Hugging Face's logo

openpecha
/
wav2vec2-large-xlsr-tibetan

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Inference Endpoints
Model card Files Community
1
TenzinGayche commited on
Commit
6bbda78
•
1 Parent(s): 37b1cf0

Training in progress, step 2100

Browse files
Files changed (19) hide show
  1. {checkpoint-1000 → checkpoint-2000}/config.json +0 -0
  2. {checkpoint-1000 → checkpoint-2000}/optimizer.pt +1 -1
  3. {checkpoint-1000 → checkpoint-2000}/preprocessor_config.json +0 -0
  4. {checkpoint-900 → checkpoint-2000}/pytorch_model.bin +1 -1
  5. {checkpoint-900 → checkpoint-2000}/rng_state.pth +1 -1
  6. {checkpoint-900 → checkpoint-2000}/scaler.pt +1 -1
  7. {checkpoint-1000 → checkpoint-2000}/scheduler.pt +1 -1
  8. {checkpoint-1000 → checkpoint-2000}/trainer_state.json +213 -3
  9. {checkpoint-1000 → checkpoint-2000}/training_args.bin +0 -0
  10. {checkpoint-900 → checkpoint-2100}/config.json +0 -0
  11. {checkpoint-900 → checkpoint-2100}/optimizer.pt +1 -1
  12. {checkpoint-900 → checkpoint-2100}/preprocessor_config.json +0 -0
  13. {checkpoint-1000 → checkpoint-2100}/pytorch_model.bin +1 -1
  14. {checkpoint-1000 → checkpoint-2100}/rng_state.pth +1 -1
  15. {checkpoint-1000 → checkpoint-2100}/scaler.pt +1 -1
  16. {checkpoint-900 → checkpoint-2100}/scheduler.pt +1 -1
  17. {checkpoint-900 → checkpoint-2100}/trainer_state.json +255 -3
  18. {checkpoint-900 → checkpoint-2100}/training_args.bin +0 -0
  19. pytorch_model.bin +1 -1
{checkpoint-1000 → checkpoint-2000}/config.json RENAMED
File without changes
{checkpoint-1000 → checkpoint-2000}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9047a1dbb440cc38970a45319a5646304b15ba0a0843ebcec9895bbdda0ce321
3
  size 2490946501
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3400ad82ded439b14c2bac67c692eb6cb3776ec6471beac1430b11b22c8fb5b5
3
  size 2490946501
{checkpoint-1000 → checkpoint-2000}/preprocessor_config.json RENAMED
File without changes
{checkpoint-900 → checkpoint-2000}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:53556d92279470e1de7995319963f9e791416b3ad7a6b250e76df709d9088b4a
3
  size 1262344621
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cd38ec9830107698c5d0ace7830fd9cf63e7332cde8b2dc9f42453c1fe96228
3
  size 1262344621
{checkpoint-900 → checkpoint-2000}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9629ea4ed6e8d0fb67511638b19dee016cfb18ca8f32f139b61c340918553503
3
  size 14639
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5c1cece2539dcc21e5f05a104f8a7b0d17369adf61ee02eabacdc5688c8ce77
3
  size 14639
{checkpoint-900 → checkpoint-2000}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0fce1b7d1b885f4cc828f1540e3b7795f526f4e8af335f562266ab79d5d296ce
3
  size 557
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa1f6377b029d7c93ef2025a47117a476c1020f2bd22a0097fe33acd14b35ae7
3
  size 557
{checkpoint-1000 → checkpoint-2000}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e835e668d21436e5880ed5cc917e1d91cd36b1736933800a6be8ba358688f2ff
