Hugging Face's logo

LegolasTheElf
/
Wav2Vec2_xls_r_300m_hi_final

Automatic Speech Recognition
Transformers
PyTorch
Hindi
wav2vec2
Openslr Multilingual
mozilla-foundation/common_voice_7_0
Generated from Trainer
Inference Endpoints
Model card Files Community
LegolasTheElf commited on
Commit
478bbba
•
1 Parent(s): 8d3838d

End of training

Browse files
Files changed (3) hide show
  1. all_results.json +8 -0
  2. train_results.json +8 -0
  3. trainer_state.json +217 -0
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 8.0,
3
+ "total_flos": 3.347893330222721e+19,
4
+ "train_loss": 0.47035410618170714,
5
+ "train_runtime": 23681.5399,
6
+ "train_samples_per_second": 13.496,
7
+ "train_steps_per_second": 0.211
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 8.0,
3
+ "total_flos": 3.347893330222721e+19,
4
+ "train_loss": 0.47035410618170714,
5
+ "train_runtime": 23681.5399,
6
+ "train_samples_per_second": 13.496,
7
+ "train_steps_per_second": 0.211
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,217 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 7.999599519423308,
5
+ "global_step": 4992,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.64,
12
+ "learning_rate": 9.386753883892068e-05,
13
+ "loss": 0.9821,
14
+ "step": 400
15
+ },
16
+ {
17
+ "epoch": 0.64,
18
+ "eval_cer": 0.15726841629986146,
19
+ "eval_loss": 0.5059255957603455,
20
+ "eval_runtime": 297.9957,
21
+ "eval_samples_per_second": 12.896,
22
+ "eval_steps_per_second": 0.406,
23
+ "eval_wer": 0.47830450053447937,
24
+ "step": 400
25
+ },
26
+ {
27
+ "epoch": 1.28,
28
+ "learning_rate": 8.569092395748161e-05,
29
+ "loss": 0.6861,
30
+ "step": 800
31
+ },
32
+ {
33
+ "epoch": 1.28,
34
+ "eval_cer": 0.13556268941095262,
35
+ "eval_loss": 0.42013171315193176,
36
+ "eval_runtime": 295.6579,
37
+ "eval_samples_per_second": 12.998,
38
+ "eval_steps_per_second": 0.409,
39
+ "eval_wer": 0.4247474857653963,
40
+ "step": 800
41
+ },
42
+ {
43
+ "epoch": 1.92,
44
+ "learning_rate": 7.751430907604253e-05,
45
+ "loss": 0.585,
46
+ "step": 1200
47
+ },
48
+ {
49
+ "epoch": 1.92,
50
+ "eval_cer": 0.1209960711263244,
51
+ "eval_loss": 0.3797381818294525,
52
+ "eval_runtime": 295.6141,
53
+ "eval_samples_per_second": 13.0,
54
+ "eval_steps_per_second": 0.409,
55
+ "eval_wer": 0.3810947010187831,
56
+ "step": 1200
57
+ },
58
+ {
59
+ "epoch": 2.56,
60
+ "learning_rate": 6.933769419460344e-05,
61
+ "loss": 0.5193,
62
+ "step": 1600
63
+ },
64
+ {
65
+ "epoch": 2.56,
66
+ "eval_cer": 0.11522804960498509,
67
+ "eval_loss": 0.3577031195163727,
68
+ "eval_runtime": 295.3778,
69
+ "eval_samples_per_second": 13.01,
70
+ "eval_steps_per_second": 0.41,
71
+ "eval_wer": 0.36521302820742163,
72
+ "step": 1600
73
+ },
74
+ {
75
+ "epoch": 3.21,
76
+ "learning_rate": 6.116107931316435e-05,
77
+ "loss": 0.4583,
78
+ "step": 2000
79
+ },
80
+ {
81
+ "epoch": 3.21,
82
+ "eval_cer": 0.11111951624265405,
83
+ "eval_loss": 0.34224170446395874,
84
+ "eval_runtime": 295.5987,
85
+ "eval_samples_per_second": 13.001,
86
+ "eval_steps_per_second": 0.409,
87
+ "eval_wer": 0.35192739806714807,
88
+ "step": 2000
89
+ },
90
+ {
91
+ "epoch": 3.85,
92
+ "learning_rate": 5.298446443172527e-05,
93
