ALM-AHME commited on
Commit
6f9c3d3
1 Parent(s): 0d439c0

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 +631 -0
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.96,
3
+ "eval_accuracy": 0.9716500553709856,
4
+ "eval_loss": 0.0727071464061737,
5
+ "eval_runtime": 53.9726,
6
+ "eval_samples_per_second": 83.653,
7
+ "eval_steps_per_second": 5.243,
8
+ "total_flos": 8.138660625246413e+18,
9
+ "train_loss": 0.23672226468722027,
10
+ "train_runtime": 4894.1951,
11
+ "train_samples_per_second": 6.114,
12
+ "train_steps_per_second": 0.095
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.96,
3
+ "eval_accuracy": 0.9716500553709856,
4
+ "eval_loss": 0.0727071464061737,
5
+ "eval_runtime": 53.9726,
6
+ "eval_samples_per_second": 83.653,
7
+ "eval_steps_per_second": 5.243
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.96,
3
+ "total_flos": 8.138660625246413e+18,
4
+ "train_loss": 0.23672226468722027,
5
+ "train_runtime": 4894.1951,
6
+ "train_samples_per_second": 6.114,
7
+ "train_steps_per_second": 0.095
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,631 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9716500553709856,
3
+ "best_model_checkpoint": "beit-large-patch16-224-finetuned-LungCancer-Classification-LC25000-AH-40-30-30-Shuffled-3rd/checkpoint-465",
4
+ "epoch": 4.96,
5
+ "eval_steps": 500,
6
+ "global_step": 465,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.05,
13
+ "learning_rate": 1.072961373390558e-05,
14
+ "loss": 1.469,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.11,
19
+ "learning_rate": 2.145922746781116e-05,
20
+ "loss": 0.6329,
21
+ "step": 10
22
+ },
23
+ {
24
+ "epoch": 0.16,
25
+ "learning_rate": 3.2188841201716734e-05,
26
+ "loss": 0.2408,
27
+ "step": 15
28
+ },
29
+ {
30
+ "epoch": 0.21,
31
+ "learning_rate": 4.291845493562232e-05,
32
+ "loss": 0.2413,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 0.27,
37
+ "learning_rate": 5.36480686695279e-05,
38
+ "loss": 0.1766,
39
+ "step": 25
40
+ },
41
+ {
42
+ "epoch": 0.32,
43
+ "learning_rate": 6.437768240343347e-05,
44
+ "loss": 0.1375,
45
+ "step": 30
46
+ },
47
+ {
48
+ "epoch": 0.37,
49
+ "learning_rate": 7.510729613733907e-05,
50
+ "loss": 0.5266,
51
+ "step": 35
52
+ },
53
+ {
54
+ "epoch": 0.43,
55
+ "learning_rate": 8.583690987124464e-05,
56
+ "loss": 0.192,
57
+ "step": 40
58
+ },
59
+ {
60
+ "epoch": 0.48,
61
+ "learning_rate": 9.656652360515021e-05,
62
+ "loss": 0.2199,
63
+ "step": 45
64
+ },
65
+ {
66
+ "epoch": 0.53,
67
+ "learning_rate": 0.0001072961373390558,
68
+ "loss": 0.1779,
69
+ "step": 50
70
+ },
71
+ {
72
+ "epoch": 0.59,
73
+ "learning_rate": 0.00011802575107296138,
74
+ "loss": 0.1609,
75
+ "step": 55
76
+ },
77
+ {
78
+ "epoch": 0.64,
79
+ "learning_rate": 0.00012875536480686693,
80
+ "loss": 0.1332,
81
+ "step": 60
82
+ },
83
+ {
84
+ "epoch": 0.69,
85
+ "learning_rate": 0.00013948497854077252,
86
+ "loss": 0.122,
87
+ "step": 65
88
+ },
89
+ {
90
+ "epoch": 0.75,
91
+ "learning_rate": 0.00015021459227467814,
92
+ "loss": 0.1725,
93
+ "step": 70
94
+ },
95
+ {
96
+ "epoch": 0.8,
97
+ "learning_rate": 0.0001609442060085837,
98
+ "loss": 0.2223,
99
+ "step": 75
100
+ },
101
+ {
102
+ "epoch": 0.85,
103
+ "learning_rate": 0.00017167381974248928,
104
+ "loss": 0.2739,
105
+ "step": 80
106
+ },
107
+ {
