Abdulwahab Sahyoun commited on
Commit
52b498d
1 Parent(s): fe84a79

End of training

Browse files
Files changed (4) hide show
  1. all_results.json +14 -0
  2. eval_results.json +9 -0
  3. train_results.json +8 -0
  4. trainer_state.json +367 -0
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 15.0,
3
+ "eval_loss": 0.6945873498916626,
4
+ "eval_runtime": 205.608,
5
+ "eval_samples": 4899,
6
+ "eval_samples_per_second": 23.827,
7
+ "eval_steps_per_second": 0.749,
8
+ "eval_wer": 0.3939828080229226,
9
+ "train_loss": 1.2329752604166666,
10
+ "train_runtime": 16157.973,
11
+ "train_samples": 14730,
12
+ "train_samples_per_second": 13.674,
13
+ "train_steps_per_second": 0.214
14
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 15.0,
3
+ "eval_loss": 0.6945873498916626,
4
+ "eval_runtime": 205.608,
5
+ "eval_samples": 4899,
6
+ "eval_samples_per_second": 23.827,
7
+ "eval_steps_per_second": 0.749,
8
+ "eval_wer": 0.3939828080229226
9
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 15.0,
3
+ "train_loss": 1.2329752604166666,
4
+ "train_runtime": 16157.973,
5
+ "train_samples": 14730,
6
+ "train_samples_per_second": 13.674,
7
+ "train_steps_per_second": 0.214
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,367 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 14.997830802603037,
5
+ "global_step": 3450,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.43,
12
+ "eval_loss": 3.6933705806732178,
13
+ "eval_runtime": 193.3371,
14
+ "eval_samples_per_second": 25.339,
15
+ "eval_steps_per_second": 0.797,
16
+ "eval_wer": 1.0,
17
+ "step": 100
18
+ },
19
+ {
20
+ "epoch": 0.87,
21
+ "eval_loss": 3.0763044357299805,
22
+ "eval_runtime": 193.2997,
23
+ "eval_samples_per_second": 25.344,
24
+ "eval_steps_per_second": 0.797,
25
+ "eval_wer": 1.0,
26
+ "step": 200
27
+ },
28
+ {
29
+ "epoch": 1.3,
30
+ "eval_loss": 2.9737112522125244,
31
+ "eval_runtime": 192.6243,
32
+ "eval_samples_per_second": 25.433,
33
+ "eval_steps_per_second": 0.799,
34
+ "eval_wer": 1.0,
35
+ "step": 300
36
+ },
37
+ {
38
+ "epoch": 1.74,
39
+ "eval_loss": 2.5733721256256104,
40
+ "eval_runtime": 192.0804,
41
+ "eval_samples_per_second": 25.505,
42
+ "eval_steps_per_second": 0.802,
43
+ "eval_wer": 1.0,
44
+ "step": 400
45
+ },
46
+ {
47
+ "epoch": 2.17,
48
+ "learning_rate": 0.0002982,
49
+ "loss": 5.0957,
50
+ "step": 500
51
+ },
52
+ {
53
+ "epoch": 2.17,
54
+ "eval_loss": 1.1900373697280884,
55
+ "eval_runtime": 192.7964,
56
+ "eval_samples_per_second": 25.41,
57
+ "eval_steps_per_second": 0.799,
58
+ "eval_wer": 0.9011191003946586,
59
+ "step": 500
60
+ },
61
+ {
62
+ "epoch": 2.61,
63
+ "eval_loss": 0.9725514054298401,
64
+ "eval_runtime": 192.2119,
65
+ "eval_samples_per_second": 25.487,
66
+ "eval_steps_per_second": 0.801,
67
+ "eval_wer": 0.7572309023084824,
68
+ "step": 600
69
+ },
70
+ {
71
+ "epoch": 3.04,
72
+ "eval_loss": 0.8960436582565308,
73
+ "eval_runtime": 193.7378,
74
+ "eval_samples_per_second": 25.287,
75
+ "eval_steps_per_second": 0.795,
76
+ "eval_wer": 0.6209114991620263,
77
+ "step": 700
78
+ },
79
+ {
80
+ "epoch": 3.48,
81
+ "eval_loss": 0.7850536704063416,
82
+ "eval_runtime": 193.8724,
83
+ "eval_samples_per_second": 25.269,
84
+ "eval_steps_per_second": 0.794,
85
+ "eval_wer": 0.5515488998215927,
86
+ "step": 800
87
+ },
88
+ {
89
+ "epoch": 3.91,
90
+ "eval_loss": 0.7271264791488647,
91
+ "eval_runtime": 195.9681,
92
+ "eval_samples_per_second": 24.999,
93
+ "eval_steps_per_second": 0.786,
94
+ "eval_wer": 0.511542412283073,
95
+ "step": 900
96
+ },
97
+ {
98
+ "epoch": 4.35,
99
