infinitejoy commited on
Commit
ee86223
1 Parent(s): 25f7543

End of training

Browse files
all_results.json CHANGED
@@ -1,14 +1,14 @@
1
  {
2
- "epoch": 1.0,
3
- "eval_loss": 14.633834838867188,
4
- "eval_runtime": 13.0798,
5
  "eval_samples": 301,
6
- "eval_samples_per_second": 23.013,
7
- "eval_steps_per_second": 0.765,
8
- "eval_wer": 1.0,
9
- "train_loss": 15.28215859153054,
10
- "train_runtime": 50.1491,
11
  "train_samples": 704,
12
- "train_samples_per_second": 14.038,
13
- "train_steps_per_second": 0.439
14
  }
 
1
  {
2
+ "epoch": 100.0,
3
+ "eval_loss": 0.535879373550415,
4
+ "eval_runtime": 13.4178,
5
  "eval_samples": 301,
6
+ "eval_samples_per_second": 22.433,
7
+ "eval_steps_per_second": 0.745,
8
+ "eval_wer": 0.6191872643485546,
9
+ "train_loss": 1.9645296894420277,
10
+ "train_runtime": 5564.2506,
11
  "train_samples": 704,
12
+ "train_samples_per_second": 12.652,
13
+ "train_steps_per_second": 0.395
14
  }
eval_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 1.0,
3
- "eval_loss": 14.633834838867188,
4
- "eval_runtime": 13.0798,
5
  "eval_samples": 301,
6
- "eval_samples_per_second": 23.013,
7
- "eval_steps_per_second": 0.765,
8
- "eval_wer": 1.0
9
  }
 
1
  {
2
+ "epoch": 100.0,
3
+ "eval_loss": 0.535879373550415,
4
+ "eval_runtime": 13.4178,
5
  "eval_samples": 301,
6
+ "eval_samples_per_second": 22.433,
7
+ "eval_steps_per_second": 0.745,
8
+ "eval_wer": 0.6191872643485546
9
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:048b851ca4920043c2b10c6c2c9a342537084ee9653667ef3732a7b7a03750db
3
  size 1262108145
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78b042e9bf1f74a24f8ca542b89658c28b5ae650b6d32eaaa99d0607f8b042e2
3
  size 1262108145
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 1.0,
3
- "train_loss": 15.28215859153054,
4
- "train_runtime": 50.1491,
5
  "train_samples": 704,
6
- "train_samples_per_second": 14.038,
7
- "train_steps_per_second": 0.439
8
  }
 