+ "loss": 0.4282,
94
+ "step": 2400
95
+ },
96
+ {
97
+ "epoch": 3.85,
98
+ "eval_cer": 0.10712445215615264,
99
+ "eval_loss": 0.32608968019485474,
100
+ "eval_runtime": 295.3154,
101
+ "eval_samples_per_second": 13.013,
102
+ "eval_steps_per_second": 0.41,
103
+ "eval_wer": 0.34501188943912386,
104
+ "step": 2400
105
+ },
106
+ {
107
+ "epoch": 4.49,
108
+ "learning_rate": 4.4807849550286185e-05,
109
+ "loss": 0.3951,
110
+ "step": 2800
111
+ },
112
+ {
113
+ "epoch": 4.49,
114
+ "eval_cer": 0.1047794204556737,
115
+ "eval_loss": 0.3201062083244324,
116
+ "eval_runtime": 296.053,
117
+ "eval_samples_per_second": 12.981,
118
+ "eval_steps_per_second": 0.409,
119
+ "eval_wer": 0.332489801260935,
120
+ "step": 2800
121
+ },
122
+ {
123
+ "epoch": 5.13,
124
+ "learning_rate": 3.66312346688471e-05,
125
+ "loss": 0.3619,
126
+ "step": 3200
127
+ },
128
+ {
129
+ "epoch": 5.13,
130
+ "eval_cer": 0.1030111909073287,
131
+ "eval_loss": 0.3167116940021515,
132
+ "eval_runtime": 295.6579,
133
+ "eval_samples_per_second": 12.998,
134
+ "eval_steps_per_second": 0.409,
135
+ "eval_wer": 0.3296101572896442,
136
+ "step": 3200
137
+ },
138
+ {
139
+ "epoch": 5.77,
140
+ "learning_rate": 2.8454619787408017e-05,
141
+ "loss": 0.345,
142
+ "step": 3600
143
+ },
144
+ {
145
+ "epoch": 5.77,
146
+ "eval_cer": 0.10128551233741981,
147
+ "eval_loss": 0.31573495268821716,
148
+ "eval_runtime": 296.0935,
149
+ "eval_samples_per_second": 12.979,
150
+ "eval_steps_per_second": 0.409,
151
+ "eval_wer": 0.3209712253757717,
152
+ "step": 3600
153
+ },
154
+ {
155
+ "epoch": 6.41,
156
+ "learning_rate": 2.027800490596893e-05,
157
+ "loss": 0.338,
158
+ "step": 4000
159
+ },
160
+ {
161
+ "epoch": 6.41,
162
+ "eval_cer": 0.0982123861170341,
163
+ "eval_loss": 0.305124431848526,
164
+ "eval_runtime": 295.7779,
165
+ "eval_samples_per_second": 12.993,
166
+ "eval_steps_per_second": 0.409,
167
+ "eval_wer": 0.31429568707868843,
168
+ "step": 4000
169
+ },
170
+ {
171
+ "epoch": 7.05,
172
+ "learning_rate": 1.2121831561733443e-05,
173
+ "loss": 0.3155,
174
+ "step": 4400
175
+ },
176
+ {
177
+ "epoch": 7.05,
178
+ "eval_cer": 0.09860007280945199,
179
+ "eval_loss": 0.30590805411338806,
180
+ "eval_runtime": 296.0673,
181
+ "eval_samples_per_second": 12.98,
182
+ "eval_steps_per_second": 0.409,
183
+ "eval_wer": 0.315408276794869,
184
+ "step": 4400
185
+ },
186
+ {
187
+ "epoch": 7.69,
188
+ "learning_rate": 3.945216680294359e-06,
189
+ "loss": 0.3057,
190
+ "step": 4800
191
+ },
192
+ {
193
+ "epoch": 7.69,
194
+ "eval_cer": 0.09723844149949648,
195
+ "eval_loss": 0.3034806251525879,
196
+ "eval_runtime": 295.665,
197
+ "eval_samples_per_second": 12.998,
198
+ "eval_steps_per_second": 0.409,
199
+ "eval_wer": 0.3137066689936517,
200
+ "step": 4800
201
+ },
202
+ {
203
+ "epoch": 8.0,
204
+ "step": 4992,
205
+ "total_flos": 3.347893330222721e+19,
206
+ "train_loss": 0.47035410618170714,
207
+ "train_runtime": 23681.5399,
208
+ "train_samples_per_second": 13.496,
209
+ "train_steps_per_second": 0.211
210
+ }
211
+ ],
212
+ "max_steps": 4992,
213
+ "num_train_epochs": 8,
214
+ "total_flos": 3.347893330222721e+19,
215
+ "trial_name": null,
216
+ "trial_params": null
217
+ }