LBolitho commited on
Commit
b4c601f
·
verified ·
1 Parent(s): 768c823

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "gelu",
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "WhisperForAudioClassification"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": [
10
+ 220,
11
+ 50256
12
+ ],
13
+ "bos_token_id": 50257,
14
+ "classifier_proj_size": 256,
15
+ "d_model": 1280,
16
+ "decoder_attention_heads": 20,
17
+ "decoder_ffn_dim": 5120,
18
+ "decoder_layerdrop": 0.0,
19
+ "decoder_layers": 4,
20
+ "decoder_start_token_id": 50258,
21
+ "dropout": 0.0,
22
+ "dtype": "float32",
23
+ "encoder_attention_heads": 20,
24
+ "encoder_ffn_dim": 5120,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 32,
27
+ "eos_token_id": 50257,
28
+ "init_std": 0.02,
29
+ "is_encoder_decoder": true,
30
+ "mask_feature_length": 10,
31
+ "mask_feature_min_masks": 0,
32
+ "mask_feature_prob": 0.0,
33
+ "mask_time_length": 10,
34
+ "mask_time_min_masks": 2,
35
+ "mask_time_prob": 0.05,
36
+ "max_source_positions": 50,
37
+ "max_target_positions": 448,
38
+ "median_filter_width": 7,
39
+ "model_type": "whisper",
40
+ "num_hidden_layers": 32,
41
+ "num_mel_bins": 128,
42
+ "pad_token_id": 50257,
43
+ "scale_embedding": false,
44
+ "transformers_version": "4.57.1",
45
+ "use_cache": true,
46
+ "use_weighted_layer_sum": false,
47
+ "vocab_size": 51866
48
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:182a210d4f728b8bab148dbc3da03370809b1e34a9903e1523b9b31030f44049
3
+ size 2541820056
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a8d08d8bbef1ce66c38b9ae8b785ff5d5a56beda6776ba974c9c36010a5dfe8
3
+ size 5083449089
preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 128,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "right",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6468252dc5a2985ff1e08329210107b45976ae610a03cbae5464e914893393d
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70cce385c313cbac907fbd19641a17caa3dbfc56b9e5a124b58aff94079b5c89
3
+ size 1465
trainer_state.json ADDED
@@ -0,0 +1,446 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 524,
3
+ "best_metric": 1.0,
4
+ "best_model_checkpoint": "wav2vec2_frog_classifier_sew_d/checkpoint-524",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 524,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.07633587786259542,
14
+ "grad_norm": 11.720809936523438,
15
+ "learning_rate": 2.2900763358778629e-07,
16
+ "loss": 0.7534,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.15267175572519084,
21
+ "grad_norm": 7.877044200897217,
22
+ "learning_rate": 4.834605597964377e-07,
23
+ "loss": 0.6457,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.22900763358778625,
28
+ "grad_norm": 6.020662784576416,
29
+ "learning_rate": 7.37913486005089e-07,
30
+ "loss": 0.5034,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.3053435114503817,
35
+ "grad_norm": 11.8593111038208,
36
+ "learning_rate": 9.923664122137404e-07,
37
+ "loss": 0.3672,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.3816793893129771,
42
+ "grad_norm": 19.238187789916992,
43
+ "learning_rate": 1.246819338422392e-06,
44
+ "loss": 0.3239,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.4580152671755725,
49
+ "grad_norm": 29.412681579589844,
50
+ "learning_rate": 1.5012722646310435e-06,
51
+ "loss": 0.4459,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.5343511450381679,
56