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb073859e728ca55c339a2f6d3ed17539f043fe9dec39e46f41d5b139c046481
3
  size 627
{checkpoint-1000 → checkpoint-2000}/trainer_state.json RENAMED
@@ -1,8 +1,8 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 6.802721088435375,
5
- "global_step": 1000,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
@@ -216,11 +216,221 @@
216
  "eval_samples_per_second": 23.898,
217
  "eval_steps_per_second": 3.008,
218
  "step": 1000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
219
  }
220
  ],
221
  "max_steps": 3675,
222
  "num_train_epochs": 25,
223
- "total_flos": 3.8998743326073446e+18,
224
  "trial_name": null,
225
  "trial_params": null
226
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 13.60544217687075,
5
+ "global_step": 2000,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
 
216
  "eval_samples_per_second": 23.898,
217
  "eval_steps_per_second": 3.008,
218
  "step": 1000
219
+ },
220
+ {
221
+ "epoch": 7.14,
222
+ "learning_rate": 0.0002483149606299212,
223
+ "loss": 0.6349,
224
+ "step": 1050
225
+ },
226
+ {
227
+ "epoch": 7.48,
228
+ "learning_rate": 0.00024359055118110235,
229
+ "loss": 0.5935,
230
+ "step": 1100
231
+ },
232
+ {
233
+ "epoch": 7.48,
234
+ "eval_cer": 0.17746616663949127,
235
+ "eval_loss": 0.46862325072288513,
236
+ "eval_runtime": 24.2024,
237
+ "eval_samples_per_second": 24.295,
238
+ "eval_steps_per_second": 3.058,
239
+ "step": 1100
240
+ },
241
+ {
242
+ "epoch": 7.82,
243
+ "learning_rate": 0.00023886614173228342,
244
+ "loss": 0.576,
245
+ "step": 1150
246
+ },
247
+ {
248
+ "epoch": 8.16,
249
+ "learning_rate": 0.00023414173228346455,
250
+ "loss": 0.5239,
251
+ "step": 1200
252
+ },
253
+ {
254
+ "epoch": 8.16,
255
+ "eval_cer": 0.1918799934779064,
256
+ "eval_loss": 0.4271094799041748,
257
+ "eval_runtime": 24.6968,
258
+ "eval_samples_per_second": 23.809,
259
+ "eval_steps_per_second": 2.996,
260
+ "step": 1200
261
+ },
262
+ {
263
+ "epoch": 8.5,
264
+ "learning_rate": 0.00022941732283464564,
265
+ "loss": 0.5207,
266
+ "step": 1250
267
+ },
268
+ {
269
+ "epoch": 8.84,
270
+ "learning_rate": 0.00022469291338582677,
271
+ "loss": 0.4855,
272
+ "step": 1300
273
+ },
274
+ {
275
+ "epoch": 8.84,
276
+ "eval_cer": 0.17550953856187837,
277
+ "eval_loss": 0.40729257464408875,
278
+ "eval_runtime": 24.885,
279
+ "eval_samples_per_second": 23.629,
280
+ "eval_steps_per_second": 2.974,
281
+ "step": 1300
282
+ },
283
+ {
284
+ "epoch": 9.18,
285
+ "learning_rate": 0.00021996850393700784,
286
+ "loss": 0.4621,
287
+ "step": 1350
288
+ },
289
+ {
290
+ "epoch": 9.52,
291
+ "learning_rate": 0.00021524409448818896,
292
+ "loss": 0.4583,
293
+ "step": 1400
294
+ },
295
+ {
296
+ "epoch": 9.52,
297
+ "eval_cer": 0.1760313060492418,
298
+ "eval_loss": 0.41417357325553894,
299
+ "eval_runtime": 24.8425,
300
+ "eval_samples_per_second": 23.669,
301
+ "eval_steps_per_second": 2.979,
302
+ "step": 1400
303
+ },
304
+ {
305
+ "epoch": 9.86,
306
+ "learning_rate": 0.00021051968503937006,
307
+ "loss": 0.4415,
308
+ "step": 1450
309
