Abhi4 commited on
Commit
6311fbb
1 Parent(s): 6fb613e

End of training

Browse files
Files changed (4) hide show
  1. all_results.json +13 -0
  2. eval_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +589 -0
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9594602592176126,
4
+ "eval_loss": 0.1155962273478508,
5
+ "eval_runtime": 284.4248,
6
+ "eval_samples_per_second": 59.408,
7
+ "eval_steps_per_second": 0.468,
8
+ "total_flos": 3.5333106012326707e+19,
9
+ "train_loss": 0.27140078694464514,
10
+ "train_runtime": 14713.6433,
11
+ "train_samples_per_second": 31.006,
12
+ "train_steps_per_second": 0.061
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9594602592176126,
4
+ "eval_loss": 0.1155962273478508,
5
+ "eval_runtime": 284.4248,
6
+ "eval_samples_per_second": 59.408,
7
+ "eval_steps_per_second": 0.468
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 3.5333106012326707e+19,
4
+ "train_loss": 0.27140078694464514,
5
+ "train_runtime": 14713.6433,
6
+ "train_samples_per_second": 31.006,
7
+ "train_steps_per_second": 0.061
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,589 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9594602592176126,
3
+ "best_model_checkpoint": "vit-base-patch16-224-abhi1-finetuned/checkpoint-891",
4
+ "epoch": 2.9974768713204374,
5
+ "eval_steps": 500,
6
+ "global_step": 891,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03,
13
+ "learning_rate": 5.555555555555556e-06,
14
+ "loss": 1.5094,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.07,
19
+ "learning_rate": 1.1111111111111112e-05,
20
+ "loss": 1.1818,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.1,
25
+ "learning_rate": 1.6666666666666667e-05,
26
+ "loss": 0.9093,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.13,
31
+ "learning_rate": 2.2222222222222223e-05,
32
+ "loss": 0.6775,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.17,
37
+ "learning_rate": 2.777777777777778e-05,
38
+ "loss": 0.5135,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.2,
43
+ "learning_rate": 3.3333333333333335e-05,
44
+ "loss": 0.42,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.24,
49
+ "learning_rate": 3.888888888888889e-05,
50
+ "loss": 0.3438,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.27,
55
+ "learning_rate": 4.4444444444444447e-05,
56
+ "loss": 0.3251,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.3,
61
+ "learning_rate": 5e-05,
62
+ "loss": 0.3169,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.34,
67
+ "learning_rate": 4.937578027465668e-05,
68
+ "loss": 0.3051,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.37,
73
+ "learning_rate": 4.875156054931336e-05,
74
+ "loss": 0.3075,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.4,
79
+ "learning_rate": 4.812734082397004e-05,
80
+ "loss": 0.3102,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.44,
85
+ "learning_rate": 4.750312109862672e-05,
86
+ "loss": 0.2866,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.47,
91
+ "learning_rate": 4.6878901373283394e-05,
92
+ "loss": 0.2821,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.5,
97
+ "learning_rate": 4.625468164794008e-05,
98
+ "loss": 0.2753,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.54,
103
+ "learning_rate": 4.563046192259676e-05,
104
+ "loss": 0.2698,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.57,
109
+ "learning_rate": 4.500624219725343e-05,
110
