humeur commited on
Commit
897b438
1 Parent(s): 820c0fc

Training in progress, step 4000

Browse files
checkpoint-4000/config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "./lab2_id2223",
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": 768,
15
+ "decoder_attention_heads": 12,
16
+ "decoder_ffn_dim": 3072,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 50258,
20
+ "dropout": 0.0,
21
+ "encoder_attention_heads": 12,
22
+ "encoder_ffn_dim": 3072,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 12,
25
+ "eos_token_id": 50257,
26
+ "forced_decoder_ids": null,
27
+ "init_std": 0.02,
28
+ "is_encoder_decoder": true,
29
+ "max_length": 448,
30
+ "max_source_positions": 1500,
31
+ "max_target_positions": 448,
32
+ "model_type": "whisper",
33
+ "num_hidden_layers": 12,
34
+ "num_mel_bins": 80,
35
+ "pad_token_id": 50257,
36
+ "scale_embedding": false,
37
+ "suppress_tokens": [],
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.26.0.dev0",
40
+ "use_cache": false,
41
+ "vocab_size": 51865
42
+ }
checkpoint-4000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdc4aa00da4171dc8ffd5bd2125b265635cc15f79410d2be715884b0fa9fbceb
3
+ size 1934161093
checkpoint-4000/preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-4000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a4bc9cba3cd08052d394628414c4dc529dfd23dc6b6cff9fbbbf823cdad791d
3
+ size 967102601
checkpoint-4000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99d9b901fa1744b001e6e6d6fb39b1cec852c15ecf4468ad398014cd4e7bccf6
3
+ size 14575
checkpoint-4000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bd4364e1615042c8cfee389cc8e3b78fd57173186257c1c246601e07d8413a7
3
+ size 557
checkpoint-4000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:704b1a0b049fec0acafde69e4d157128ccfe20ff6f043233156342121f536311
3
+ size 627
checkpoint-4000/trainer_state.json ADDED
@@ -0,0 +1,1012 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 19.11929903392496,
3
+ "best_model_checkpoint": "./lab2_id2223/checkpoint-4000",
4
+ "epoch": 5.174644243208279,
5
+ "global_step": 4000,
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": 4.800000000000001e-07,
13
+ "loss": 0.2225,
14
+ "step": 25
15
+ },
16
+ {
17
+ "epoch": 0.06,
18
+ "learning_rate": 9.800000000000001e-07,
19
+ "loss": 0.2081,
20
+ "step": 50
21
+ },
22
+ {
23
+ "epoch": 0.1,
24
+ "learning_rate": 1.48e-06,
25
+ "loss": 0.1992,
26
+ "step": 75
27
+ },
28
+ {
29
+ "epoch": 0.13,
30
+ "learning_rate": 1.98e-06,
31
+ "loss": 0.2014,
32
+ "step": 100
33
+ },
34
+ {
35
+ "epoch": 0.16,
36
+ "learning_rate": 2.4800000000000004e-06,
37
+ "loss": 0.1914,
38
+ "step": 125
39
+ },
40
+ {
41
+ "epoch": 0.19,
42
+ "learning_rate": 2.9800000000000003e-06,
43
+ "loss": 0.1978,
44
+ "step": 150
45
+ },
46
+ {
47
+ "epoch": 0.23,
48
+ "learning_rate": 3.48e-06,
49
+ "loss": 0.1988,
50
+ "step": 175
51
+ },
52
+ {
53
+ "epoch": 0.26,
54
+ "learning_rate": 3.980000000000001e-06,
55
+ "loss": 0.2153,
56
+ "step": 200
57
+ },
58
+ {
59
+ "epoch": 0.29,
60