+ },
310
+ {
311
+ "epoch": 10.2,
312
+ "learning_rate": 0.00020579527559055118,
313
+ "loss": 0.417,
314
+ "step": 1500
315
+ },
316
+ {
317
+ "epoch": 10.2,
318
+ "eval_cer": 0.17156367193869232,
319
+ "eval_loss": 0.3833578824996948,
320
+ "eval_runtime": 24.455,
321
+ "eval_samples_per_second": 24.044,
322
+ "eval_steps_per_second": 3.026,
323
+ "step": 1500
324
+ },
325
+ {
326
+ "epoch": 10.54,
327
+ "learning_rate": 0.00020107086614173225,
328
+ "loss": 0.3959,
329
+ "step": 1550
330
+ },
331
+ {
332
+ "epoch": 10.88,
333
+ "learning_rate": 0.00019634645669291338,
334
+ "loss": 0.4074,
335
+ "step": 1600
336
+ },
337
+ {
338
+ "epoch": 10.88,
339
+ "eval_cer": 0.16301972933311593,
340
+ "eval_loss": 0.3626195192337036,
341
+ "eval_runtime": 24.5184,
342
+ "eval_samples_per_second": 23.982,
343
+ "eval_steps_per_second": 3.018,
344
+ "step": 1600
345
+ },
346
+ {
347
+ "epoch": 11.22,
348
+ "learning_rate": 0.00019162204724409448,
349
+ "loss": 0.3584,
350
+ "step": 1650
351
+ },
352
+ {
353
+ "epoch": 11.56,
354
+ "learning_rate": 0.0001868976377952756,
355
+ "loss": 0.3682,
356
+ "step": 1700
357
+ },
358
+ {
359
+ "epoch": 11.56,
360
+ "eval_cer": 0.15685635088863525,
361
+ "eval_loss": 0.36927542090415955,
362
+ "eval_runtime": 24.3751,
363
+ "eval_samples_per_second": 24.123,
364
+ "eval_steps_per_second": 3.036,
365
+ "step": 1700
366
+ },
367
+ {
368
+ "epoch": 11.9,
369
+ "learning_rate": 0.00018217322834645667,
370
+ "loss": 0.3455,
371
+ "step": 1750
372
+ },
373
+ {
374
+ "epoch": 12.24,
375
+ "learning_rate": 0.00017744881889763777,
376
+ "loss": 0.3245,
377
+ "step": 1800
378
+ },
379
+ {
380
+ "epoch": 12.24,
381
+ "eval_cer": 0.15816076960704387,
382
+ "eval_loss": 0.3740461468696594,
383
+ "eval_runtime": 24.7162,
384
+ "eval_samples_per_second": 23.79,
385
+ "eval_steps_per_second": 2.994,
386
+ "step": 1800
387
+ },
388
+ {
389
+ "epoch": 12.59,
390
+ "learning_rate": 0.0001727244094488189,
391
+ "loss": 0.3208,
392
+ "step": 1850
393
+ },
394
+ {
395
+ "epoch": 12.93,
396
+ "learning_rate": 0.000168,
397
+ "loss": 0.3063,
398
+ "step": 1900
399
+ },
400
+ {
401
+ "epoch": 12.93,
402
+ "eval_cer": 0.15904125224196966,
403
+ "eval_loss": 0.3622555434703827,
404
+ "eval_runtime": 24.4729,
405
+ "eval_samples_per_second": 24.027,
406
+ "eval_steps_per_second": 3.024,
407
+ "step": 1900
408
+ },
409
+ {
410
+ "epoch": 13.27,
411
+ "learning_rate": 0.0001632755905511811,
412
+ "loss": 0.3019,
413
+ "step": 1950
414
+ },
415
+ {
416
+ "epoch": 13.61,
417
+ "learning_rate": 0.00015855118110236219,
418
+ "loss": 0.2945,
419
+ "step": 2000
420
+ },
421
+ {
422
+ "epoch": 13.61,
423
+ "eval_cer": 0.16634599706505787,
424
+ "eval_loss": 0.3725011348724365,
425
+ "eval_runtime": 25.0023,
426
+ "eval_samples_per_second": 23.518,
427
+ "eval_steps_per_second": 2.96,
428
+ "step": 2000
429
  }
430
  ],
431
  "max_steps": 3675,
432
  "num_train_epochs": 25,
433
+ "total_flos": 7.778933854152041e+18,
434
  "trial_name": null,
435
  "trial_params": null
436
  }