+ "grad_norm": 23.955923080444336,
57
+ "learning_rate": 1.7557251908396948e-06,
58
+ "loss": 0.3843,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.6106870229007634,
63
+ "grad_norm": 33.24127960205078,
64
+ "learning_rate": 2.010178117048346e-06,
65
+ "loss": 0.1934,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.6870229007633588,
70
+ "grad_norm": 8.673649787902832,
71
+ "learning_rate": 2.2646310432569978e-06,
72
+ "loss": 0.2595,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.7633587786259542,
77
+ "grad_norm": 1.9229257106781006,
78
+ "learning_rate": 2.5190839694656487e-06,
79
+ "loss": 0.2245,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.8396946564885496,
84
+ "grad_norm": 13.961572647094727,
85
+ "learning_rate": 2.7735368956743004e-06,
86
+ "loss": 0.2427,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.916030534351145,
91
+ "grad_norm": 5.140026092529297,
92
+ "learning_rate": 3.0279898218829517e-06,
93
+ "loss": 0.1903,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.9923664122137404,
98
+ "grad_norm": 0.7627769112586975,
99
+ "learning_rate": 3.2824427480916034e-06,
100
+ "loss": 0.1594,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 1.0,
105
+ "eval_f1": 0.9463087248322147,
106
+ "eval_fbeta": 0.9778085139994803,
107
+ "eval_loss": 0.08893176913261414,
108
+ "eval_precision": 1.0,
109
+ "eval_recall": 0.8980891719745223,
110
+ "eval_runtime": 13.2655,
111
+ "eval_samples_per_second": 67.468,
112
+ "eval_steps_per_second": 8.443,
113
+ "step": 131
114
+ },
115
+ {
116
+ "epoch": 1.0687022900763359,
117
+ "grad_norm": 0.26965653896331787,
118
+ "learning_rate": 3.5368956743002547e-06,
119
+ "loss": 0.0149,
120
+ "step": 140
121
+ },
122
+ {
123
+ "epoch": 1.1450381679389312,
124
+ "grad_norm": 0.1886816918849945,
125
+ "learning_rate": 3.791348600508906e-06,
126
+ "loss": 0.1284,
127
+ "step": 150
128
+ },
129
+ {
130
+ "epoch": 1.2213740458015268,
131
+ "grad_norm": 36.349491119384766,
132
+ "learning_rate": 4.045801526717557e-06,
133
+ "loss": 0.1203,
134
+ "step": 160
135
+ },
136
+ {
137
+ "epoch": 1.297709923664122,
138
+ "grad_norm": 0.12028646469116211,
139
+ "learning_rate": 4.300254452926209e-06,
140
+ "loss": 0.0069,
141
+ "step": 170
142
+ },
143
+ {
144
+ "epoch": 1.3740458015267176,
145
+ "grad_norm": 0.1057005226612091,
146
+ "learning_rate": 4.554707379134861e-06,
147
+ "loss": 0.0351,
148
+ "step": 180
149
+ },
150
+ {
151
+ "epoch": 1.450381679389313,
152
+ "grad_norm": 0.13390018045902252,
153
+ "learning_rate": 4.8091603053435125e-06,
154
+ "loss": 0.0688,
155
+ "step": 190
156
+ },
157
+ {
158
+ "epoch": 1.5267175572519083,
159
+ "grad_norm": 188.4539031982422,
160
+ "learning_rate": 5.063613231552163e-06,
161
+ "loss": 0.1442,
162
+ "step": 200
163
+ },
164
+ {
165
+ "epoch": 1.6030534351145038,
166
+ "grad_norm": 0.15123459696769714,
167
+ "learning_rate": 5.318066157760815e-06,
168
+ "loss": 0.0292,
169
+ "step": 210
170
+ },
171
+ {
172
+ "epoch": 1.6793893129770994,
173
+ "grad_norm": 0.10352899134159088,
174
+ "learning_rate": 5.572519083969467e-06,
175
+ "loss": 0.0328,
176
+ "step": 220
177
+ },
178
+ {
179
+ "epoch": 1.7557251908396947,
180
+ "grad_norm": 0.2473313808441162,
181
+ "learning_rate": 5.826972010178118e-06,
182
+ "loss": 0.0049,
183
+ "step": 230
184
+ },
185
+ {
186
+ "epoch": 1.83206106870229,
187
+ "grad_norm": 0.10870244354009628,
