tuananh7198 commited on
Commit
65a05de
1 Parent(s): 4af3796

End of training

Browse files
all_results.json CHANGED
@@ -1,7 +1,12 @@
1
  {
2
- "epoch": 1.0,
3
- "train_loss": 1.7669422626495361,
4
- "train_runtime": 101.3421,
5
- "train_samples_per_second": 0.316,
6
- "train_steps_per_second": 0.01
 
 
 
 
 
7
  }
 
1
  {
2
+ "epoch": 4.01,
3
+ "eval_loss": 0.5422283411026001,
4
+ "eval_runtime": 892.2169,
5
+ "eval_samples_per_second": 1.386,
6
+ "eval_steps_per_second": 0.087,
7
+ "eval_wer": 20.04825619653433,
8
+ "train_loss": 0.19602714616060257,
9
+ "train_runtime": 17744.1511,
10
+ "train_samples_per_second": 0.902,
11
+ "train_steps_per_second": 0.056
12
  }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.01,
3
+ "eval_loss": 0.5422283411026001,
4
+ "eval_runtime": 892.2169,
5
+ "eval_samples_per_second": 1.386,
6
+ "eval_steps_per_second": 0.087,
7
+ "eval_wer": 20.04825619653433
8
+ }
runs/Dec16_16-42-37_ip-172-16-16-209.ec2.internal/events.out.tfevents.1671227845.ip-172-16-16-209.ec2.internal.21913.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60be2dae64a3a14b8cbf8a5181118533e640096037b9693e09e37a4b7e6fd44f
3
+ size 358
train_results.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "epoch": 1.0,
3
- "train_loss": 1.7669422626495361,
4
- "train_runtime": 101.3421,
5
- "train_samples_per_second": 0.316,
6
- "train_steps_per_second": 0.01
7
  }
 
1
  {
2
+ "epoch": 4.01,
3
+ "train_loss": 0.19602714616060257,
4
+ "train_runtime": 17744.1511,
5
+ "train_samples_per_second": 0.902,
6
+ "train_steps_per_second": 0.056
7
  }
trainer_state.json CHANGED
@@ -1,25 +1,274 @@
1
  {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
- "epoch": 1.0,
5
- "global_step": 1,
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": 1,
13
- "total_flos": 4082419630080000.0,
14
- "train_loss": 1.7669422626495361,
15
- "train_runtime": 101.3421,
16
- "train_samples_per_second": 0.316,
17
- "train_steps_per_second": 0.01
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
18
  }
19
  ],
20
- "max_steps": 1,
21
  "num_train_epochs": 9223372036854775807,
22
- "total_flos": 4082419630080000.0,
23
  "trial_name": null,
24
  "trial_params": null
25
  }
 