+ "learning_rate": 4.48e-06,
61
+ "loss": 0.1907,
62
+ "step": 225
63
+ },
64
+ {
65
+ "epoch": 0.32,
66
+ "learning_rate": 4.980000000000001e-06,
67
+ "loss": 0.1976,
68
+ "step": 250
69
+ },
70
+ {
71
+ "epoch": 0.36,
72
+ "learning_rate": 5.480000000000001e-06,
73
+ "loss": 0.2256,
74
+ "step": 275
75
+ },
76
+ {
77
+ "epoch": 0.39,
78
+ "learning_rate": 5.98e-06,
79
+ "loss": 0.1805,
80
+ "step": 300
81
+ },
82
+ {
83
+ "epoch": 0.42,
84
+ "learning_rate": 6.480000000000001e-06,
85
+ "loss": 0.2144,
86
+ "step": 325
87
+ },
88
+ {
89
+ "epoch": 0.45,
90
+ "learning_rate": 6.98e-06,
91
+ "loss": 0.195,
92
+ "step": 350
93
+ },
94
+ {
95
+ "epoch": 0.49,
96
+ "learning_rate": 7.48e-06,
97
+ "loss": 0.2053,
98
+ "step": 375
99
+ },
100
+ {
101
+ "epoch": 0.52,
102
+ "learning_rate": 7.980000000000002e-06,
103
+ "loss": 0.2157,
104
+ "step": 400
105
+ },
106
+ {
107
+ "epoch": 0.55,
108
+ "learning_rate": 8.48e-06,
109
+ "loss": 0.2062,
110
+ "step": 425
111
+ },
112
+ {
113
+ "epoch": 0.58,
114
+ "learning_rate": 8.96e-06,
115
+ "loss": 0.1979,
116
+ "step": 450
117
+ },
118
+ {
119
+ "epoch": 0.61,
120
+ "learning_rate": 9.460000000000001e-06,
121
+ "loss": 0.1868,
122
+ "step": 475
123
+ },
124
+ {
125
+ "epoch": 0.65,
126
+ "learning_rate": 9.960000000000001e-06,
127
+ "loss": 0.2211,
128
+ "step": 500
129
+ },
130
+ {
131
+ "epoch": 0.68,
132
+ "learning_rate": 9.934285714285715e-06,
133
+ "loss": 0.2049,
134
+ "step": 525
135
+ },
136
+ {
137
+ "epoch": 0.71,
138
+ "learning_rate": 9.862857142857144e-06,
139
+ "loss": 0.1989,
140
+ "step": 550
141
+ },
142
+ {
143
+ "epoch": 0.74,
144
+ "learning_rate": 9.791428571428571e-06,
145
+ "loss": 0.2127,
146
+ "step": 575
147
+ },
148
+ {
149
+ "epoch": 0.78,
150
+ "learning_rate": 9.72e-06,
151
+ "loss": 0.2261,
152
+ "step": 600
153
+ },
154
+ {
155
+ "epoch": 0.81,
156
+ "learning_rate": 9.648571428571429e-06,
157
+ "loss": 0.2081,
158
+ "step": 625
159
+ },
160
+ {
161
+ "epoch": 0.84,
162
+ "learning_rate": 9.577142857142858e-06,
163
+ "loss": 0.2281,
164
+ "step": 650
165
+ },
166
+ {
167
+ "epoch": 0.87,
168
+ "learning_rate": 9.505714285714287e-06,
169
+ "loss": 0.204,
170
+ "step": 675
171
+ },
172
+ {
173
+ "epoch": 0.91,
174
+ "learning_rate": 9.434285714285714e-06,
175
+ "loss": 0.2053,
176
+ "step": 700
177
+ },
178
+ {
179
+ "epoch": 0.94,
180
+ "learning_rate": 9.362857142857143e-06,
181
+ "loss": 0.2208,
182
+ "step": 725
183
+ },
184
+ {
185
+ "epoch": 0.97,
186
+ "learning_rate": 9.291428571428572e-06,
187
+ "loss": 0.2076,
188
+ "step": 750
189
+ },
190
+ {
191
+ "epoch": 1.0,
192
+ "learning_rate": 9.220000000000002e-06,
193
+ "loss": 0.1941,
194
+ "step": 775
195
+ },
196
+ {
197
+ "epoch": 1.03,
198
+ "learning_rate": 9.148571428571429e-06,
199
+ "loss": 0.0949,
200
+ "step": 800
201
+ },
202
+ {
203
+ "epoch": 1.07,
204
+ "learning_rate": 9.077142857142858e-06,
205
+ "loss": 0.1011,
206
+ "step": 825
207
+ },
208
+ {
209
+ "epoch": 1.1,
210
+ "learning_rate": 9.005714285714287e-06,
211
+ "loss": 0.1078,
212