188
+ "learning_rate": 6.081424936386769e-06,
189
+ "loss": 0.0361,
190
+ "step": 240
191
+ },
192
+ {
193
+ "epoch": 1.9083969465648853,
194
+ "grad_norm": 0.0995243638753891,
195
+ "learning_rate": 6.335877862595419e-06,
196
+ "loss": 0.0387,
197
+ "step": 250
198
+ },
199
+ {
200
+ "epoch": 1.984732824427481,
201
+ "grad_norm": 0.07525702565908432,
202
+ "learning_rate": 6.590330788804071e-06,
203
+ "loss": 0.0037,
204
+ "step": 260
205
+ },
206
+ {
207
+ "epoch": 2.0,
208
+ "eval_f1": 0.9968253968253968,
209
+ "eval_fbeta": 0.9949302118129412,
210
+ "eval_loss": 0.00947254616767168,
211
+ "eval_precision": 0.9936708860759493,
212
+ "eval_recall": 1.0,
213
+ "eval_runtime": 13.3328,
214
+ "eval_samples_per_second": 67.128,
215
+ "eval_steps_per_second": 8.4,
216
+ "step": 262
217
+ },
218
+ {
219
+ "epoch": 2.0610687022900764,
220
+ "grad_norm": 0.06131064146757126,
221
+ "learning_rate": 6.844783715012723e-06,
222
+ "loss": 0.0032,
223
+ "step": 270
224
+ },
225
+ {
226
+ "epoch": 2.1374045801526718,
227
+ "grad_norm": 0.05664181336760521,
228
+ "learning_rate": 7.0992366412213746e-06,
229
+ "loss": 0.0029,
230
+ "step": 280
231
+ },
232
+ {
233
+ "epoch": 2.213740458015267,
234
+ "grad_norm": 0.051850203424692154,
235
+ "learning_rate": 7.3536895674300254e-06,
236
+ "loss": 0.0026,
237
+ "step": 290
238
+ },
239
+ {
240
+ "epoch": 2.2900763358778624,
241
+ "grad_norm": 0.04679035767912865,
242
+ "learning_rate": 7.608142493638677e-06,
243
+ "loss": 0.0024,
244
+ "step": 300
245
+ },
246
+ {
247
+ "epoch": 2.366412213740458,
248
+ "grad_norm": 0.043534696102142334,
249
+ "learning_rate": 7.862595419847328e-06,
250
+ "loss": 0.0022,
251
+ "step": 310
252
+ },
253
+ {
254
+ "epoch": 2.4427480916030535,
255
+ "grad_norm": 0.06631364673376083,
256
+ "learning_rate": 8.11704834605598e-06,
257
+ "loss": 0.0501,
258
+ "step": 320
259
+ },
260
+ {
261
+ "epoch": 2.519083969465649,
262
+ "grad_norm": 0.2406100630760193,
263
+ "learning_rate": 8.371501272264631e-06,
264
+ "loss": 0.1185,
265
+ "step": 330
266
+ },
267
+ {
268
+ "epoch": 2.595419847328244,
269
+ "grad_norm": 0.15562430024147034,
270
+ "learning_rate": 8.625954198473283e-06,
271
+ "loss": 0.1555,
272
+ "step": 340
273
+ },
274
+ {
275
+ "epoch": 2.67175572519084,
276
+ "grad_norm": 0.07086818665266037,
277
+ "learning_rate": 8.880407124681935e-06,
278
+ "loss": 0.0396,
279
+ "step": 350
280
+ },
281
+ {
282
+ "epoch": 2.7480916030534353,
283
+ "grad_norm": 3.227670192718506,
284
+ "learning_rate": 9.134860050890587e-06,
285
+ "loss": 0.1098,
286
+ "step": 360
287
+ },
288
+ {
289
+ "epoch": 2.8244274809160306,
290
+ "grad_norm": 2.484853744506836,
291
+ "learning_rate": 9.389312977099237e-06,
292
+ "loss": 0.2296,
293
+ "step": 370
294
+ },
295
+ {
296
+ "epoch": 2.900763358778626,
297
+ "grad_norm": 1.2200217247009277,
298
+ "learning_rate": 9.643765903307888e-06,
299
+ "loss": 0.0674,
300
+ "step": 380
301
+ },
302
+ {
303
+ "epoch": 2.9770992366412212,
304
+ "grad_norm": 0.7877931594848633,
305
+ "learning_rate": 9.89821882951654e-06,
306
+ "loss": 0.1377,
307
+ "step": 390
308
+ },
309
+ {
310
+ "epoch": 3.0,
311
+ "eval_f1": 0.9354838709677419,
312
+ "eval_fbeta": 0.9427827536107456,
313
+ "eval_loss": 0.1109568178653717,
314
+ "eval_precision": 0.9477124183006536,
315
+ "eval_recall": 0.9235668789808917,
316
+ "eval_runtime": 13.3172,
317
+ "eval_samples_per_second": 67.206,
