noflm commited on
Commit
e64cae7
1 Parent(s): da850fd

Training in progress, step 7000

Browse files
checkpoint-7000/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openai/whisper-base",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50257
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "d_model": 512,
15
+ "decoder_attention_heads": 8,
16
+ "decoder_ffn_dim": 2048,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 8,
22
+ "encoder_ffn_dim": 2048,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 6,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": [
27
+ [
28
+ 1,
29
+ 50266
30
+ ],
31
+ [
32
+ 2,
33
+ 50359
34
+ ],
35
+ [
36
+ 3,
37
+ 50363
38
+ ]
39
+ ],
40
+ "init_std": 0.02,
41
+ "is_encoder_decoder": true,
42
+ "max_length": 448,
43
+ "max_source_positions": 1500,
44
+ "max_target_positions": 448,
45
+ "model_type": "whisper",
46
+ "num_hidden_layers": 6,
47
+ "num_mel_bins": 80,
48
+ "pad_token_id": 50257,
49
+ "scale_embedding": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.26.0.dev0",
52
+ "use_cache": false,
53
+ "vocab_size": 51865
54
+ }
checkpoint-7000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db533fde3f64336561ce2878b8e4038b718e68343949d83d8511159b70001353
3
+ size 580895237
checkpoint-7000/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-7000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5bcae64ae2a6ac8902c7cafa598f375ad75277df9c489d88a2605619107181f
3
+ size 290458721
checkpoint-7000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad7660ffd37ae757846069ec68f062fb4c56bbbe2878b46d27dd05ee291628e6
3
+ size 14575
checkpoint-7000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a188787883d2abb8426969b8b379decb7256016b9381af48f1bdb8bebeb7e3f
3
+ size 557
checkpoint-7000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:605bf64ff9f162204cfcef93f745b02e81865784f8b37d950374025d238ff002
3
+ size 627
checkpoint-7000/trainer_state.json ADDED
@@ -0,0 +1,499 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 777.0007,
5
+ "global_step": 7000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 11.0,
12
+ "learning_rate": 4.85e-06,
13
+ "loss": 0.716,
14
+ "step": 100
15
+ },
16
+ {
17
+ "epoch": 22.0,
18
+ "learning_rate": 9.85e-06,
19
+ "loss": 0.0615,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 33.0,
24
+ "learning_rate": 1e-05,
25
+ "loss": 0.0043,
26
+ "step": 300
27
+ },
28
+ {
29
+ "epoch": 44.0,
30
+ "learning_rate": 1e-05,
31
+ "loss": 0.002,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 55.0,
36
+ "learning_rate": 1e-05,
37
+ "loss": 0.0008,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 66.0,
42
+ "learning_rate": 1e-05,
43
+ "loss": 0.0006,
44
+ "step": 600
45
+ },
46
+ {
47
+ "epoch": 77.0,
48
+ "learning_rate": 1e-05,
49
+ "loss": 0.0004,
50
+ "step": 700
51
+ },
52
+ {
53
+ "epoch": 88.0,
54
+ "learning_rate": 1e-05,
55
+ "loss": 0.0003,
56
+ "step": 800
57
+ },
58
+ {
59
+ "epoch": 99.0,
60
+ "learning_rate": 1e-05,
61
+ "loss": 0.0003,
62
+ "step": 900
63
+ },
64
+ {
65
+ "epoch": 111.0,