+ "step": 850
213
+ },
214
+ {
215
+ "epoch": 1.13,
216
+ "learning_rate": 8.934285714285716e-06,
217
+ "loss": 0.0956,
218
+ "step": 875
219
+ },
220
+ {
221
+ "epoch": 1.16,
222
+ "learning_rate": 8.862857142857143e-06,
223
+ "loss": 0.1093,
224
+ "step": 900
225
+ },
226
+ {
227
+ "epoch": 1.2,
228
+ "learning_rate": 8.791428571428572e-06,
229
+ "loss": 0.1133,
230
+ "step": 925
231
+ },
232
+ {
233
+ "epoch": 1.23,
234
+ "learning_rate": 8.720000000000001e-06,
235
+ "loss": 0.1128,
236
+ "step": 950
237
+ },
238
+ {
239
+ "epoch": 1.26,
240
+ "learning_rate": 8.64857142857143e-06,
241
+ "loss": 0.1068,
242
+ "step": 975
243
+ },
244
+ {
245
+ "epoch": 1.29,
246
+ "learning_rate": 8.577142857142858e-06,
247
+ "loss": 0.1015,
248
+ "step": 1000
249
+ },
250
+ {
251
+ "epoch": 1.29,
252
+ "eval_loss": 0.2880099415779114,
253
+ "eval_runtime": 2234.5268,
254
+ "eval_samples_per_second": 1.343,
255
+ "eval_steps_per_second": 0.168,
256
+ "eval_wer": 20.41339024938216,
257
+ "step": 1000
258
+ },
259
+ {
260
+ "epoch": 1.33,
261
+ "learning_rate": 8.505714285714287e-06,
262
+ "loss": 0.1154,
263
+ "step": 1025
264
+ },
265
+ {
266
+ "epoch": 1.36,
267
+ "learning_rate": 8.434285714285716e-06,
268
+ "loss": 0.1035,
269
+ "step": 1050
270
+ },
271
+ {
272
+ "epoch": 1.39,
273
+ "learning_rate": 8.362857142857143e-06,
274
+ "loss": 0.103,
275
+ "step": 1075
276
+ },
277
+ {
278
+ "epoch": 1.42,
279
+ "learning_rate": 8.291428571428572e-06,
280
+ "loss": 0.1089,
281
+ "step": 1100
282
+ },
283
+ {
284
+ "epoch": 1.46,
285
+ "learning_rate": 8.220000000000001e-06,
286
+ "loss": 0.0986,
287
+ "step": 1125
288
+ },
289
+ {
290
+ "epoch": 1.49,
291
+ "learning_rate": 8.148571428571428e-06,
292
+ "loss": 0.0999,
293
+ "step": 1150
294
+ },
295
+ {
296
+ "epoch": 1.52,
297
+ "learning_rate": 8.077142857142857e-06,
298
+ "loss": 0.1075,
299
+ "step": 1175
300
+ },
301
+ {
302
+ "epoch": 1.55,
303
+ "learning_rate": 8.005714285714286e-06,
304
+ "loss": 0.1155,
305
+ "step": 1200
306
+ },
307
+ {
308
+ "epoch": 1.58,
309
+ "learning_rate": 7.934285714285715e-06,
310
+ "loss": 0.0978,
311
+ "step": 1225
312
+ },
313
+ {
314
+ "epoch": 1.62,
315
+ "learning_rate": 7.862857142857143e-06,
316
+ "loss": 0.1035,
317
+ "step": 1250
318
+ },
319
+ {
320
+ "epoch": 1.65,
321
+ "learning_rate": 7.791428571428572e-06,
322
+ "loss": 0.1085,
323
+ "step": 1275
324
+ },
325
+ {
326
+ "epoch": 1.68,
327
+ "learning_rate": 7.72e-06,
328
+ "loss": 0.1008,
329
+ "step": 1300
330
+ },
331
+ {
332
+ "epoch": 1.71,
333
+ "learning_rate": 7.64857142857143e-06,
334
+ "loss": 0.1083,
335
+ "step": 1325
336
+ },
337
+ {
338
+ "epoch": 1.75,
339
+ "learning_rate": 7.577142857142857e-06,
340
+ "loss": 0.0918,
341
+ "step": 1350
342
+ },
343
+ {
344
+ "epoch": 1.78,
345
+ "learning_rate": 7.505714285714286e-06,
346
+ "loss": 0.1124,
347
+ "step": 1375
348
+ },
349
+ {
350
+ "epoch": 1.81,
351
+ "learning_rate": 7.434285714285715e-06,
352
+ "loss": 0.0997,
353
+ "step": 1400
354
+ },
355
+ {
356
+ "epoch": 1.84,
357