{checkpoint-1000 → checkpoint-2000}/training_args.bin RENAMED
File without changes
{checkpoint-900 → checkpoint-2100}/config.json RENAMED
File without changes
{checkpoint-900 → checkpoint-2100}/optimizer.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:107d92d4ddd5bc4cd7edde59bd7485ad60137cc09e6a3dd29c36502c0498fb02
3
  size 2490946501
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87e2080de9b12e77e557f50a48e30b7ba4194eca0fd9c1a98f3b79a795706b37
3
  size 2490946501
{checkpoint-900 → checkpoint-2100}/preprocessor_config.json RENAMED
File without changes
{checkpoint-1000 → checkpoint-2100}/pytorch_model.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f063532b60c9d8ebea8e407d99824e1467e0ec25d6e28485f2e79240c5f3e9e9
3
  size 1262344621
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec5ea49be8b746ea48d869dab13f1c8fb1f407557d206b2a7a9ee54fc7002b55
3
  size 1262344621
{checkpoint-1000 → checkpoint-2100}/rng_state.pth RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a57d9d544f974865071e68a80b84de5de664a230baeb249be510db0a0d96d04
3
  size 14639
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d26c90a74252f9ad5cdc4c91db335ffbbece95883ba82e495b95ac3d28e6ea52
3
  size 14639
{checkpoint-1000 → checkpoint-2100}/scaler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:31f7c33f7d4b3d298a27a7dc6b22c035e0393e714287cc7d69f421614213cb75
3
  size 557
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f255336918f2820e02ae648c756202ba55ed3193e9fd1950955ca2f6fb0457f
3
  size 557
{checkpoint-900 → checkpoint-2100}/scheduler.pt RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d74636d9d96b3d2802b618f989e7624586129edb69921d8923707bdbc8fc257a
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c20e8e4f4623e1333635d4d6f55117df2428ab91ba217c9207d1a49a91c2552
3
  size 627
{checkpoint-900 → checkpoint-2100}/trainer_state.json RENAMED
@@ -1,8 +1,8 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 6.122448979591836,
5
- "global_step": 900,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
@@ -195,11 +195,263 @@
195
  "eval_samples_per_second": 23.922,
196
  "eval_steps_per_second": 3.011,
197
  "step": 900
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
198
  }
199
  ],
200
  "max_steps": 3675,
201
  "num_train_epochs": 25,
202
- "total_flos": 3.513728116263168e+18,
203
  "trial_name": null,
204
  "trial_params": null
205
  }
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 14.285714285714286,
5
+ "global_step": 2100,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
 
195
  "eval_samples_per_second": 23.922,
196
  "eval_steps_per_second": 3.011,
197
  "step": 900
198
+ },
199
+ {
200
+ "epoch": 6.46,
201
+ "learning_rate": 0.00025776377952755903,
202
+ "loss": 0.6711,
203
+ "step": 950
204
+ },
205
+ {
206
+ "epoch": 6.8,
207
+ "learning_rate": 0.00025303937007874016,
208
+ "loss": 0.7003,
209
+ "step": 1000
210
+ },
211
+ {
212
+ "epoch": 6.8,
213
+ "eval_cer": 0.19279308658079244,
214
+ "eval_loss": 0.5100580453872681,
215
+ "eval_runtime": 24.6049,
216