+ "loss": 0.2627,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.61,
115
+ "learning_rate": 4.438202247191011e-05,
116
+ "loss": 0.259,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.64,
121
+ "learning_rate": 4.37578027465668e-05,
122
+ "loss": 0.2448,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.67,
127
+ "learning_rate": 4.313358302122347e-05,
128
+ "loss": 0.2691,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.71,
133
+ "learning_rate": 4.250936329588015e-05,
134
+ "loss": 0.2573,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.74,
139
+ "learning_rate": 4.188514357053683e-05,
140
+ "loss": 0.2609,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.77,
145
+ "learning_rate": 4.126092384519351e-05,
146
+ "loss": 0.2503,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.81,
151
+ "learning_rate": 4.063670411985019e-05,
152
+ "loss": 0.2488,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.84,
157
+ "learning_rate": 4.001248439450687e-05,
158
+ "loss": 0.2468,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.87,
163
+ "learning_rate": 3.938826466916355e-05,
164
+ "loss": 0.2433,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.91,
169
+ "learning_rate": 3.876404494382023e-05,
170
+ "loss": 0.248,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.94,
175
+ "learning_rate": 3.813982521847691e-05,
176
+ "loss": 0.239,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.98,
181
+ "learning_rate": 3.751560549313358e-05,
182
+ "loss": 0.2361,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 1.0,
187
+ "eval_accuracy": 0.9521808605077825,
188
+ "eval_loss": 0.14011575281620026,
189
+ "eval_runtime": 329.6902,
190
+ "eval_samples_per_second": 51.251,
191
+ "eval_steps_per_second": 0.403,
192
+ "step": 297
193
+ },
194
+ {
195
+ "epoch": 1.01,
196
+ "learning_rate": 3.689138576779027e-05,
197
+ "loss": 0.2282,
198
+ "step": 300
199
+ },
200
+ {
201
+ "epoch": 1.04,
202
+ "learning_rate": 3.626716604244694e-05,
203
+ "loss": 0.2398,
204
+ "step": 310
205
+ },
206
+ {
207
+ "epoch": 1.08,
208
+ "learning_rate": 3.564294631710362e-05,
209
+ "loss": 0.2279,
210
+ "step": 320
211
+ },
212
+ {
213
+ "epoch": 1.11,
214
+ "learning_rate": 3.5018726591760305e-05,
215
+ "loss": 0.2427,
216
+ "step": 330
217
+ },
218
+ {
219
+ "epoch": 1.14,
220
+ "learning_rate": 3.439450686641698e-05,
221
+ "loss": 0.2261,
222
+ "step": 340
223
+ },
224
+ {
225
+ "epoch": 1.18,
226
+ "learning_rate": 3.377028714107366e-05,
227
+ "loss": 0.2423,
228
+ "step": 350
229
+ },
230
+ {
231
+ "epoch": 1.21,
232
+ "learning_rate": 3.314606741573034e-05,
233
+ "loss": 0.2285,
234
+ "step": 360
235
+ },
236
+ {
237
+ "epoch": 1.24,
238
+ "learning_rate": 3.252184769038702e-05,
239
+ "loss": 0.2412,
240
+ "step": 370
241
+ },
242
+ {
243
+ "epoch": 1.28,
244
+ "learning_rate": 3.1897627965043697e-05,
245
+ "loss": 0.2139,
246
+ "step": 380
247
+ },
248
+ {
249
+ "epoch": 1.31,
250
+ "learning_rate": 3.1273408239700376e-05,
251
+ "loss": 0.2369,
252
+ "step": 390
253
+ },
254
+ {
255
+ "epoch": 1.35,
256
+ "learning_rate": 3.0649188514357056e-05,
257
+ "loss": 0.2439,
258
+ "step": 400
259
+ },
260
+ {
261
+ "epoch": 1.38,
262
+ "learning_rate": 3.0024968789013735e-05,
263
+ "loss": 0.2149,
264
+ "step": 410
265
+ },
266
+ {
267
+ "epoch": 1.41,
268
+ "learning_rate": 2.940074906367041e-05,
269
+ "loss": 0.2219,
270
+ "step": 420
271
+ },