+ "learning_rate": 7.362857142857144e-06,
358
+ "loss": 0.0954,
359
+ "step": 1425
360
+ },
361
+ {
362
+ "epoch": 1.88,
363
+ "learning_rate": 7.291428571428571e-06,
364
+ "loss": 0.0946,
365
+ "step": 1450
366
+ },
367
+ {
368
+ "epoch": 1.91,
369
+ "learning_rate": 7.22e-06,
370
+ "loss": 0.1,
371
+ "step": 1475
372
+ },
373
+ {
374
+ "epoch": 1.94,
375
+ "learning_rate": 7.148571428571429e-06,
376
+ "loss": 0.0961,
377
+ "step": 1500
378
+ },
379
+ {
380
+ "epoch": 1.97,
381
+ "learning_rate": 7.077142857142858e-06,
382
+ "loss": 0.1074,
383
+ "step": 1525
384
+ },
385
+ {
386
+ "epoch": 2.01,
387
+ "learning_rate": 7.0057142857142865e-06,
388
+ "loss": 0.0821,
389
+ "step": 1550
390
+ },
391
+ {
392
+ "epoch": 2.04,
393
+ "learning_rate": 6.934285714285715e-06,
394
+ "loss": 0.0392,
395
+ "step": 1575
396
+ },
397
+ {
398
+ "epoch": 2.07,
399
+ "learning_rate": 6.862857142857144e-06,
400
+ "loss": 0.0371,
401
+ "step": 1600
402
+ },
403
+ {
404
+ "epoch": 2.1,
405
+ "learning_rate": 6.791428571428572e-06,
406
+ "loss": 0.0441,
407
+ "step": 1625
408
+ },
409
+ {
410
+ "epoch": 2.13,
411
+ "learning_rate": 6.720000000000001e-06,
412
+ "loss": 0.0416,
413
+ "step": 1650
414
+ },
415
+ {
416
+ "epoch": 2.17,
417
+ "learning_rate": 6.648571428571429e-06,
418
+ "loss": 0.0434,
419
+ "step": 1675
420
+ },
421
+ {
422
+ "epoch": 2.2,
423
+ "learning_rate": 6.577142857142857e-06,
424
+ "loss": 0.0457,
425
+ "step": 1700
426
+ },
427
+ {
428
+ "epoch": 2.23,
429
+ "learning_rate": 6.505714285714286e-06,
430
+ "loss": 0.0376,
431
+ "step": 1725
432
+ },
433
+ {
434
+ "epoch": 2.26,
435
+ "learning_rate": 6.434285714285715e-06,
436
+ "loss": 0.039,
437
+ "step": 1750
438
+ },
439
+ {
440
+ "epoch": 2.3,
441
+ "learning_rate": 6.3628571428571426e-06,
442
+ "loss": 0.0405,
443
+ "step": 1775
444
+ },
445
+ {
446
+ "epoch": 2.33,
447
+ "learning_rate": 6.2914285714285716e-06,
448
+ "loss": 0.0352,
449
+ "step": 1800
450
+ },
451
+ {
452
+ "epoch": 2.36,
453
+ "learning_rate": 6.220000000000001e-06,
454
+ "loss": 0.0397,
455
+ "step": 1825
456
+ },
457
+ {
458
+ "epoch": 2.39,
459
+ "learning_rate": 6.14857142857143e-06,
460
+ "loss": 0.0396,
461
+ "step": 1850
462
+ },
463
+ {
464
+ "epoch": 2.43,
465
+ "learning_rate": 6.077142857142858e-06,
466
+ "loss": 0.0388,
467
+ "step": 1875
468
+ },
469
+ {
470
+ "epoch": 2.46,
471
+ "learning_rate": 6.005714285714286e-06,
472
+ "loss": 0.038,
473
+ "step": 1900
474
+ },
475
+ {
476
+ "epoch": 2.49,
477
+ "learning_rate": 5.934285714285715e-06,
478
+ "loss": 0.0391,
479
+ "step": 1925
480
+ },
481
+ {
482
+ "epoch": 2.52,
483
+ "learning_rate": 5.862857142857143e-06,
484
+ "loss": 0.0358,
485
+ "step": 1950
486
+ },
487
+ {
488
+ "epoch": 2.55,
489
+ "learning_rate": 5.791428571428572e-06,
490
+ "loss": 0.0399,
491
+ "step": 1975
492
+ },
493
+ {
494
+ "epoch": 2.59,
495
+ "learning_rate": 5.72e-06,
496
+ "loss": 0.0387,
497
+ "step": 2000
498
+ },
499
+ {
500
+ "epoch": 2.59,
501
+ "eval_loss": 0.29592692852020264,
502
+ "eval_runtime": 2465.3193,