108
+ "epoch": 0.91,
109
+ "learning_rate": 0.00018240343347639484,
110
+ "loss": 0.2776,
111
+ "step": 85
112
+ },
113
+ {
114
+ "epoch": 0.96,
115
+ "learning_rate": 0.00019313304721030043,
116
+ "loss": 0.3123,
117
+ "step": 90
118
+ },
119
+ {
120
+ "epoch": 0.99,
121
+ "eval_accuracy": 0.7769656699889258,
122
+ "eval_loss": 0.7103577852249146,
123
+ "eval_runtime": 1628.9879,
124
+ "eval_samples_per_second": 2.772,
125
+ "eval_steps_per_second": 0.174,
126
+ "step": 93
127
+ },
128
+ {
129
+ "epoch": 1.01,
130
+ "learning_rate": 0.00020386266094420602,
131
+ "loss": 0.387,
132
+ "step": 95
133
+ },
134
+ {
135
+ "epoch": 1.07,
136
+ "learning_rate": 0.0002145922746781116,
137
+ "loss": 0.1535,
138
+ "step": 100
139
+ },
140
+ {
141
+ "epoch": 1.12,
142
+ "learning_rate": 0.00022532188841201716,
143
+ "loss": 0.2921,
144
+ "step": 105
145
+ },
146
+ {
147
+ "epoch": 1.17,
148
+ "learning_rate": 0.00023605150214592275,
149
+ "loss": 0.1437,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 1.23,
154
+ "learning_rate": 0.0002467811158798283,
155
+ "loss": 0.2731,
156
+ "step": 115
157
+ },
158
+ {
159
+ "epoch": 1.28,
160
+ "learning_rate": 0.00025751072961373387,
161
+ "loss": 0.2365,
162
+ "step": 120
163
+ },
164
+ {
165
+ "epoch": 1.33,
166
+ "learning_rate": 0.0002682403433476395,
167
+ "loss": 0.1692,
168
+ "step": 125
169
+ },
170
+ {
171
+ "epoch": 1.39,
172
+ "learning_rate": 0.00027896995708154504,
173
+ "loss": 0.3372,
174
+ "step": 130
175
+ },
176
+ {
177
+ "epoch": 1.44,
178
+ "learning_rate": 0.00028969957081545066,
179
+ "loss": 0.2857,
180
+ "step": 135
181
+ },
182
+ {
183
+ "epoch": 1.49,
184
+ "learning_rate": 0.00030042918454935627,
185
+ "loss": 0.2505,
186
+ "step": 140
187
+ },
188
+ {
189
+ "epoch": 1.55,
190
+ "learning_rate": 0.00031115879828326183,
191
+ "loss": 0.2508,
192
+ "step": 145
193
+ },
194
+ {
195
+ "epoch": 1.6,
196
+ "learning_rate": 0.0003218884120171674,
197
+ "loss": 0.172,
198
+ "step": 150
199
+ },
200
+ {
201
+ "epoch": 1.65,
202
+ "learning_rate": 0.00033261802575107295,
203
+ "loss": 0.6477,
204
+ "step": 155
205
+ },
206
+ {
207
+ "epoch": 1.71,
208
+ "learning_rate": 0.00034334763948497857,
209
+ "loss": 0.3342,
210
+ "step": 160
211
+ },
212
+ {
213
+ "epoch": 1.76,
214
+ "learning_rate": 0.0003540772532188841,
215
+ "loss": 0.2687,
216
+ "step": 165
217
+ },
218
+ {
219
+ "epoch": 1.81,
220
+ "learning_rate": 0.0003648068669527897,
221
+ "loss": 0.1948,
222
+ "step": 170
223
+ },
224
+ {
225
+ "epoch": 1.87,
226
+ "learning_rate": 0.00037553648068669525,
227
+ "loss": 0.2025,
228
+ "step": 175
229
+ },
230
+ {
231
+ "epoch": 1.92,
232
+ "learning_rate": 0.00038626609442060086,
233
+ "loss": 0.3164,
234
+ "step": 180
235
+ },
236
+ {
237
+ "epoch": 1.97,
238
+ "learning_rate": 0.0003969957081545064,
239
+ "loss": 0.261,
240
+ "step": 185
241
+ },
242
+ {
243
+ "epoch": 1.99,
244
+ "eval_accuracy": 0.8172757475083057,
245
+ "eval_loss": 0.4562360942363739,
246
+ "eval_runtime": 54.0995,
247
+ "eval_samples_per_second": 83.457,
248
+ "eval_steps_per_second": 5.231,
249
+ "step": 187
250
+ },
251
+ {
252
+ "epoch": 2.03,
253
+ "learning_rate": 0.00040772532188841203,
254
+ "loss": 0.4598,
255
+ "step": 190
256
+ },
257
+ {
258
+ "epoch": 2.08,
259
+ "learning_rate": 0.00041845493562231765,