272
+ {
273
+ "epoch": 1.45,
274
+ "learning_rate": 2.877652933832709e-05,
275
+ "loss": 0.2193,
276
+ "step": 430
277
+ },
278
+ {
279
+ "epoch": 1.48,
280
+ "learning_rate": 2.8152309612983774e-05,
281
+ "loss": 0.2173,
282
+ "step": 440
283
+ },
284
+ {
285
+ "epoch": 1.51,
286
+ "learning_rate": 2.752808988764045e-05,
287
+ "loss": 0.2316,
288
+ "step": 450
289
+ },
290
+ {
291
+ "epoch": 1.55,
292
+ "learning_rate": 2.690387016229713e-05,
293
+ "loss": 0.2143,
294
+ "step": 460
295
+ },
296
+ {
297
+ "epoch": 1.58,
298
+ "learning_rate": 2.6279650436953806e-05,
299
+ "loss": 0.212,
300
+ "step": 470
301
+ },
302
+ {
303
+ "epoch": 1.61,
304
+ "learning_rate": 2.565543071161049e-05,
305
+ "loss": 0.2286,
306
+ "step": 480
307
+ },
308
+ {
309
+ "epoch": 1.65,
310
+ "learning_rate": 2.503121098626717e-05,
311
+ "loss": 0.2189,
312
+ "step": 490
313
+ },
314
+ {
315
+ "epoch": 1.68,
316
+ "learning_rate": 2.4406991260923848e-05,
317
+ "loss": 0.2175,
318
+ "step": 500
319
+ },
320
+ {
321
+ "epoch": 1.72,
322
+ "learning_rate": 2.3782771535580524e-05,
323
+ "loss": 0.2289,
324
+ "step": 510
325
+ },
326
+ {
327
+ "epoch": 1.75,
328
+ "learning_rate": 2.3158551810237207e-05,
329
+ "loss": 0.2254,
330
+ "step": 520
331
+ },
332
+ {
333
+ "epoch": 1.78,
334
+ "learning_rate": 2.2534332084893883e-05,
335
+ "loss": 0.2203,
336
+ "step": 530
337
+ },
338
+ {
339
+ "epoch": 1.82,
340
+ "learning_rate": 2.1910112359550563e-05,
341
+ "loss": 0.2029,
342
+ "step": 540
343
+ },
344
+ {
345
+ "epoch": 1.85,
346
+ "learning_rate": 2.1285892634207243e-05,
347
+ "loss": 0.1987,
348
+ "step": 550
349
+ },
350
+ {
351
+ "epoch": 1.88,
352
+ "learning_rate": 2.0661672908863922e-05,
353
+ "loss": 0.2075,
354
+ "step": 560
355
+ },
356
+ {
357
+ "epoch": 1.92,
358
+ "learning_rate": 2.00374531835206e-05,
359
+ "loss": 0.1962,
360
+ "step": 570
361
+ },
362
+ {
363
+ "epoch": 1.95,
364
+ "learning_rate": 1.9413233458177278e-05,
365
+ "loss": 0.2216,
366
+ "step": 580
367
+ },
368
+ {
369
+ "epoch": 1.98,
370
+ "learning_rate": 1.8789013732833958e-05,
371
+ "loss": 0.2012,
372
+ "step": 590
373
+ },
374
+ {
375
+ "epoch": 2.0,
376
+ "eval_accuracy": 0.9573297034976623,
377
+ "eval_loss": 0.12352792918682098,
378
+ "eval_runtime": 333.5253,
379
+ "eval_samples_per_second": 50.662,
380
+ "eval_steps_per_second": 0.399,
381
+ "step": 594
382
+ },
383
+ {
384
+ "epoch": 2.02,
385
+ "learning_rate": 1.8164794007490637e-05,
386
+ "loss": 0.2048,
387
+ "step": 600
388
+ },
389
+ {
390
+ "epoch": 2.05,
391
+ "learning_rate": 1.7540574282147317e-05,
392
+ "loss": 0.2041,
393
+ "step": 610
394
+ },
395
+ {
396
+ "epoch": 2.09,
397
+ "learning_rate": 1.6916354556803996e-05,
398
+ "loss": 0.2072,
399
+ "step": 620
400
+ },
401
+ {
402
+ "epoch": 2.12,
403
+ "learning_rate": 1.6292134831460676e-05,
404
+ "loss": 0.1977,
405
+ "step": 630
406
+ },
407
+ {
408
+ "epoch": 2.15,
409
+ "learning_rate": 1.5667915106117352e-05,
410
+ "loss": 0.2101,
411
+ "step": 640
412
+ },
413
+ {
414
+ "epoch": 2.19,
415
+ "learning_rate": 1.5043695380774033e-05,
416
+ "loss": 0.1979,
417
+ "step": 650
418
+ },
419
+ {
420
+ "epoch": 2.22,
421
+ "learning_rate": 1.4419475655430711e-05,
422
+ "loss": 0.1969,
423
+ "step": 660
424
+ },
425
+ {
426
+ "epoch": 2.25,
427
+ "learning_rate": 1.3795255930087393e-05,
428
+ "loss": 0.1941,
429
+ "step": 670
430
+ },
431
+ {
432
+ "epoch": 2.29,
433