503
+ "eval_samples_per_second": 1.217,
504
+ "eval_steps_per_second": 0.152,
505
+ "eval_wer": 19.680970568411592,
506
+ "step": 2000
507
+ },
508
+ {
509
+ "epoch": 2.62,
510
+ "learning_rate": 5.6485714285714285e-06,
511
+ "loss": 0.0349,
512
+ "step": 2025
513
+ },
514
+ {
515
+ "epoch": 2.65,
516
+ "learning_rate": 5.5771428571428575e-06,
517
+ "loss": 0.0395,
518
+ "step": 2050
519
+ },
520
+ {
521
+ "epoch": 2.68,
522
+ "learning_rate": 5.5057142857142865e-06,
523
+ "loss": 0.0403,
524
+ "step": 2075
525
+ },
526
+ {
527
+ "epoch": 2.72,
528
+ "learning_rate": 5.4342857142857155e-06,
529
+ "loss": 0.043,
530
+ "step": 2100
531
+ },
532
+ {
533
+ "epoch": 2.75,
534
+ "learning_rate": 5.362857142857143e-06,
535
+ "loss": 0.0443,
536
+ "step": 2125
537
+ },
538
+ {
539
+ "epoch": 2.78,
540
+ "learning_rate": 5.291428571428572e-06,
541
+ "loss": 0.0421,
542
+ "step": 2150
543
+ },
544
+ {
545
+ "epoch": 2.81,
546
+ "learning_rate": 5.220000000000001e-06,
547
+ "loss": 0.0354,
548
+ "step": 2175
549
+ },
550
+ {
551
+ "epoch": 2.85,
552
+ "learning_rate": 5.14857142857143e-06,
553
+ "loss": 0.0393,
554
+ "step": 2200
555
+ },
556
+ {
557
+ "epoch": 2.88,
558
+ "learning_rate": 5.077142857142857e-06,
559
+ "loss": 0.0371,
560
+ "step": 2225
561
+ },
562
+ {
563
+ "epoch": 2.91,
564
+ "learning_rate": 5.005714285714286e-06,
565
+ "loss": 0.0354,
566
+ "step": 2250
567
+ },
568
+ {
569
+ "epoch": 2.94,
570
+ "learning_rate": 4.934285714285715e-06,
571
+ "loss": 0.043,
572
+ "step": 2275
573
+ },
574
+ {
575
+ "epoch": 2.98,
576
+ "learning_rate": 4.862857142857143e-06,
577
+ "loss": 0.0356,
578
+ "step": 2300
579
+ },
580
+ {
581
+ "epoch": 3.01,
582
+ "learning_rate": 4.7914285714285715e-06,
583
+ "loss": 0.038,
584
+ "step": 2325
585
+ },
586
+ {
587
+ "epoch": 3.04,
588
+ "learning_rate": 4.7200000000000005e-06,
589
+ "loss": 0.0162,
590
+ "step": 2350
591
+ },
592
+ {
593
+ "epoch": 3.07,
594
+ "learning_rate": 4.648571428571429e-06,
595
+ "loss": 0.0175,
596
+ "step": 2375
597
+ },
598
+ {
599
+ "epoch": 3.1,
600
+ "learning_rate": 4.577142857142858e-06,
601
+ "loss": 0.0143,
602
+ "step": 2400
603
+ },
604
+ {
605
+ "epoch": 3.14,
606
+ "learning_rate": 4.505714285714286e-06,
607
+ "loss": 0.0165,
608
+ "step": 2425
609
+ },
610
+ {
611
+ "epoch": 3.17,
612
+ "learning_rate": 4.434285714285715e-06,
613
+ "loss": 0.0176,
614
+ "step": 2450
615
+ },
616
+ {
617
+ "epoch": 3.2,
618
+ "learning_rate": 4.362857142857143e-06,
619
+ "loss": 0.0148,
620
+ "step": 2475
621
+ },
622
+ {
623
+ "epoch": 3.23,
624
+ "learning_rate": 4.291428571428572e-06,
625
+ "loss": 0.0157,
626
+ "step": 2500
627
+ },
628
+ {
629
+ "epoch": 3.27,
630
+ "learning_rate": 4.22e-06,
631
+ "loss": 0.0147,
632
+ "step": 2525
633
+ },
634
+ {
635
+ "epoch": 3.3,
636
+ "learning_rate": 4.151428571428572e-06,
637
+ "loss": 0.016,
638
+ "step": 2550
639
+ },
640
+ {
641
+ "epoch": 3.33,
642
+ "learning_rate": 4.08e-06,
643
+ "loss": 0.0153,
644
+ "step": 2575
645
+ },
646
+ {
647
+ "epoch": 3.36,
648