+ "learning_rate": 0.00024945762711864404,
100
+ "loss": 1.0312,
101
+ "step": 1000
102
+ },
103
+ {
104
+ "epoch": 4.35,
105
+ "eval_loss": 0.7053087949752808,
106
+ "eval_runtime": 192.6577,
107
+ "eval_samples_per_second": 25.429,
108
+ "eval_steps_per_second": 0.799,
109
+ "eval_wer": 0.49548575444666704,
110
+ "step": 1000
111
+ },
112
+ {
113
+ "epoch": 4.78,
114
+ "eval_loss": 0.6822740435600281,
115
+ "eval_runtime": 193.7373,
116
+ "eval_samples_per_second": 25.287,
117
+ "eval_steps_per_second": 0.795,
118
+ "eval_wer": 0.47369843758447316,
119
+ "step": 1100
120
+ },
121
+ {
122
+ "epoch": 5.22,
123
+ "eval_loss": 0.6767629384994507,
124
+ "eval_runtime": 192.3102,
125
+ "eval_samples_per_second": 25.474,
126
+ "eval_steps_per_second": 0.801,
127
+ "eval_wer": 0.45945288425150027,
128
+ "step": 1200
129
+ },
130
+ {
131
+ "epoch": 5.65,
132
+ "eval_loss": 0.6635140776634216,
133
+ "eval_runtime": 192.3892,
134
+ "eval_samples_per_second": 25.464,
135
+ "eval_steps_per_second": 0.8,
136
+ "eval_wer": 0.4488295399253933,
137
+ "step": 1300
138
+ },
139
+ {
140
+ "epoch": 6.09,
141
+ "eval_loss": 0.6602210402488708,
142
+ "eval_runtime": 191.9311,
143
+ "eval_samples_per_second": 25.525,
144
+ "eval_steps_per_second": 0.802,
145
+ "eval_wer": 0.4389630750932584,
146
+ "step": 1400
147
+ },
148
+ {
149
+ "epoch": 6.52,
150
+ "learning_rate": 0.0001986101694915254,
151
+ "loss": 0.6815,
152
+ "step": 1500
153
+ },
154
+ {
155
+ "epoch": 6.52,
156
+ "eval_loss": 0.6463663578033447,
157
+ "eval_runtime": 193.1979,
158
+ "eval_samples_per_second": 25.357,
159
+ "eval_steps_per_second": 0.797,
160
+ "eval_wer": 0.4310158404065524,
161
+ "step": 1500
162
+ },
163
+ {
164
+ "epoch": 6.95,
165
+ "eval_loss": 0.6455022096633911,
166
+ "eval_runtime": 198.417,
167
+ "eval_samples_per_second": 24.69,
168
+ "eval_steps_per_second": 0.776,
169
+ "eval_wer": 0.43942260907174135,
170
+ "step": 1600
171
+ },
172
+ {
173
+ "epoch": 7.39,
174
+ "eval_loss": 0.6629974842071533,
175
+ "eval_runtime": 197.9313,
176
+ "eval_samples_per_second": 24.751,
177
+ "eval_steps_per_second": 0.778,
178
+ "eval_wer": 0.4311780288695464,
179
+ "step": 1700
180
+ },
181
+ {
182
+ "epoch": 7.82,
183
+ "eval_loss": 0.6521152853965759,
184
+ "eval_runtime": 202.8016,
185
+ "eval_samples_per_second": 24.157,
186
+ "eval_steps_per_second": 0.759,
187
+ "eval_wer": 0.4125804184462345,
188
+ "step": 1800
189
+ },
190
+ {
191
+ "epoch": 8.26,
192
+ "eval_loss": 0.6281991600990295,
193
+ "eval_runtime": 195.8017,
194
+ "eval_samples_per_second": 25.02,
195
+ "eval_steps_per_second": 0.787,
196
+ "eval_wer": 0.42839379358814944,
197
+ "step": 1900
198
+ },
199
+ {
200
+ "epoch": 8.69,
201
+ "learning_rate": 0.00014776271186440676,
202
+ "loss": 0.544,
203
+ "step": 2000
204
+ },
205
+ {
206
+ "epoch": 8.69,
207
+ "eval_loss": 0.6247670650482178,
208
+ "eval_runtime": 197.0538,
209
+ "eval_samples_per_second": 24.861,
210
+ "eval_steps_per_second": 0.782,
211
+ "eval_wer": 0.4178245120830405,
212
+ "step": 2000
213
+ },
214
+ {
215
+ "epoch": 9.13,
216
+ "eval_loss": 0.6510103940963745,
217
+ "eval_runtime": 199.9527,
218
+ "eval_samples_per_second": 24.501,
219
+ "eval_steps_per_second": 0.77,
220
+ "eval_wer": 0.4103638427853165,
221
+ "step": 2100
222
+ },
223
+ {
224
+ "epoch": 9.56,
225
+ "eval_loss": 0.6527135968208313,
226
+ "eval_runtime": 202.4561,
227
+ "eval_samples_per_second": 24.198,
228
+ "eval_steps_per_second": 0.761,
229
+ "eval_wer": 0.4012542574471536,
230
+ "step": 2200
231
+ },
232
+ {
233
+ "epoch": 10.0,
234
+ "eval_loss": 0.6510817408561707,