1
  {
2
+ "epoch": 100.0,
3
+ "train_loss": 1.9645296894420277,
4
+ "train_runtime": 5564.2506,
5
  "train_samples": 704,
6
+ "train_samples_per_second": 12.652,
7
+ "train_steps_per_second": 0.395
8
  }
trainer_state.json CHANGED
@@ -1,25 +1,193 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.0,
5
- "global_step": 22,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
- "epoch": 1.0,
12
- "step": 22,
13
- "total_flos": 1.5056241143943168e+17,
14
- "train_loss": 15.28215859153054,
15
- "train_runtime": 50.1491,
16
- "train_samples_per_second": 14.038,
17
- "train_steps_per_second": 0.439
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  }
19
  ],
20
- "max_steps": 22,
21
- "num_train_epochs": 1,
22
- "total_flos": 1.5056241143943168e+17,
23
  "trial_name": null,
24
  "trial_params": null
25
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 100.0,
5
+ "global_step": 2200,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
+ "epoch": 4.55,
12
+ "learning_rate": 3.4299999999999993e-05,
13
+ "loss": 9.7952,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 9.09,
18
+ "learning_rate": 6.929999999999999e-05,
19
+ "loss": 3.6097,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 13.64,
24
+ "learning_rate": 6.657e-05,
25
+ "loss": 3.0981,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 18.18,
30
+ "learning_rate": 6.306999999999999e-05,
31
+ "loss": 3.0036,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 22.73,
36
+ "learning_rate": 5.9569999999999994e-05,
37
+ "loss": 2.8617,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 22.73,
42
+ "eval_loss": 2.626376152038574,
43
+ "eval_runtime": 13.7667,
44
+ "eval_samples_per_second": 21.864,
45
+ "eval_steps_per_second": 0.726,
46
+ "eval_wer": 1.0012568077084205,
47
+ "step": 500
48
+ },
49
+ {
50
+ "epoch": 27.27,
51
+ "learning_rate": 5.607e-05,
52
+ "loss": 2.4115,
53
+ "step": 600
54
+ },
55
+ {
56
+ "epoch": 31.82,
57
+ "learning_rate": 5.256999999999999e-05,
58
+ "loss": 1.8202,
59
+ "step": 700
60
+ },
61
+ {
62
+ "epoch": 36.36,
63
+ "learning_rate": 4.906999999999999e-05,
64
+ "loss": 1.5225,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 40.91,
69
+ "learning_rate": 4.557e-05,
70
+ "loss": 1.3681,
71
+ "step": 900
72
+ },
73
+ {
74
+ "epoch": 45.45,
75
+ "learning_rate": 4.2069999999999995e-05,
76
+ "loss": 1.2716,
77
+ "step": 1000
78
+ },
79
+ {
80
+ "epoch": 45.45,
81
+ "eval_loss": 0.6217501163482666,
82
+ "eval_runtime": 13.948,
83
+ "eval_samples_per_second": 21.58,
84
+ "eval_steps_per_second": 0.717,
85
+ "eval_wer": 0.6941767909509845,
86
+ "step": 1000
87
+ },
88
+ {
89
+ "epoch": 50.0,
90
+ "learning_rate": 3.857e-05,
91
+ "loss": 1.2079,
92
+ "step": 1100
93
+ },
94
+ {
95
+ "epoch": 54.55,
96
+ "learning_rate": 3.5069999999999995e-05,
97
+ "loss": 1.154,
98
+ "step": 1200
99
+ },
100
+ {
101
+ "epoch": 59.09,
102
+ "learning_rate": 3.157e-05,
103
+ "loss": 1.1124,
104
+ "step": 1300
105
+ },
106
+ {
107
+ "epoch": 63.64,
108
+ "learning_rate": 2.807e-05,
109
+ "loss": 1.087,
110
+ "step": 1400
111
+ },
112
+ {
113
+ "epoch": 68.18,
114
+ "learning_rate": 2.4569999999999997e-05,
115
+ "loss": 1.049,
116
+ "step": 1500
117
+ },
118
+ {
119
+ "epoch": 68.18,
120
+ "eval_loss": 0.544226348400116,
121
+ "eval_runtime": 13.3563,
122
+ "eval_samples_per_second": 22.536,
123
+ "eval_steps_per_second": 0.749,
124
+ "eval_wer": 0.6367825722664432,
125
+ "step": 1500
126
+ },
127
+ {
128
+ "epoch": 72.73,
129
+ "learning_rate": 2.1069999999999996e-05,
130
+ "loss": 1.023,
131
+ "step": 1600
132
+ },
133
+ {
134
+ "epoch": 77.27,
135
+ "learning_rate": 1.757e-05,
136
+ "loss": 1.0046,
137
+ "step": 1700
138
+ },
139
+ {
140
+ "epoch": 81.82,
141
+ "learning_rate": 1.4069999999999999e-05,
142
+ "loss": 0.9863,
143
+ "step": 1800
144
+ },
145
+ {
146
+ "epoch": 86.36,
147
+ "learning_rate": 1.0569999999999999e-05,
148
+ "loss": 0.9772,
149
+ "step": 1900
150
+ },
151
+ {
152
+ "epoch": 90.91,
153
+ "learning_rate": 7.07e-06,
154
+ "loss": 0.9632,
155
+ "step": 2000
156
+ },
157
+ {
158
+ "epoch": 90.91,
159
+ "eval_loss": 0.5364273190498352,
160
+ "eval_runtime": 13.6559,
161
+ "eval_samples_per_second": 22.042,
162
+ "eval_steps_per_second": 0.732,
163
+ "eval_wer": 0.6242144951822372,
164
+ "step": 2000
165
+ },
166
+ {
167
+ "epoch": 95.45,
168
+ "learning_rate": 3.5699999999999993e-06,
169
+ "loss": 0.9443,
170
+ "step": 2100
171
+ },
172
+ {
173
+ "epoch": 100.0,
174
+ "learning_rate": 6.999999999999999e-08,
175
+ "loss": 0.9485,
176
+ "step": 2200
177
+ },
178
+ {
179
+ "epoch": 100.0,
180
+ "step": 2200,
181
+ "total_flos": 1.4967759850398351e+19,
182
+ "train_loss": 1.9645296894420277,
183
+ "train_runtime": 5564.2506,
184
+ "train_samples_per_second": 12.652,
185
+ "train_steps_per_second": 0.395
186
  }
187
  ],
188
+ "max_steps": 2200,
189
+ "num_train_epochs": 100,
190
+ "total_flos": 1.4967759850398351e+19,
191
  "trial_name": null,
192
  "trial_params": null
193
  }