+ "learning_rate": 1.3171036204744069e-05,
434
+ "loss": 0.2061,
435
+ "step": 680
436
+ },
437
+ {
438
+ "epoch": 2.32,
439
+ "learning_rate": 1.254681647940075e-05,
440
+ "loss": 0.1901,
441
+ "step": 690
442
+ },
443
+ {
444
+ "epoch": 2.35,
445
+ "learning_rate": 1.192259675405743e-05,
446
+ "loss": 0.2116,
447
+ "step": 700
448
+ },
449
+ {
450
+ "epoch": 2.39,
451
+ "learning_rate": 1.1298377028714107e-05,
452
+ "loss": 0.204,
453
+ "step": 710
454
+ },
455
+ {
456
+ "epoch": 2.42,
457
+ "learning_rate": 1.0674157303370787e-05,
458
+ "loss": 0.1881,
459
+ "step": 720
460
+ },
461
+ {
462
+ "epoch": 2.46,
463
+ "learning_rate": 1.0049937578027465e-05,
464
+ "loss": 0.1998,
465
+ "step": 730
466
+ },
467
+ {
468
+ "epoch": 2.49,
469
+ "learning_rate": 9.425717852684144e-06,
470
+ "loss": 0.1922,
471
+ "step": 740
472
+ },
473
+ {
474
+ "epoch": 2.52,
475
+ "learning_rate": 8.801498127340826e-06,
476
+ "loss": 0.1883,
477
+ "step": 750
478
+ },
479
+ {
480
+ "epoch": 2.56,
481
+ "learning_rate": 8.177278401997504e-06,
482
+ "loss": 0.1932,
483
+ "step": 760
484
+ },
485
+ {
486
+ "epoch": 2.59,
487
+ "learning_rate": 7.553058676654183e-06,
488
+ "loss": 0.2086,
489
+ "step": 770
490
+ },
491
+ {
492
+ "epoch": 2.62,
493
+ "learning_rate": 6.928838951310862e-06,
494
+ "loss": 0.1895,
495
+ "step": 780
496
+ },
497
+ {
498
+ "epoch": 2.66,
499
+ "learning_rate": 6.304619225967541e-06,
500
+ "loss": 0.1846,
501
+ "step": 790
502
+ },
503
+ {
504
+ "epoch": 2.69,
505
+ "learning_rate": 5.68039950062422e-06,
506
+ "loss": 0.2035,
507
+ "step": 800
508
+ },
509
+ {
510
+ "epoch": 2.72,
511
+ "learning_rate": 5.056179775280899e-06,
512
+ "loss": 0.2232,
513
+ "step": 810
514
+ },
515
+ {
516
+ "epoch": 2.76,
517
+ "learning_rate": 4.431960049937578e-06,
518
+ "loss": 0.1987,
519
+ "step": 820
520
+ },
521
+ {
522
+ "epoch": 2.79,
523
+ "learning_rate": 3.8077403245942577e-06,
524
+ "loss": 0.1939,
525
+ "step": 830
526
+ },
527
+ {
528
+ "epoch": 2.83,
529
+ "learning_rate": 3.1835205992509364e-06,
530
+ "loss": 0.1967,
531
+ "step": 840
532
+ },
533
+ {
534
+ "epoch": 2.86,
535
+ "learning_rate": 2.5593008739076156e-06,
536
+ "loss": 0.2037,
537
+ "step": 850
538
+ },
539
+ {
540
+ "epoch": 2.89,
541
+ "learning_rate": 1.9350811485642947e-06,
542
+ "loss": 0.1999,
543
+ "step": 860
544
+ },
545
+ {
546
+ "epoch": 2.93,
547
+ "learning_rate": 1.310861423220974e-06,
548
+ "loss": 0.1858,
549
+ "step": 870
550
+ },
551
+ {
552
+ "epoch": 2.96,
553
+ "learning_rate": 6.86641697877653e-07,
554
+ "loss": 0.2124,
555
+ "step": 880
556
+ },
557
+ {
558
+ "epoch": 2.99,
559
+ "learning_rate": 6.242197253433209e-08,
560
+ "loss": 0.2045,
561
+ "step": 890
562
+ },
563
+ {
564
+ "epoch": 3.0,
565
+ "eval_accuracy": 0.9594602592176126,
566
+ "eval_loss": 0.1155962273478508,
567
+ "eval_runtime": 329.7961,
568
+ "eval_samples_per_second": 51.235,
569
+ "eval_steps_per_second": 0.403,
570
+ "step": 891
571
+ },
572
+ {
573
+ "epoch": 3.0,
574
+ "step": 891,
575
+ "total_flos": 3.5333106012326707e+19,
576
+ "train_loss": 0.27140078694464514,
577
+ "train_runtime": 14713.6433,
578
+ "train_samples_per_second": 31.006,
579
+ "train_steps_per_second": 0.061
580
+ }
581
+ ],
582
+ "logging_steps": 10,
583
+ "max_steps": 891,
584
+ "num_train_epochs": 3,
585
+ "save_steps": 500,
586
+ "total_flos": 3.5333106012326707e+19,
587
+ "trial_name": null,
588
+ "trial_params": null
589
+ }