+ "learning_rate": 4.008571428571429e-06,
649
+ "loss": 0.0135,
650
+ "step": 2600
651
+ },
652
+ {
653
+ "epoch": 3.4,
654
+ "learning_rate": 3.937142857142858e-06,
655
+ "loss": 0.0194,
656
+ "step": 2625
657
+ },
658
+ {
659
+ "epoch": 3.43,
660
+ "learning_rate": 3.865714285714286e-06,
661
+ "loss": 0.0155,
662
+ "step": 2650
663
+ },
664
+ {
665
+ "epoch": 3.46,
666
+ "learning_rate": 3.7942857142857147e-06,
667
+ "loss": 0.0127,
668
+ "step": 2675
669
+ },
670
+ {
671
+ "epoch": 3.49,
672
+ "learning_rate": 3.722857142857143e-06,
673
+ "loss": 0.0148,
674
+ "step": 2700
675
+ },
676
+ {
677
+ "epoch": 3.53,
678
+ "learning_rate": 3.651428571428572e-06,
679
+ "loss": 0.0167,
680
+ "step": 2725
681
+ },
682
+ {
683
+ "epoch": 3.56,
684
+ "learning_rate": 3.58e-06,
685
+ "loss": 0.0181,
686
+ "step": 2750
687
+ },
688
+ {
689
+ "epoch": 3.59,
690
+ "learning_rate": 3.508571428571429e-06,
691
+ "loss": 0.0155,
692
+ "step": 2775
693
+ },
694
+ {
695
+ "epoch": 3.62,
696
+ "learning_rate": 3.437142857142857e-06,
697
+ "loss": 0.0157,
698
+ "step": 2800
699
+ },
700
+ {
701
+ "epoch": 3.65,
702
+ "learning_rate": 3.3657142857142862e-06,
703
+ "loss": 0.0142,
704
+ "step": 2825
705
+ },
706
+ {
707
+ "epoch": 3.69,
708
+ "learning_rate": 3.2942857142857144e-06,
709
+ "loss": 0.0156,
710
+ "step": 2850
711
+ },
712
+ {
713
+ "epoch": 3.72,
714
+ "learning_rate": 3.222857142857143e-06,
715
+ "loss": 0.0156,
716
+ "step": 2875
717
+ },
718
+ {
719
+ "epoch": 3.75,
720
+ "learning_rate": 3.151428571428572e-06,
721
+ "loss": 0.0189,
722
+ "step": 2900
723
+ },
724
+ {
725
+ "epoch": 3.78,
726
+ "learning_rate": 3.08e-06,
727
+ "loss": 0.0133,
728
+ "step": 2925
729
+ },
730
+ {
731
+ "epoch": 3.82,
732
+ "learning_rate": 3.008571428571429e-06,
733
+ "loss": 0.0151,
734
+ "step": 2950
735
+ },
736
+ {
737
+ "epoch": 3.85,
738
+ "learning_rate": 2.9371428571428573e-06,
739
+ "loss": 0.0176,
740
+ "step": 2975
741
+ },
742
+ {
743
+ "epoch": 3.88,
744
+ "learning_rate": 2.865714285714286e-06,
745
+ "loss": 0.0126,
746
+ "step": 3000
747
+ },
748
+ {
749
+ "epoch": 3.88,
750
+ "eval_loss": 0.3103422224521637,
751
+ "eval_runtime": 2600.9697,
752
+ "eval_samples_per_second": 1.153,
753
+ "eval_steps_per_second": 0.144,
754
+ "eval_wer": 19.299033924960682,
755
+ "step": 3000
756
+ },
757
+ {
758
+ "epoch": 3.91,
759
+ "learning_rate": 2.7942857142857145e-06,
760
+ "loss": 0.0173,
761
+ "step": 3025
762
+ },
763
+ {
764
+ "epoch": 3.95,
765
+ "learning_rate": 2.722857142857143e-06,
766
+ "loss": 0.0144,
767
+ "step": 3050
768
+ },
769
+ {
770
+ "epoch": 3.98,
771
+ "learning_rate": 2.6514285714285713e-06,
772
+ "loss": 0.016,
773
+ "step": 3075
774
+ },
775
+ {
776
+ "epoch": 4.01,
777
+ "learning_rate": 2.5800000000000003e-06,
778
+ "loss": 0.012,
779
+ "step": 3100
780
+ },
781
+ {
782
+ "epoch": 4.04,
783
+ "learning_rate": 2.5085714285714285e-06,
784
+ "loss": 0.0071,
785
+ "step": 3125
786
+ },
787
+ {
788
+ "epoch": 4.08,
789
+ "learning_rate": 2.4371428571428575e-06,
790
+ "loss": 0.0069,
