "grad_norm": 4.451031684875488,
83
+ "learning_rate": 3.75e-05,
84
+ "loss": 0.1752,
85
+ "step": 140
86
+ },
87
+ {
88
+ "epoch": 5.079365079365079,
89
+ "grad_norm": 4.960638046264648,
90
+ "learning_rate": 3.75e-05,
91
+ "loss": 0.1486,
92
+ "step": 160
93
+ },
94
+ {
95
+ "epoch": 5.079365079365079,
96
+ "eval_loss": 0.4561767578125,
97
+ "eval_runtime": 54.4171,
98
+ "eval_samples_per_second": 1.838,
99
+ "eval_steps_per_second": 1.838,
100
+ "eval_wer": 52.1079258010118,
101
+ "step": 160
102
+ },
103
+ {
104
+ "epoch": 5.714285714285714,
105
+ "grad_norm": 3.8416409492492676,
106
+ "learning_rate": 3.75e-05,
107
+ "loss": 0.1021,
108
+ "step": 180
109
+ },
110
+ {
111
+ "epoch": 6.349206349206349,
112
+ "grad_norm": 3.137124538421631,
113
+ "learning_rate": 3.75e-05,
114
+ "loss": 0.0868,
115
+ "step": 200
116
+ },
117
+ {
118
+ "epoch": 6.349206349206349,
119
+ "eval_loss": 0.46873587369918823,
120
+ "eval_runtime": 54.2397,
121
+ "eval_samples_per_second": 1.844,
122
+ "eval_steps_per_second": 1.844,
123
+ "eval_wer": 50.5902192242833,
124
+ "step": 200
125
+ },
126
+ {
127
+ "epoch": 6.349206349206349,
128
+ "step": 200,
129
+ "total_flos": 1.5637915828224e+17,
130
+ "train_loss": 0.4222871881723404,
131
+ "train_runtime": 1142.3513,
132
+ "train_samples_per_second": 5.602,
133
+ "train_steps_per_second": 0.175
134
+ }
135
+ ],
136
+ "logging_steps": 20,
137
+ "max_steps": 200,
138
+ "num_input_tokens_seen": 0,
139
+ "num_train_epochs": 7,
140
+ "save_steps": 40,
141
+ "stateful_callbacks": {
142
+ "TrainerControl": {
143
+ "args": {
144
+ "should_epoch_stop": false,
145
+ "should_evaluate": false,
146
+ "should_log": false,
147
+ "should_save": true,
148
+ "should_training_stop": true
149
+ },
150
+ "attributes": {}
151
+ }
152
+ },
153
+ "total_flos": 1.5637915828224e+17,
154
+ "train_batch_size": 16,
155
+ "trial_name": null,
156
+ "trial_params": null
157
+ }