1
  {
2
+ "best_metric": 20.04825619653433,
3
+ "best_model_checkpoint": "./checkpoint-1000",
4
+ "epoch": 4.008,
5
+ "global_step": 1000,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 2.1000000000000002e-06,
13
+ "loss": 1.7151,
14
+ "step": 25
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 4.600000000000001e-06,
19
+ "loss": 0.7627,
20
+ "step": 50
21
+ },
22
+ {
23
+ "epoch": 0.07,
24
+ "learning_rate": 7.100000000000001e-06,
25
+ "loss": 0.3638,
26
+ "step": 75
27
+ },
28
+ {
29
+ "epoch": 0.1,
30
+ "learning_rate": 9.600000000000001e-06,
31
+ "loss": 0.3364,
32
+ "step": 100
33
+ },
34
+ {
35
+ "epoch": 0.12,
36
+ "learning_rate": 9.766666666666667e-06,
37
+ "loss": 0.3486,
38
+ "step": 125
39
+ },
40
+ {
41
+ "epoch": 0.15,
42
+ "learning_rate": 9.48888888888889e-06,
43
+ "loss": 0.3894,
44
+ "step": 150
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "learning_rate": 9.211111111111111e-06,
49
+ "loss": 0.4697,
50
+ "step": 175
51
+ },
52
+ {
53
+ "epoch": 0.2,
54
+ "learning_rate": 8.933333333333333e-06,
55
+ "loss": 0.3761,
56
+ "step": 200
57
+ },
58
+ {
59
+ "epoch": 0.23,
60
+ "learning_rate": 8.655555555555557e-06,
61
+ "loss": 0.3839,
62
+ "step": 225
63
+ },
64
  {
65
  "epoch": 1.0,
66
+ "learning_rate": 8.377777777777779e-06,
67
+ "loss": 0.3711,
68
+ "step": 250
69
+ },
70
+ {
71
+ "epoch": 1.03,
72
+ "learning_rate": 8.1e-06,
73
+ "loss": 0.1781,
74
+ "step": 275
75
+ },
76
+ {
77
+ "epoch": 1.05,
78
+ "learning_rate": 7.822222222222224e-06,
79
+ "loss": 0.1591,
80
+ "step": 300
81
+ },
82
+ {
83
+ "epoch": 1.08,
84
+ "learning_rate": 7.544444444444445e-06,
85
+ "loss": 0.1354,
86
+ "step": 325
87
+ },
88
+ {
89
+ "epoch": 1.1,
90
+ "learning_rate": 7.266666666666668e-06,
91
+ "loss": 0.1823,
92
+ "step": 350
93
+ },
94
+ {
95
+ "epoch": 1.13,
96
+ "learning_rate": 6.9888888888888895e-06,
97
+ "loss": 0.1651,
98
+ "step": 375
99
+ },
100
+ {
101
+ "epoch": 1.15,
102
+ "learning_rate": 6.711111111111111e-06,
103
+ "loss": 0.1603,
104
+ "step": 400
105
+ },
106
+ {
107
+ "epoch": 1.18,
108
+ "learning_rate": 6.433333333333333e-06,
109
+ "loss": 0.1426,
110
+ "step": 425
111
+ },
112
+ {
113
+ "epoch": 1.2,
114
+ "learning_rate": 6.155555555555556e-06,
115
+ "loss": 0.178,
116
+ "step": 450
117
+ },
118
+ {
119
+ "epoch": 1.23,
120
+ "learning_rate": 5.877777777777778e-06,
121
+ "loss": 0.1491,
122
+ "step": 475
123
+ },
124
+ {
125
+ "epoch": 2.0,
126
+ "learning_rate": 5.600000000000001e-06,
127
+ "loss": 0.0891,
128
+ "step": 500
129
+ },
130
+ {
131
+ "epoch": 2.03,
132
+ "learning_rate": 5.322222222222223e-06,
133
+ "loss": 0.0439,
134
+ "step": 525
135
+ },
136
+ {
137
+ "epoch": 2.05,
138
+ "learning_rate": 5.044444444444445e-06,
139
+ "loss": 0.0815,
140
+ "step": 550
141
+ },
142
+ {
143
+ "epoch": 2.08,
144
+ "learning_rate": 4.766666666666667e-06,
145
+ "loss": 0.0708,
146
+ "step": 575
147
+ },
148
+ {
149
+ "epoch": 2.1,
150
+ "learning_rate": 4.488888888888889e-06,
151
+ "loss": 0.0662,
152
+ "step": 600
153
+ },
154
+ {
155
+ "epoch": 2.13,
156
+ "learning_rate": 4.211111111111112e-06,
157
+ "loss": 0.076,
158
+ "step": 625
159
+ },
160
+ {
161
+ "epoch": 2.15,
162
+ "learning_rate": 3.9333333333333335e-06,
163
+ "loss": 0.0632,
164
+ "step": 650
165
+ },
166
+ {
167
+ "epoch": 2.18,
168
+ "learning_rate": 3.6555555555555562e-06,
169
+ "loss": 0.0461,
170
+ "step": 675
171
+ },
172
+ {
173
+ "epoch": 2.2,
174
+ "learning_rate": 3.377777777777778e-06,
175
+ "loss": 0.0826,
176
+ "step": 700
177
+ },
178
+ {
179
+ "epoch": 2.23,
180
+ "learning_rate": 3.1000000000000004e-06,
181
+ "loss": 0.0674,
182
+ "step": 725
183
+ },
184
+ {
185
+ "epoch": 3.01,
186
+ "learning_rate": 2.8222222222222223e-06,
187
+ "loss": 0.0247,
188
+ "step": 750
189
+ },
190
+ {
191
+ "epoch": 3.03,
192
+ "learning_rate": 2.5444444444444446e-06,
193
+ "loss": 0.0147,
194
+ "step": 775
195
+ },
196
+ {
197
+ "epoch": 3.06,
198
+ "learning_rate": 2.266666666666667e-06,
199
+ "loss": 0.0141,
200
+ "step": 800
201
+ },
202
+ {
203
+ "epoch": 3.08,
204
+ "learning_rate": 1.988888888888889e-06,
205
+ "loss": 0.0255,
206
+ "step": 825
207
+ },
208
+ {
209
+ "epoch": 3.11,
210
+ "learning_rate": 1.7111111111111112e-06,
211
+ "loss": 0.011,
212
+ "step": 850
213
+ },
214
+ {
215
+ "epoch": 3.13,
216
+ "learning_rate": 1.4333333333333335e-06,
217
+ "loss": 0.0099,
218
+ "step": 875
219
+ },
220
+ {
221
+ "epoch": 3.16,
222
+ "learning_rate": 1.1555555555555556e-06,
223
+ "loss": 0.013,
224
+ "step": 900
225
+ },
226
+ {
227
+ "epoch": 3.18,
228
+ "learning_rate": 8.777777777777778e-07,
229
+ "loss": 0.0213,
230
+ "step": 925
231
+ },
232
+ {
233
+ "epoch": 3.21,
234
+ "learning_rate": 6.000000000000001e-07,
235
+ "loss": 0.0119,
236
+ "step": 950
237
+ },
238
+ {
239
+ "epoch": 3.23,
240
+ "learning_rate": 3.2222222222222227e-07,
241
+ "loss": 0.0173,
242
+ "step": 975
243
+ },
244
+ {
245
+ "epoch": 4.01,
246
+ "learning_rate": 4.444444444444445e-08,
247
+ "loss": 0.0241,
248
+ "step": 1000
249
+ },
250
+ {
251
+ "epoch": 4.01,
252
+ "eval_loss": 0.5422283411026001,
253
+ "eval_runtime": 909.8399,
254
+ "eval_samples_per_second": 1.36,
255
+ "eval_steps_per_second": 0.086,
256
+ "eval_wer": 20.04825619653433,
257
+ "step": 1000
258
+ },
259
+ {
260
+ "epoch": 4.01,
261
+ "step": 1000,
262
+ "total_flos": 2.04120981504e+18,
263
+ "train_loss": 0.19602714616060257,
264
+ "train_runtime": 17744.1511,
265
+ "train_samples_per_second": 0.902,
266
+ "train_steps_per_second": 0.056
267
  }
268
  ],
269
+ "max_steps": 1000,
270
  "num_train_epochs": 9223372036854775807,
271
+ "total_flos": 2.04120981504e+18,
272
  "trial_name": null,
273
  "trial_params": null
274
  }