791
+ "step": 3150
792
+ },
793
+ {
794
+ "epoch": 4.11,
795
+ "learning_rate": 2.365714285714286e-06,
796
+ "loss": 0.0069,
797
+ "step": 3175
798
+ },
799
+ {
800
+ "epoch": 4.14,
801
+ "learning_rate": 2.2942857142857146e-06,
802
+ "loss": 0.0061,
803
+ "step": 3200
804
+ },
805
+ {
806
+ "epoch": 4.17,
807
+ "learning_rate": 2.222857142857143e-06,
808
+ "loss": 0.0052,
809
+ "step": 3225
810
+ },
811
+ {
812
+ "epoch": 4.2,
813
+ "learning_rate": 2.1514285714285714e-06,
814
+ "loss": 0.0064,
815
+ "step": 3250
816
+ },
817
+ {
818
+ "epoch": 4.24,
819
+ "learning_rate": 2.08e-06,
820
+ "loss": 0.0062,
821
+ "step": 3275
822
+ },
823
+ {
824
+ "epoch": 4.27,
825
+ "learning_rate": 2.0085714285714286e-06,
826
+ "loss": 0.006,
827
+ "step": 3300
828
+ },
829
+ {
830
+ "epoch": 4.3,
831
+ "learning_rate": 1.9371428571428576e-06,
832
+ "loss": 0.0073,
833
+ "step": 3325
834
+ },
835
+ {
836
+ "epoch": 4.33,
837
+ "learning_rate": 1.865714285714286e-06,
838
+ "loss": 0.0055,
839
+ "step": 3350
840
+ },
841
+ {
842
+ "epoch": 4.37,
843
+ "learning_rate": 1.7942857142857146e-06,
844
+ "loss": 0.0059,
845
+ "step": 3375
846
+ },
847
+ {
848
+ "epoch": 4.4,
849
+ "learning_rate": 1.7228571428571432e-06,
850
+ "loss": 0.0051,
851
+ "step": 3400
852
+ },
853
+ {
854
+ "epoch": 4.43,
855
+ "learning_rate": 1.6514285714285715e-06,
856
+ "loss": 0.0057,
857
+ "step": 3425
858
+ },
859
+ {
860
+ "epoch": 4.46,
861
+ "learning_rate": 1.5800000000000001e-06,
862
+ "loss": 0.0051,
863
+ "step": 3450
864
+ },
865
+ {
866
+ "epoch": 4.5,
867
+ "learning_rate": 1.5085714285714287e-06,
868
+ "loss": 0.0054,
869
+ "step": 3475
870
+ },
871
+ {
872
+ "epoch": 4.53,
873
+ "learning_rate": 1.4371428571428573e-06,
874
+ "loss": 0.0055,
875
+ "step": 3500
876
+ },
877
+ {
878
+ "epoch": 4.56,
879
+ "learning_rate": 1.3657142857142857e-06,
880
+ "loss": 0.0059,
881
+ "step": 3525
882
+ },
883
+ {
884
+ "epoch": 4.59,
885
+ "learning_rate": 1.2942857142857143e-06,
886
+ "loss": 0.0049,
887
+ "step": 3550
888
+ },
889
+ {
890
+ "epoch": 4.62,
891
+ "learning_rate": 1.222857142857143e-06,
892
+ "loss": 0.0049,
893
+ "step": 3575
894
+ },
895
+ {
896
+ "epoch": 4.66,
897
+ "learning_rate": 1.1514285714285714e-06,
898
+ "loss": 0.0056,
899
+ "step": 3600
900
+ },
901
+ {
902
+ "epoch": 4.69,
903
+ "learning_rate": 1.08e-06,
904
+ "loss": 0.0066,
905
+ "step": 3625
906
+ },
907
+ {
908
+ "epoch": 4.72,
909
+ "learning_rate": 1.0085714285714286e-06,
910
+ "loss": 0.0056,
911
+ "step": 3650
912
+ },
913
+ {
914
+ "epoch": 4.75,
915
+ "learning_rate": 9.371428571428571e-07,
916
+ "loss": 0.0052,
917
+ "step": 3675
918
+ },
919
+ {
920
+ "epoch": 4.79,
921
+ "learning_rate": 8.657142857142858e-07,
922
+ "loss": 0.0056,
923
+ "step": 3700
924
+ },
925
+ {
926
+ "epoch": 4.82,
927
+ "learning_rate": 7.942857142857144e-07,
928
+ "loss": 0.0047,
929
+ "step": 3725
930
+ },
931
+ {
932
+ "epoch": 4.85,
933
+ "learning_rate": 7.228571428571429e-07,
934
+ "loss": 0.0061,
935
+ "step": 3750
936
+ },
937
+ {
938