260
+ "loss": 0.668,
261
+ "step": 195
262
+ },
263
+ {
264
+ "epoch": 2.13,
265
+ "learning_rate": 0.0004291845493562232,
266
+ "loss": 0.3559,
267
+ "step": 200
268
+ },
269
+ {
270
+ "epoch": 2.19,
271
+ "learning_rate": 0.00043991416309012877,
272
+ "loss": 0.3004,
273
+ "step": 205
274
+ },
275
+ {
276
+ "epoch": 2.24,
277
+ "learning_rate": 0.0004506437768240343,
278
+ "loss": 0.3047,
279
+ "step": 210
280
+ },
281
+ {
282
+ "epoch": 2.29,
283
+ "learning_rate": 0.00046137339055793994,
284
+ "loss": 0.2776,
285
+ "step": 215
286
+ },
287
+ {
288
+ "epoch": 2.35,
289
+ "learning_rate": 0.0004721030042918455,
290
+ "loss": 0.3193,
291
+ "step": 220
292
+ },
293
+ {
294
+ "epoch": 2.4,
295
+ "learning_rate": 0.00048283261802575106,
296
+ "loss": 0.25,
297
+ "step": 225
298
+ },
299
+ {
300
+ "epoch": 2.45,
301
+ "learning_rate": 0.0004935622317596566,
302
+ "loss": 0.2472,
303
+ "step": 230
304
+ },
305
+ {
306
+ "epoch": 2.51,
307
+ "learning_rate": 0.0004956896551724138,
308
+ "loss": 0.2343,
309
+ "step": 235
310
+ },
311
+ {
312
+ "epoch": 2.56,
313
+ "learning_rate": 0.0004849137931034483,
314
+ "loss": 0.1439,
315
+ "step": 240
316
+ },
317
+ {
318
+ "epoch": 2.61,
319
+ "learning_rate": 0.00047413793103448276,
320
+ "loss": 0.2208,
321
+ "step": 245
322
+ },
323
+ {
324
+ "epoch": 2.67,
325
+ "learning_rate": 0.00046336206896551726,
326
+ "loss": 0.2339,
327
+ "step": 250
328
+ },
329
+ {
330
+ "epoch": 2.72,
331
+ "learning_rate": 0.0004525862068965517,
332
+ "loss": 0.1704,
333
+ "step": 255
334
+ },
335
+ {
336
+ "epoch": 2.77,
337
+ "learning_rate": 0.0004418103448275862,
338
+ "loss": 0.2218,
339
+ "step": 260
340
+ },
341
+ {
342
+ "epoch": 2.83,
343
+ "learning_rate": 0.0004310344827586207,
344
+ "loss": 0.1341,
345
+ "step": 265
346
+ },
347
+ {
348
+ "epoch": 2.88,
349
+ "learning_rate": 0.0004202586206896552,
350
+ "loss": 0.2466,
351
+ "step": 270
352
+ },
353
+ {
354
+ "epoch": 2.93,
355
+ "learning_rate": 0.00040948275862068967,
356
+ "loss": 0.2176,
357
+ "step": 275
358
+ },
359
+ {
360
+ "epoch": 2.99,
361
+ "learning_rate": 0.00039870689655172416,
362
+ "loss": 0.2012,
363
+ "step": 280
364
+ },
365
+ {
366
+ "epoch": 3.0,
367
+ "eval_accuracy": 0.9508305647840531,
368
+ "eval_loss": 0.12906160950660706,
369
+ "eval_runtime": 54.2749,
370
+ "eval_samples_per_second": 83.188,
371
+ "eval_steps_per_second": 5.214,
372
+ "step": 281
373
+ },
374
+ {
375
+ "epoch": 3.04,
376
+ "learning_rate": 0.0003879310344827586,
377
+ "loss": 0.2101,
378
+ "step": 285
379
+ },
380
+ {
381
+ "epoch": 3.09,
382
+ "learning_rate": 0.0003771551724137931,
383
+ "loss": 0.1814,
384
+ "step": 290
385
+ },
386
+ {
387
+ "epoch": 3.15,
388
+ "learning_rate": 0.0003663793103448276,
389
+ "loss": 0.1757,
390
+ "step": 295
391
+ },
392
+ {
393
+ "epoch": 3.2,
394
+ "learning_rate": 0.00035560344827586203,
395
+ "loss": 0.1658,
396
+ "step": 300
397
+ },
398
+ {
399
+ "epoch": 3.25,
400
+ "learning_rate": 0.0003448275862068966,
401
+ "loss": 0.1867,
402
+ "step": 305
403
+ },
404
+ {
405
+ "epoch": 3.31,
406
+ "learning_rate": 0.0003340517241379311,
407
+ "loss": 0.189,
408
+ "step": 310
409
+ },
410
+ {
411
+ "epoch": 3.36,
412
+ "learning_rate": 0.0003232758620689655,
413
+ "loss": 0.2099,
414
+ "step": 315
415
+ },
416
+ {
417
+ "epoch": 3.41,