318
+ "eval_steps_per_second": 8.41,
319
+ "step": 393
320
+ },
321
+ {
322
+ "epoch": 3.053435114503817,
323
+ "grad_norm": 0.1212519109249115,
324
+ "learning_rate": 9.961832061068703e-06,
325
+ "loss": 0.0141,
326
+ "step": 400
327
+ },
328
+ {
329
+ "epoch": 3.1297709923664123,
330
+ "grad_norm": 0.18422600626945496,
331
+ "learning_rate": 9.89821882951654e-06,
332
+ "loss": 0.1976,
333
+ "step": 410
334
+ },
335
+ {
336
+ "epoch": 3.2061068702290076,
337
+ "grad_norm": 3.665578842163086,
338
+ "learning_rate": 9.834605597964377e-06,
339
+ "loss": 0.0717,
340
+ "step": 420
341
+ },
342
+ {
343
+ "epoch": 3.282442748091603,
344
+ "grad_norm": 0.1137157678604126,
345
+ "learning_rate": 9.770992366412214e-06,
346
+ "loss": 0.0075,
347
+ "step": 430
348
+ },
349
+ {
350
+ "epoch": 3.3587786259541983,
351
+ "grad_norm": 0.0800420492887497,
352
+ "learning_rate": 9.707379134860051e-06,
353
+ "loss": 0.0049,
354
+ "step": 440
355
+ },
356
+ {
357
+ "epoch": 3.435114503816794,
358
+ "grad_norm": 0.06329266726970673,
359
+ "learning_rate": 9.643765903307888e-06,
360
+ "loss": 0.0036,
361
+ "step": 450
362
+ },
363
+ {
364
+ "epoch": 3.5114503816793894,
365
+ "grad_norm": 0.058372776955366135,
366
+ "learning_rate": 9.580152671755725e-06,
367
+ "loss": 0.0391,
368
+ "step": 460
369
+ },
370
+ {
371
+ "epoch": 3.5877862595419847,
372
+ "grad_norm": 0.05802847445011139,
373
+ "learning_rate": 9.516539440203563e-06,
374
+ "loss": 0.0029,
375
+ "step": 470
376
+ },
377
+ {
378
+ "epoch": 3.66412213740458,
379
+ "grad_norm": 0.050598662346601486,
380
+ "learning_rate": 9.452926208651401e-06,
381
+ "loss": 0.0026,
382
+ "step": 480
383
+ },
384
+ {
385
+ "epoch": 3.7404580152671754,
386
+ "grad_norm": 0.045668765902519226,
387
+ "learning_rate": 9.389312977099237e-06,
388
+ "loss": 0.0023,
389
+ "step": 490
390
+ },
391
+ {
392
+ "epoch": 3.816793893129771,
393
+ "grad_norm": 0.040798306465148926,
394
+ "learning_rate": 9.325699745547074e-06,
395
+ "loss": 0.0021,
396
+ "step": 500
397
+ },
398
+ {
399
+ "epoch": 3.8931297709923665,
400
+ "grad_norm": 0.038592200726270676,
401
+ "learning_rate": 9.26208651399491e-06,
402
+ "loss": 0.0019,
403
+ "step": 510
404
+ },
405
+ {
406
+ "epoch": 3.969465648854962,
407
+ "grad_norm": 0.03555968776345253,
408
+ "learning_rate": 9.19847328244275e-06,
409
+ "loss": 0.0017,
410
+ "step": 520
411
+ },
412
+ {
413
+ "epoch": 4.0,
414
+ "eval_f1": 1.0,
415
+ "eval_fbeta": 0.9999999200000064,
416
+ "eval_loss": 0.0016206938307732344,
417
+ "eval_precision": 1.0,
418
+ "eval_recall": 1.0,
419
+ "eval_runtime": 13.3404,
420
+ "eval_samples_per_second": 67.089,
421
+ "eval_steps_per_second": 8.396,
422
+ "step": 524
423
+ }
424
+ ],
425
+ "logging_steps": 10,
426
+ "max_steps": 1965,
427
+ "num_input_tokens_seen": 0,
428
+ "num_train_epochs": 15,
429
+ "save_steps": 500,
430
+ "stateful_callbacks": {
431
+ "TrainerControl": {
432
+ "args": {
433
+ "should_epoch_stop": false,
434
+ "should_evaluate": false,
435
+ "should_log": false,
436
+ "should_save": true,
437
+ "should_training_stop": false
438
+ },
439
+ "attributes": {}
440
+ }
441
+ },
442
+ "total_flos": 4.07357231514624e+17,
443
+ "train_batch_size": 16,
444
+ "trial_name": null,
445
+ "trial_params": null
446
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30512d3c7b53a461013c5fedd16f10990cfee8a1d78f2784969f161fbf3a4fce
3
+ size 5841