+ "eval_samples_per_second": 23.898,
217
+ "eval_steps_per_second": 3.008,
218
+ "step": 1000
219
+ },
220
+ {
221
+ "epoch": 7.14,
222
+ "learning_rate": 0.0002483149606299212,
223
+ "loss": 0.6349,
224
+ "step": 1050
225
+ },
226
+ {
227
+ "epoch": 7.48,
228
+ "learning_rate": 0.00024359055118110235,
229
+ "loss": 0.5935,
230
+ "step": 1100
231
+ },
232
+ {
233
+ "epoch": 7.48,
234
+ "eval_cer": 0.17746616663949127,
235
+ "eval_loss": 0.46862325072288513,
236
+ "eval_runtime": 24.2024,
237
+ "eval_samples_per_second": 24.295,
238
+ "eval_steps_per_second": 3.058,
239
+ "step": 1100
240
+ },
241
+ {
242
+ "epoch": 7.82,
243
+ "learning_rate": 0.00023886614173228342,
244
+ "loss": 0.576,
245
+ "step": 1150
246
+ },
247
+ {
248
+ "epoch": 8.16,
249
+ "learning_rate": 0.00023414173228346455,
250
+ "loss": 0.5239,
251
+ "step": 1200
252
+ },
253
+ {
254
+ "epoch": 8.16,
255
+ "eval_cer": 0.1918799934779064,
256
+ "eval_loss": 0.4271094799041748,
257
+ "eval_runtime": 24.6968,
258
+ "eval_samples_per_second": 23.809,
259
+ "eval_steps_per_second": 2.996,
260
+ "step": 1200
261
+ },
262
+ {
263
+ "epoch": 8.5,
264
+ "learning_rate": 0.00022941732283464564,
265
+ "loss": 0.5207,
266
+ "step": 1250
267
+ },
268
+ {
269
+ "epoch": 8.84,
270
+ "learning_rate": 0.00022469291338582677,
271
+ "loss": 0.4855,
272
+ "step": 1300
273
+ },
274
+ {
275
+ "epoch": 8.84,
276
+ "eval_cer": 0.17550953856187837,
277
+ "eval_loss": 0.40729257464408875,
278
+ "eval_runtime": 24.885,
279
+ "eval_samples_per_second": 23.629,
280
+ "eval_steps_per_second": 2.974,
281
+ "step": 1300
282
+ },
283
+ {
284
+ "epoch": 9.18,
285
+ "learning_rate": 0.00021996850393700784,
286
+ "loss": 0.4621,
287
+ "step": 1350
288
+ },
289
+ {
290
+ "epoch": 9.52,
291
+ "learning_rate": 0.00021524409448818896,
292
+ "loss": 0.4583,
293
+ "step": 1400
294
+ },
295
+ {
296
+ "epoch": 9.52,
297
+ "eval_cer": 0.1760313060492418,
298
+ "eval_loss": 0.41417357325553894,
299
+ "eval_runtime": 24.8425,
300
+ "eval_samples_per_second": 23.669,
301
+ "eval_steps_per_second": 2.979,
302
+ "step": 1400
303
+ },
304
+ {
305
+ "epoch": 9.86,
306
+ "learning_rate": 0.00021051968503937006,
307
+ "loss": 0.4415,
308
+ "step": 1450
309
+ },
310
+ {
311
+ "epoch": 10.2,
312
+ "learning_rate": 0.00020579527559055118,
313
+ "loss": 0.417,
314
+ "step": 1500
315
+ },
316
+ {
317
+ "epoch": 10.2,
318
+ "eval_cer": 0.17156367193869232,
319
+ "eval_loss": 0.3833578824996948,
320
+ "eval_runtime": 24.455,
321
+ "eval_samples_per_second": 24.044,
322
+ "eval_steps_per_second": 3.026,
323
+ "step": 1500
324
+ },
325
+ {
326
+ "epoch": 10.54,
327
+ "learning_rate": 0.00020107086614173225,
328
+ "loss": 0.3959,
329
+ "step": 1550
330
+ },
331
+ {
332
+ "epoch": 10.88,
333
+ "learning_rate": 0.00019634645669291338,
334
+ "loss": 0.4074,
335
+ "step": 1600
336
+ },
337
+ {
338
+ "epoch": 10.88,
339
+ "eval_cer": 0.16301972933311593,
340
+ "eval_loss": 0.3626195192337036,
341
+ "eval_runtime": 24.5184,
342
+ "eval_samples_per_second": 23.982,
343
+ "eval_steps_per_second": 3.018,