66
+ "learning_rate": 1e-05,
67
+ "loss": 0.0002,
68
+ "step": 1000
69
+ },
70
+ {
71
+ "epoch": 111.0,
72
+ "eval_loss": 0.21552081406116486,
73
+ "eval_runtime": 10.773,
74
+ "eval_samples_per_second": 1.671,
75
+ "eval_steps_per_second": 0.093,
76
+ "eval_wer": 9.75609756097561,
77
+ "step": 1000
78
+ },
79
+ {
80
+ "epoch": 122.0,
81
+ "learning_rate": 1e-05,
82
+ "loss": 0.0002,
83
+ "step": 1100
84
+ },
85
+ {
86
+ "epoch": 133.0,
87
+ "learning_rate": 1e-05,
88
+ "loss": 0.0002,
89
+ "step": 1200
90
+ },
91
+ {
92
+ "epoch": 144.0,
93
+ "learning_rate": 1e-05,
94
+ "loss": 0.0001,
95
+ "step": 1300
96
+ },
97
+ {
98
+ "epoch": 155.0,
99
+ "learning_rate": 1e-05,
100
+ "loss": 0.0001,
101
+ "step": 1400
102
+ },
103
+ {
104
+ "epoch": 166.0,
105
+ "learning_rate": 1e-05,
106
+ "loss": 0.0001,
107
+ "step": 1500
108
+ },
109
+ {
110
+ "epoch": 177.0,
111
+ "learning_rate": 1e-05,
112
+ "loss": 0.0001,
113
+ "step": 1600
114
+ },
115
+ {
116
+ "epoch": 188.0,
117
+ "learning_rate": 1e-05,
118
+ "loss": 0.0001,
119
+ "step": 1700
120
+ },
121
+ {
122
+ "epoch": 199.0,
123
+ "learning_rate": 1e-05,
124
+ "loss": 0.0001,
125
+ "step": 1800
126
+ },
127
+ {
128
+ "epoch": 211.0,
129
+ "learning_rate": 1e-05,
130
+ "loss": 0.0001,
131
+ "step": 1900
132
+ },
133
+ {
134
+ "epoch": 222.0,
135
+ "learning_rate": 1e-05,
136
+ "loss": 0.0001,
137
+ "step": 2000
138
+ },
139
+ {
140
+ "epoch": 222.0,
141
+ "eval_loss": 0.2447899729013443,
142
+ "eval_runtime": 14.1456,
143
+ "eval_samples_per_second": 1.272,
144
+ "eval_steps_per_second": 0.071,
145
+ "eval_wer": 12.195121951219512,
146
+ "step": 2000
147
+ },
148
+ {
149
+ "epoch": 233.0,
150
+ "learning_rate": 1e-05,
151
+ "loss": 0.0001,
152
+ "step": 2100
153
+ },
154
+ {
155
+ "epoch": 244.0,
156
+ "learning_rate": 1e-05,
157
+ "loss": 0.0001,
158
+ "step": 2200
159
+ },
160
+ {
161
+ "epoch": 255.0,
162
+ "learning_rate": 1e-05,
163
+ "loss": 0.0001,
164
+ "step": 2300
165
+ },
166
+ {
167
+ "epoch": 266.0,
168
+ "learning_rate": 1e-05,
169
+ "loss": 0.0001,
170
+ "step": 2400
171
+ },
172
+ {
173
+ "epoch": 277.0,
174
+ "learning_rate": 1e-05,
175
+ "loss": 0.0,
176
+ "step": 2500
177
+ },
178
+ {
179
+ "epoch": 288.0,
180
+ "learning_rate": 1e-05,
181
+ "loss": 0.0,
182
+ "step": 2600
183
+ },
184
+ {
185
+ "epoch": 299.0,
186
+ "learning_rate": 1e-05,
187
+ "loss": 0.0,
188
+ "step": 2700
189
+ },
190
+ {
191
+ "epoch": 311.0,
192
+ "learning_rate": 1e-05,
193
+ "loss": 0.0,
194
+ "step": 2800
195
+ },
196
+ {
197
+ "epoch": 322.0,
198
+ "learning_rate": 1e-05,
199
+ "loss": 0.0,
200
+ "step": 2900
201
+ },
202
+ {
203
+ "epoch": 333.0,
204
+ "learning_rate": 1e-05,
205
+ "loss": 0.0,
206
+ "step": 3000
207
+ },
208
+ {
209
+ "epoch": 333.0,
210
+ "eval_loss": 0.2674005329608917,
211
+ "eval_runtime": 8.0249,
212
+ "eval_samples_per_second": 2.243,
213
+ "eval_steps_per_second": 0.125,
214
+ "eval_wer": 13.414634146341465,
215
+ "step": 3000
216
+ },
217
+ {
218
+ "epoch": 344.0,
219
+ "learning_rate": 1e-05,
220
+ "loss": 0.0,
221