+ "epoch": 4.88,
939
+ "learning_rate": 6.514285714285715e-07,
940
+ "loss": 0.0049,
941
+ "step": 3775
942
+ },
943
+ {
944
+ "epoch": 4.92,
945
+ "learning_rate": 5.800000000000001e-07,
946
+ "loss": 0.0041,
947
+ "step": 3800
948
+ },
949
+ {
950
+ "epoch": 4.95,
951
+ "learning_rate": 5.085714285714286e-07,
952
+ "loss": 0.005,
953
+ "step": 3825
954
+ },
955
+ {
956
+ "epoch": 4.98,
957
+ "learning_rate": 4.371428571428572e-07,
958
+ "loss": 0.0058,
959
+ "step": 3850
960
+ },
961
+ {
962
+ "epoch": 5.01,
963
+ "learning_rate": 3.657142857142858e-07,
964
+ "loss": 0.0038,
965
+ "step": 3875
966
+ },
967
+ {
968
+ "epoch": 5.05,
969
+ "learning_rate": 2.942857142857143e-07,
970
+ "loss": 0.0031,
971
+ "step": 3900
972
+ },
973
+ {
974
+ "epoch": 5.08,
975
+ "learning_rate": 2.228571428571429e-07,
976
+ "loss": 0.0031,
977
+ "step": 3925
978
+ },
979
+ {
980
+ "epoch": 5.11,
981
+ "learning_rate": 1.5142857142857144e-07,
982
+ "loss": 0.0033,
983
+ "step": 3950
984
+ },
985
+ {
986
+ "epoch": 5.14,
987
+ "learning_rate": 8e-08,
988
+ "loss": 0.0033,
989
+ "step": 3975
990
+ },
991
+ {
992
+ "epoch": 5.17,
993
+ "learning_rate": 8.571428571428572e-09,
994
+ "loss": 0.0035,
995
+ "step": 4000
996
+ },
997
+ {
998
+ "epoch": 5.17,
999
+ "eval_loss": 0.3309653699398041,
1000
+ "eval_runtime": 2605.9672,
1001
+ "eval_samples_per_second": 1.151,
1002
+ "eval_steps_per_second": 0.144,
1003
+ "eval_wer": 19.11929903392496,
1004
+ "step": 4000
1005
+ }
1006
+ ],
1007
+ "max_steps": 4000,
1008
+ "num_train_epochs": 6,
1009
+ "total_flos": 1.84579222044672e+19,
1010
+ "trial_name": null,
1011
+ "trial_params": null
1012
+ }
checkpoint-4000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e11a85f192287523b28a6ac51081f5592c031a75ef62ea9d55008b6f1e2bc48d
3
+ size 3579
preprocessor_config.json CHANGED
@@ -5,7 +5,7 @@
5
  "hop_length": 160,
6
  "mel_filters": [
7
  [
8
- -0.0,
9
  0.02486259490251541,
10
  0.0,
11
  0.0,
 
5
  "hop_length": 160,
6
  "mel_filters": [
7
  [
8
+ 0.0,
9
  0.02486259490251541,
10
  0.0,
11
  0.0,
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a31c2369d10e9f24b6b89d9e0513844a5716b99a443f712639b3a92841f8c6cb
3
  size 967102601
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a4bc9cba3cd08052d394628414c4dc529dfd23dc6b6cff9fbbbf823cdad791d
3
  size 967102601
runs/Dec08_16-23-13_12004c88daaa/1670516632.4447682/events.out.tfevents.1670516632.12004c88daaa.75.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c93dd752daa84344c4bc9b31223c5e366622aa4cbaf3e259f14e33add61021e5
3
+ size 5749
runs/Dec08_16-23-13_12004c88daaa/events.out.tfevents.1670516632.12004c88daaa.75.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d3f19769e9efbae3640e3439cfe07129f2a9af0d8c0d2aa4411755a1e282253
3
+ size 10837
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a547a4035d6a58e0a12161a55b096363471ab528691eaa0b1437f2281f02e977
3
  size 3579
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e11a85f192287523b28a6ac51081f5592c031a75ef62ea9d55008b6f1e2bc48d
3
  size 3579