Hugging Face's logo

Bronsn
/
whisper-tiny-luganda-final

Automatic Speech Recognition
Transformers
Safetensors
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
Bronsn commited on
Commit
3929501
·
verified ·
1 Parent(s): 2873c50

Upload checkpoints/checkpoint-1500/trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. checkpoints/checkpoint-1500/trainer_state.json +165 -0
checkpoints/checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,165 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7198385688576876,
3
+ "best_model_checkpoint": "whisper-tiny-luganda-v2/checkpoint-1000",
4
+ "epoch": 0.3262997607135088,
5
+ "eval_steps": 500,
6
+ "global_step": 1500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.021753317380900587,
13
+ "grad_norm": 47.60859680175781,
14
+ "learning_rate": 4.5238095238095235e-06,
15
+ "loss": 9.6365,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.04350663476180117,
20
+ "grad_norm": 41.293460845947266,
21
+ "learning_rate": 9.285714285714286e-06,
22
+ "loss": 4.282,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.06525995214270176,
27
+ "grad_norm": 29.167564392089844,
28
+ "learning_rate": 1.4047619047619048e-05,
29
+ "loss": 2.8974,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.08701326952360235,
34
+ "grad_norm": 30.911415100097656,
35
+ "learning_rate": 1.880952380952381e-05,
36
+ "loss": 2.2716,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.10876658690450294,
41
+ "grad_norm": 25.824464797973633,
42
+ "learning_rate": 2.357142857142857e-05,
43
+ "loss": 2.0498,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 0.10876658690450294,
48
+ "eval_loss": 0.9947264194488525,
49
+ "eval_runtime": 2111.5938,
50
+ "eval_samples_per_second": 6.458,
51
+ "eval_steps_per_second": 1.615,
52
+ "eval_wer": 0.8350676193361011,
53
+ "step": 500
54
+ },
55
+ {
56
+ "epoch": 0.13051990428540353,
57
+ "grad_norm": 22.513357162475586,
58
+ "learning_rate": 2.8333333333333335e-05,
59
+ "loss": 1.8357,
60
+ "step": 600
61
+ },
62
+ {
63
+ "epoch": 0.1522732216663041,
64
+ "grad_norm": 22.64017105102539,
65
+ "learning_rate": 3.309523809523809e-05,
66
+ "loss": 1.7019,
67
+ "step": 700
68
+ },
69
+ {
70
+ "epoch": 0.1740265390472047,
71
+ "grad_norm": 22.659271240234375,
72
+ "learning_rate": 3.785714285714286e-05,
73
+ "loss": 1.6458,
74
+ "step": 800
75
+ },
76
+ {
77
+ "epoch": 0.19577985642810528,
78
+ "grad_norm": 19.53955841064453,
79
+ "learning_rate": 4.261904761904762e-05,
80
+ "loss": 1.6001,
81
+ "step": 900
82
+ },
83
+ {
84
+ "epoch": 0.2175331738090059,
85
+ "grad_norm": 21.89055061340332,
86
+ "learning_rate": 4.738095238095238e-05,
87
+ "loss": 1.5134,
88
+ "step": 1000
89
+ },
90
+ {
91
+ "epoch": 0.2175331738090059,
92
+ "eval_loss": 0.7854410409927368,
93
+ "eval_runtime": 2131.7562,
94
+ "eval_samples_per_second": 6.397,
95
+ "eval_steps_per_second": 1.6,
96
+ "eval_wer": 0.7198385688576876,
97
+ "step": 1000
98
+ },
99
+ {
100
+ "epoch": 0.23928649118990647,
101
+ "grad_norm": 24.862058639526367,
102
+ "learning_rate": 4.988721804511278e-05,
103
+ "loss": 1.421,
104
+ "step": 1100
105
+ },
106
+ {
107
+ "epoch": 0.26103980857080705,
108
+ "grad_norm": 18.665178298950195,
109
+ "learning_rate": 4.963659147869674e-05,
110
+ "loss": 1.4254,
111
+ "step": 1200
112
+ },
113
+ {
114
+ "epoch": 0.28279312595170764,
115
+ "grad_norm": 16.429445266723633,
116
+ "learning_rate": 4.93859649122807e-05,
117
+ "loss": 1.3466,
118
+ "step": 1300
119
+ },
120
+ {
121
+ "epoch": 0.3045464433326082,
122
+ "grad_norm": 24.4464054107666,
123
+ "learning_rate": 4.913533834586466e-05,
124
+ "loss": 1.3342,
125
+ "step": 1400
126
+ },
127
+ {
128
+ "epoch": 0.3262997607135088,
129
+ "grad_norm": 14.964367866516113,
130
+ "learning_rate": 4.888471177944862e-05,
131
+ "loss": 1.2494,
132
+ "step": 1500
133
+ },
134
+ {
135
+ "epoch": 0.3262997607135088,
136
+ "eval_loss": 0.6914966106414795,
137
+ "eval_runtime": 2154.6429,
138
+ "eval_samples_per_second": 6.329,
139
+ "eval_steps_per_second": 1.583,
140
+ "eval_wer": 0.7817383247509934,
141
+ "step": 1500
142
+ }
143
+ ],
144
+ "logging_steps": 100,
145
+ "max_steps": 21000,
146
+ "num_input_tokens_seen": 0,
147
+ "num_train_epochs": 5,
148
+ "save_steps": 500,
149
+ "stateful_callbacks": {
150
+ "TrainerControl": {
151
+ "args": {
152
+ "should_epoch_stop": false,
153
+ "should_evaluate": false,
154
+ "should_log": false,
155
+ "should_save": true,
156
+ "should_training_stop": false
157
+ },
158
+ "attributes": {}
159
+ }
160
+ },
161
+ "total_flos": 5.9085324288e+17,
162
+ "train_batch_size": 8,
163
+ "trial_name": null,
164
+ "trial_params": null
165
+ }