+ "step": 3100
222
+ },
223
+ {
224
+ "epoch": 355.0,
225
+ "learning_rate": 1e-05,
226
+ "loss": 0.0,
227
+ "step": 3200
228
+ },
229
+ {
230
+ "epoch": 366.0,
231
+ "learning_rate": 1e-05,
232
+ "loss": 0.0,
233
+ "step": 3300
234
+ },
235
+ {
236
+ "epoch": 377.0,
237
+ "learning_rate": 1e-05,
238
+ "loss": 0.0,
239
+ "step": 3400
240
+ },
241
+ {
242
+ "epoch": 388.0,
243
+ "learning_rate": 1e-05,
244
+ "loss": 0.0,
245
+ "step": 3500
246
+ },
247
+ {
248
+ "epoch": 399.0,
249
+ "learning_rate": 1e-05,
250
+ "loss": 0.0,
251
+ "step": 3600
252
+ },
253
+ {
254
+ "epoch": 411.0,
255
+ "learning_rate": 1e-05,
256
+ "loss": 0.0,
257
+ "step": 3700
258
+ },
259
+ {
260
+ "epoch": 422.0,
261
+ "learning_rate": 1e-05,
262
+ "loss": 0.0,
263
+ "step": 3800
264
+ },
265
+ {
266
+ "epoch": 433.0,
267
+ "learning_rate": 1e-05,
268
+ "loss": 0.0,
269
+ "step": 3900
270
+ },
271
+ {
272
+ "epoch": 444.0,
273
+ "learning_rate": 1e-05,
274
+ "loss": 0.0,
275
+ "step": 4000
276
+ },
277
+ {
278
+ "epoch": 444.0,
279
+ "eval_loss": 0.29429247975349426,
280
+ "eval_runtime": 7.7984,
281
+ "eval_samples_per_second": 2.308,
282
+ "eval_steps_per_second": 0.128,
283
+ "eval_wer": 15.853658536585366,
284
+ "step": 4000
285
+ },
286
+ {
287
+ "epoch": 455.0,
288
+ "learning_rate": 1e-05,
289
+ "loss": 0.0,
290
+ "step": 4100
291
+ },
292
+ {
293
+ "epoch": 466.0,
294
+ "learning_rate": 1e-05,
295
+ "loss": 0.0,
296
+ "step": 4200
297
+ },
298
+ {
299
+ "epoch": 477.0,
300
+ "learning_rate": 1e-05,
301
+ "loss": 0.0,
302
+ "step": 4300
303
+ },
304
+ {
305
+ "epoch": 488.0,
306
+ "learning_rate": 1e-05,
307
+ "loss": 0.0,
308
+ "step": 4400
309
+ },
310
+ {
311
+ "epoch": 499.0,
312
+ "learning_rate": 1e-05,
313
+ "loss": 0.0,
314
+ "step": 4500
315
+ },
316
+ {
317
+ "epoch": 511.0,
318
+ "learning_rate": 1e-05,
319
+ "loss": 0.0,
320
+ "step": 4600
321
+ },
322
+ {
323
+ "epoch": 522.0,
324
+ "learning_rate": 1e-05,
325
+ "loss": 0.0,
326
+ "step": 4700
327
+ },
328
+ {
329
+ "epoch": 533.0,
330
+ "learning_rate": 1e-05,
331
+ "loss": 0.0,
332
+ "step": 4800
333
+ },
334
+ {
335
+ "epoch": 544.0,
336
+ "learning_rate": 1e-05,
337
+ "loss": 0.0,
338
+ "step": 4900
339
+ },
340
+ {
341
+ "epoch": 555.0,
342
+ "learning_rate": 1e-05,
343
+ "loss": 0.0,
344
+ "step": 5000
345
+ },
346
+ {
347
+ "epoch": 555.0,
348
+ "eval_loss": 0.31815305352211,
349
+ "eval_runtime": 7.3177,
350
+ "eval_samples_per_second": 2.46,
351
+ "eval_steps_per_second": 0.137,
352
+ "eval_wer": 17.073170731707318,
353
+ "step": 5000
354
+ },
355
+ {
356
+ "epoch": 566.0,
357
+ "learning_rate": 1e-05,
358
+ "loss": 0.0,
359
+ "step": 5100
360
+ },
361
+ {
362
+ "epoch": 577.0,
363
+ "learning_rate": 1e-05,
364
+ "loss": 0.0,
365
+ "step": 5200
366
+ },
367
+ {
368
+ "epoch": 588.0,
369
+ "learning_rate": 1e-05,
370
+ "loss": 0.0,
371
+ "step": 5300
372
+ },
373
+ {
374
+ "epoch": 599.0,
375
+ "learning_rate": 1e-05,
376
+ "loss": 0.0,
377
+ "step": 5400
378
+ },
379
+ {
380
+ "epoch": 611.0,
381
+ "learning_rate": 1e-05,
382
+ "loss": 0.0,
383
