Hugging Face's logo

GENG
/
hubert_5000_checkpoint_ls

Automatic Speech Recognition
Transformers
hubert
Inference Endpoints
Model card Files Community
GENG commited on
Commit
586f8bf
•
1 Parent(s): da1db52

Upload trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +181 -0
trainer_state.json ADDED
@@ -0,0 +1,181 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 38.46153846153846,
5
+ "global_step": 5500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 3.5,
12
+ "learning_rate": 4.9800000000000004e-05,
13
+ "loss": 4.6717,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.5,
18
+ "eval_cer": 0.9966954216782684,
19
+ "eval_loss": 3.3382508754730225,
20
+ "eval_runtime": 157.3429,
21
+ "eval_samples_per_second": 7.258,
22
+ "eval_steps_per_second": 0.909,
23
+ "step": 500
24
+ },
25
+ {
26
+ "epoch": 6.99,
27
+ "learning_rate": 9.98e-05,
28
+ "loss": 3.1712,
29
+ "step": 1000
30
+ },
31
+ {
32
+ "epoch": 6.99,
33
+ "eval_cer": 0.9966954216782684,
34
+ "eval_loss": 3.281019449234009,
35
+ "eval_runtime": 163.1426,
36
+ "eval_samples_per_second": 7.0,
37
+ "eval_steps_per_second": 0.877,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 10.49,
42
+ "learning_rate": 9.190243902439025e-05,
43
+ "loss": 3.1588,
44
+ "step": 1500
45
+ },
46
+ {
47
+ "epoch": 10.49,
48
+ "eval_cer": 0.9966954216782684,
49
+ "eval_loss": 3.2297120094299316,
50
+ "eval_runtime": 161.5142,
51
+ "eval_samples_per_second": 7.071,
52
+ "eval_steps_per_second": 0.885,
53
+ "step": 1500
54
+ },
55
+ {
56
+ "epoch": 13.99,
57
+ "learning_rate": 8.377235772357724e-05,
58
+ "loss": 2.9826,
59
+ "step": 2000
60
+ },
61
+ {
62
+ "epoch": 13.99,
63
+ "eval_cer": 0.2508371402935709,
64
+ "eval_loss": 1.0106703042984009,
65
+ "eval_runtime": 160.9136,
66
+ "eval_samples_per_second": 7.097,
67
+ "eval_steps_per_second": 0.889,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 17.48,
72
+ "learning_rate": 7.564227642276424e-05,
73
+ "loss": 0.576,
74
+ "step": 2500
75
+ },
76
+ {
77
+ "epoch": 17.48,
78
+ "eval_cer": 0.048671424633916056,
79
+ "eval_loss": 0.23353584110736847,
80
+ "eval_runtime": 161.7107,
81
+ "eval_samples_per_second": 7.062,
82
+ "eval_steps_per_second": 0.884,
83
+ "step": 2500
84
+ },
85
+ {
86
+ "epoch": 20.98,
87
+ "learning_rate": 6.751219512195123e-05,
88
+ "loss": 0.251,
89
+ "step": 3000
90
+ },
91
+ {
92
+ "epoch": 20.98,
93
+ "eval_cer": 0.031453603954938104,
94
+ "eval_loss": 0.15631325542926788,
95
+ "eval_runtime": 162.3596,
96
+ "eval_samples_per_second": 7.034,
97
+ "eval_steps_per_second": 0.881,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 24.48,
102
+ "learning_rate": 5.938211382113822e-05,
103
+ "loss": 0.1958,
104
+ "step": 3500
105
+ },
106
+ {
107
+ "epoch": 24.48,
108
+ "eval_cer": 0.026911641381413432,
109
+ "eval_loss": 0.14399459958076477,
110
+ "eval_runtime": 161.0224,
111
+ "eval_samples_per_second": 7.092,
112
+ "eval_steps_per_second": 0.888,
113
+ "step": 3500
114
+ },
115
+ {
116
+ "epoch": 27.97,
117
+ "learning_rate": 5.125203252032521e-05,
118
+ "loss": 0.1513,
119
+ "step": 4000
120
+ },
121
+ {
122
+ "epoch": 27.97,
123
+ "eval_cer": 0.024079145677072034,
124
+ "eval_loss": 0.13698184490203857,
125
+ "eval_runtime": 157.5139,
126
+ "eval_samples_per_second": 7.25,
127
+ "eval_steps_per_second": 0.908,
128
+ "step": 4000
129
+ },
130
+ {
131
+ "epoch": 31.47,
132
+ "learning_rate": 4.31219512195122e-05,
133
+ "loss": 0.1178,
134
+ "step": 4500
135
+ },
136
+ {
137
+ "epoch": 31.47,
138
+ "eval_cer": 0.022228933679722732,
139
+ "eval_loss": 0.13266856968402863,
140
+ "eval_runtime": 158.1185,
141
+ "eval_samples_per_second": 7.222,
142
+ "eval_steps_per_second": 0.904,
143
+ "step": 4500
144
+ },
145
+ {
146
+ "epoch": 34.97,
147
+ "learning_rate": 3.4991869918699187e-05,
148
+ "loss": 0.1008,
149
+ "step": 5000
150
+ },
151
+ {
152
+ "epoch": 34.97,
153
+ "eval_cer": 0.02185068114777652,
154
+ "eval_loss": 0.134142205119133,
155
+ "eval_runtime": 161.0792,
156
+ "eval_samples_per_second": 7.09,
157
+ "eval_steps_per_second": 0.888,
158
+ "step": 5000
159
+ },
160
+ {
161
+ "epoch": 38.46,
162
+ "learning_rate": 2.686178861788618e-05,
163
+ "loss": 0.0933,
164
+ "step": 5500
165
+ },
166
+ {
167
+ "epoch": 38.46,
168
+ "eval_cer": 0.020815618017722158,
169
+ "eval_loss": 0.13403166830539703,
170
+ "eval_runtime": 159.8014,
171
+ "eval_samples_per_second": 7.146,
172
+ "eval_steps_per_second": 0.895,
173
+ "step": 5500
174
+ }
175
+ ],
176
+ "max_steps": 7150,
177
+ "num_train_epochs": 50,
178
+ "total_flos": 2.045078058974218e+19,
179
+ "trial_name": null,
180
+ "trial_params": null
181
+ }