344
+ "step": 1600
345
+ },
346
+ {
347
+ "epoch": 11.22,
348
+ "learning_rate": 0.00019162204724409448,
349
+ "loss": 0.3584,
350
+ "step": 1650
351
+ },
352
+ {
353
+ "epoch": 11.56,
354
+ "learning_rate": 0.0001868976377952756,
355
+ "loss": 0.3682,
356
+ "step": 1700
357
+ },
358
+ {
359
+ "epoch": 11.56,
360
+ "eval_cer": 0.15685635088863525,
361
+ "eval_loss": 0.36927542090415955,
362
+ "eval_runtime": 24.3751,
363
+ "eval_samples_per_second": 24.123,
364
+ "eval_steps_per_second": 3.036,
365
+ "step": 1700
366
+ },
367
+ {
368
+ "epoch": 11.9,
369
+ "learning_rate": 0.00018217322834645667,
370
+ "loss": 0.3455,
371
+ "step": 1750
372
+ },
373
+ {
374
+ "epoch": 12.24,
375
+ "learning_rate": 0.00017744881889763777,
376
+ "loss": 0.3245,
377
+ "step": 1800
378
+ },
379
+ {
380
+ "epoch": 12.24,
381
+ "eval_cer": 0.15816076960704387,
382
+ "eval_loss": 0.3740461468696594,
383
+ "eval_runtime": 24.7162,
384
+ "eval_samples_per_second": 23.79,
385
+ "eval_steps_per_second": 2.994,
386
+ "step": 1800
387
+ },
388
+ {
389
+ "epoch": 12.59,
390
+ "learning_rate": 0.0001727244094488189,
391
+ "loss": 0.3208,
392
+ "step": 1850
393
+ },
394
+ {
395
+ "epoch": 12.93,
396
+ "learning_rate": 0.000168,
397
+ "loss": 0.3063,
398
+ "step": 1900
399
+ },
400
+ {
401
+ "epoch": 12.93,
402
+ "eval_cer": 0.15904125224196966,
403
+ "eval_loss": 0.3622555434703827,
404
+ "eval_runtime": 24.4729,
405
+ "eval_samples_per_second": 24.027,
406
+ "eval_steps_per_second": 3.024,
407
+ "step": 1900
408
+ },
409
+ {
410
+ "epoch": 13.27,
411
+ "learning_rate": 0.0001632755905511811,
412
+ "loss": 0.3019,
413
+ "step": 1950
414
+ },
415
+ {
416
+ "epoch": 13.61,
417
+ "learning_rate": 0.00015855118110236219,
418
+ "loss": 0.2945,
419
+ "step": 2000
420
+ },
421
+ {
422
+ "epoch": 13.61,
423
+ "eval_cer": 0.16634599706505787,
424
+ "eval_loss": 0.3725011348724365,
425
+ "eval_runtime": 25.0023,
426
+ "eval_samples_per_second": 23.518,
427
+ "eval_steps_per_second": 2.96,
428
+ "step": 2000
429
+ },
430
+ {
431
+ "epoch": 13.95,
432
+ "learning_rate": 0.0001538267716535433,
433
+ "loss": 0.279,
434
+ "step": 2050
435
+ },
436
+ {
437
+ "epoch": 14.29,
438
+ "learning_rate": 0.0001491023622047244,
439
+ "loss": 0.2674,
440
+ "step": 2100
441
+ },
442
+ {
443
+ "epoch": 14.29,
444
+ "eval_cer": 0.15731289744007826,
445
+ "eval_loss": 0.3531067371368408,
446
+ "eval_runtime": 24.8381,
447
+ "eval_samples_per_second": 23.673,
448
+ "eval_steps_per_second": 2.979,
449
+ "step": 2100
450
  }
451
  ],
452
  "max_steps": 3675,
453
  "num_train_epochs": 25,
454
+ "total_flos": 8.157984479823167e+18,
455
  "trial_name": null,
456
  "trial_params": null
457
  }
{checkpoint-900 → checkpoint-2100}/training_args.bin RENAMED
File without changes
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:944bbfe53ed8e3678f0df597947df90c1f1fd1a592c5e2f1c1893cd644db8b25
3
  size 1262344621
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec5ea49be8b746ea48d869dab13f1c8fb1f407557d206b2a7a9ee54fc7002b55
3
  size 1262344621