+ "step": 5500
384
+ },
385
+ {
386
+ "epoch": 622.0,
387
+ "learning_rate": 1e-05,
388
+ "loss": 0.0,
389
+ "step": 5600
390
+ },
391
+ {
392
+ "epoch": 633.0,
393
+ "learning_rate": 1e-05,
394
+ "loss": 0.0,
395
+ "step": 5700
396
+ },
397
+ {
398
+ "epoch": 644.0,
399
+ "learning_rate": 1e-05,
400
+ "loss": 0.0,
401
+ "step": 5800
402
+ },
403
+ {
404
+ "epoch": 655.0,
405
+ "learning_rate": 1e-05,
406
+ "loss": 0.0,
407
+ "step": 5900
408
+ },
409
+ {
410
+ "epoch": 666.0,
411
+ "learning_rate": 1e-05,
412
+ "loss": 0.0,
413
+ "step": 6000
414
+ },
415
+ {
416
+ "epoch": 666.0,
417
+ "eval_loss": 0.35010817646980286,
418
+ "eval_runtime": 7.7729,
419
+ "eval_samples_per_second": 2.316,
420
+ "eval_steps_per_second": 0.129,
421
+ "eval_wer": 18.902439024390244,
422
+ "step": 6000
423
+ },
424
+ {
425
+ "epoch": 677.0,
426
+ "learning_rate": 1e-05,
427
+ "loss": 0.0,
428
+ "step": 6100
429
+ },
430
+ {
431
+ "epoch": 688.0,
432
+ "learning_rate": 1e-05,
433
+ "loss": 0.0,
434
+ "step": 6200
435
+ },
436
+ {
437
+ "epoch": 699.0,
438
+ "learning_rate": 1e-05,
439
+ "loss": 0.0,
440
+ "step": 6300
441
+ },
442
+ {
443
+ "epoch": 711.0,
444
+ "learning_rate": 1e-05,
445
+ "loss": 0.0,
446
+ "step": 6400
447
+ },
448
+ {
449
+ "epoch": 722.0,
450
+ "learning_rate": 1e-05,
451
+ "loss": 0.0,
452
+ "step": 6500
453
+ },
454
+ {
455
+ "epoch": 733.0,
456
+ "learning_rate": 1e-05,
457
+ "loss": 0.0,
458
+ "step": 6600
459
+ },
460
+ {
461
+ "epoch": 744.0,
462
+ "learning_rate": 1e-05,
463
+ "loss": 0.0,
464
+ "step": 6700
465
+ },
466
+ {
467
+ "epoch": 755.0,
468
+ "learning_rate": 1e-05,
469
+ "loss": 0.0,
470
+ "step": 6800
471
+ },
472
+ {
473
+ "epoch": 766.0,
474
+ "learning_rate": 1e-05,
475
+ "loss": 0.0,
476
+ "step": 6900
477
+ },
478
+ {
479
+ "epoch": 777.0,
480
+ "learning_rate": 1e-05,
481
+ "loss": 0.0,
482
+ "step": 7000
483
+ },
484
+ {
485
+ "epoch": 777.0,
486
+ "eval_loss": 0.37315306067466736,
487
+ "eval_runtime": 8.5166,
488
+ "eval_samples_per_second": 2.114,
489
+ "eval_steps_per_second": 0.117,
490
+ "eval_wer": 16.463414634146343,
491
+ "step": 7000
492
+ }
493
+ ],
494
+ "max_steps": 10000,
495
+ "num_train_epochs": 9223372036854775807,
496
+ "total_flos": 1.372229966757888e+19,
497
+ "trial_name": null,
498
+ "trial_params": null
499
+ }
checkpoint-7000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c36b35e37212c7c09dbca6abb82e5ea958e60ecae5b6ed572ed342d641b50f3
3
+ size 3643
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c0dc7da7b9d3e85c795530ca30bcc712fbed7b068d03c6396d428dab509f555
3
  size 290458721
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5bcae64ae2a6ac8902c7cafa598f375ad75277df9c489d88a2605619107181f
3
  size 290458721
runs/Dec29_11-05-20_srv/events.out.tfevents.1672311942.srv.13534.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:159a748f06382714106268f595ee594292278848682b82adf3b6420e05753fb7
3
- size 15756
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e63352a001220511e0e9b541b0d0a1bb7666881c3a23a3124d5e7994e9a63834
3
+ size 17644