235
+ "eval_runtime": 202.1587,
236
+ "eval_samples_per_second": 24.233,
237
+ "eval_steps_per_second": 0.762,
238
+ "eval_wer": 0.40641725685246255,
239
+ "step": 2300
240
+ },
241
+ {
242
+ "epoch": 10.43,
243
+ "eval_loss": 0.6734431982040405,
244
+ "eval_runtime": 202.8997,
245
+ "eval_samples_per_second": 24.145,
246
+ "eval_steps_per_second": 0.759,
247
+ "eval_wer": 0.4060928799264746,
248
+ "step": 2400
249
+ },
250
+ {
251
+ "epoch": 10.87,
252
+ "learning_rate": 9.691525423728813e-05,
253
+ "loss": 0.4478,
254
+ "step": 2500
255
+ },
256
+ {
257
+ "epoch": 10.87,
258
+ "eval_loss": 0.6756427884101868,
259
+ "eval_runtime": 201.7844,
260
+ "eval_samples_per_second": 24.278,
261
+ "eval_steps_per_second": 0.763,
262
+ "eval_wer": 0.41452668000216253,
263
+ "step": 2500
264
+ },
265
+ {
266
+ "epoch": 11.3,
267
+ "eval_loss": 0.6727204918861389,
268
+ "eval_runtime": 202.413,
269
+ "eval_samples_per_second": 24.203,
270
+ "eval_steps_per_second": 0.761,
271
+ "eval_wer": 0.3989565875547386,
272
+ "step": 2600
273
+ },
274
+ {
275
+ "epoch": 11.74,
276
+ "eval_loss": 0.6618625521659851,
277
+ "eval_runtime": 205.4327,
278
+ "eval_samples_per_second": 23.847,
279
+ "eval_steps_per_second": 0.75,
280
+ "eval_wer": 0.4007406606476726,
281
+ "step": 2700
282
+ },
283
+ {
284
+ "epoch": 12.17,
285
+ "eval_loss": 0.6613638997077942,
286
+ "eval_runtime": 202.1353,
287
+ "eval_samples_per_second": 24.236,
288
+ "eval_steps_per_second": 0.762,
289
+ "eval_wer": 0.40193004270962857,
290
+ "step": 2800
291
+ },
292
+ {
293
+ "epoch": 12.61,
294
+ "eval_loss": 0.6695073246955872,
295
+ "eval_runtime": 202.6466,
296
+ "eval_samples_per_second": 24.175,
297
+ "eval_steps_per_second": 0.76,
298
+ "eval_wer": 0.40036222090068657,
299
+ "step": 2900
300
+ },
301
+ {
302
+ "epoch": 13.04,
303
+ "learning_rate": 4.606779661016948e-05,
304
+ "loss": 0.3919,
305
+ "step": 3000
306
+ },
307
+ {
308
+ "epoch": 13.04,
309
+ "eval_loss": 0.6778153777122498,
310
+ "eval_runtime": 202.4696,
311
+ "eval_samples_per_second": 24.196,
312
+ "eval_steps_per_second": 0.761,
313
+ "eval_wer": 0.3966048548413256,
314
+ "step": 3000
315
+ },
316
+ {
317
+ "epoch": 13.48,
318
+ "eval_loss": 0.6872451305389404,
319
+ "eval_runtime": 201.7642,
320
+ "eval_samples_per_second": 24.281,
321
+ "eval_steps_per_second": 0.763,
322
+ "eval_wer": 0.3971184516408066,
323
+ "step": 3100
324
+ },
325
+ {
326
+ "epoch": 13.91,
327
+ "eval_loss": 0.688228964805603,
328
+ "eval_runtime": 204.8269,
329
+ "eval_samples_per_second": 23.918,
330
+ "eval_steps_per_second": 0.752,
331
+ "eval_wer": 0.39452343623290265,
332
+ "step": 3200
333
+ },
334
+ {
335
+ "epoch": 14.35,
336
+ "eval_loss": 0.6937959790229797,
337
+ "eval_runtime": 202.6788,
338
+ "eval_samples_per_second": 24.171,
339
+ "eval_steps_per_second": 0.76,
340
+ "eval_wer": 0.39365843109693466,
341
+ "step": 3300
342
+ },
343
+ {
344
+ "epoch": 14.78,
345
+ "eval_loss": 0.6927648782730103,
346
+ "eval_runtime": 201.0141,
347
+ "eval_samples_per_second": 24.371,
348
+ "eval_steps_per_second": 0.766,
349
+ "eval_wer": 0.3945504676434016,
350
+ "step": 3400
351
+ },
352
+ {
353
+ "epoch": 15.0,
354
+ "step": 3450,
355
+ "total_flos": 2.9875620378207076e+19,
356
+ "train_loss": 1.2329752604166666,
357
+ "train_runtime": 16157.973,
358
+ "train_samples_per_second": 13.674,
359
+ "train_steps_per_second": 0.214
360
+ }
361
+ ],
362
+ "max_steps": 3450,
363
+ "num_train_epochs": 15,
364
+ "total_flos": 2.9875620378207076e+19,
365
+ "trial_name": null,
366
+ "trial_params": null
367
+ }