418
+ "learning_rate": 0.0003125,
419
+ "loss": 0.1683,
420
+ "step": 320
421
+ },
422
+ {
423
+ "epoch": 3.47,
424
+ "learning_rate": 0.0003017241379310345,
425
+ "loss": 0.2641,
426
+ "step": 325
427
+ },
428
+ {
429
+ "epoch": 3.52,
430
+ "learning_rate": 0.00029094827586206894,
431
+ "loss": 0.191,
432
+ "step": 330
433
+ },
434
+ {
435
+ "epoch": 3.57,
436
+ "learning_rate": 0.00028017241379310343,
437
+ "loss": 0.2154,
438
+ "step": 335
439
+ },
440
+ {
441
+ "epoch": 3.63,
442
+ "learning_rate": 0.000269396551724138,
443
+ "loss": 0.1961,
444
+ "step": 340
445
+ },
446
+ {
447
+ "epoch": 3.68,
448
+ "learning_rate": 0.0002586206896551724,
449
+ "loss": 0.193,
450
+ "step": 345
451
+ },
452
+ {
453
+ "epoch": 3.73,
454
+ "learning_rate": 0.0002478448275862069,
455
+ "loss": 0.2128,
456
+ "step": 350
457
+ },
458
+ {
459
+ "epoch": 3.79,
460
+ "learning_rate": 0.00023706896551724138,
461
+ "loss": 0.1744,
462
+ "step": 355
463
+ },
464
+ {
465
+ "epoch": 3.84,
466
+ "learning_rate": 0.00022629310344827585,
467
+ "loss": 0.1067,
468
+ "step": 360
469
+ },
470
+ {
471
+ "epoch": 3.89,
472
+ "learning_rate": 0.00021551724137931034,
473
+ "loss": 0.1643,
474
+ "step": 365
475
+ },
476
+ {
477
+ "epoch": 3.95,
478
+ "learning_rate": 0.00020474137931034484,
479
+ "loss": 0.2163,
480
+ "step": 370
481
+ },
482
+ {
483
+ "epoch": 4.0,
484
+ "learning_rate": 0.0001939655172413793,
485
+ "loss": 0.1424,
486
+ "step": 375
487
+ },
488
+ {
489
+ "epoch": 4.0,
490
+ "eval_accuracy": 0.9508305647840531,
491
+ "eval_loss": 0.13319291174411774,
492
+ "eval_runtime": 54.0858,
493
+ "eval_samples_per_second": 83.479,
494
+ "eval_steps_per_second": 5.232,
495
+ "step": 375
496
+ },
497
+ {
498
+ "epoch": 4.05,
499
+ "learning_rate": 0.0001831896551724138,
500
+ "loss": 0.1964,
501
+ "step": 380
502
+ },
503
+ {
504
+ "epoch": 4.11,
505
+ "learning_rate": 0.0001724137931034483,
506
+ "loss": 0.179,
507
+ "step": 385
508
+ },
509
+ {
510
+ "epoch": 4.16,
511
+ "learning_rate": 0.00016163793103448276,
512
+ "loss": 0.1167,
513
+ "step": 390
514
+ },
515
+ {
516
+ "epoch": 4.21,
517
+ "learning_rate": 0.00015086206896551725,
518
+ "loss": 0.1361,
519
+ "step": 395
520
+ },
521
+ {
522
+ "epoch": 4.27,
523
+ "learning_rate": 0.00014008620689655172,
524
+ "loss": 0.1418,
525
+ "step": 400
526
+ },
527
+ {
528
+ "epoch": 4.32,
529
+ "learning_rate": 0.0001293103448275862,
530
+ "loss": 0.1664,
531
+ "step": 405
532
+ },
533
+ {
534
+ "epoch": 4.37,
535
+ "learning_rate": 0.00011853448275862069,
536
+ "loss": 0.1295,
537
+ "step": 410
538
+ },
539
+ {
540
+ "epoch": 4.43,
541
+ "learning_rate": 0.00010775862068965517,
542
+ "loss": 0.1185,
543
+ "step": 415
544
+ },
545
+ {
546
+ "epoch": 4.48,
547
+ "learning_rate": 9.698275862068965e-05,
548
+ "loss": 0.1278,
549
+ "step": 420
550
+ },
551
+ {
552
+ "epoch": 4.53,
553
+ "learning_rate": 8.620689655172414e-05,
554
+ "loss": 0.1287,
555
+ "step": 425
556
+ },
557
+ {
558
+ "epoch": 4.59,
559
+ "learning_rate": 7.543103448275863e-05,
560
+ "loss": 0.1168,
561
+ "step": 430
562
+ },
563
+ {
564
+ "epoch": 4.64,
565
+ "learning_rate": 6.46551724137931e-05,
566
+ "loss": 0.1142,
567
+ "step": 435
568
+ },
569
+ {
570
+ "epoch": 4.69,
571
+ "learning_rate": 5.3879310344827585e-05,
572
+ "loss": 0.1276,
573
+ "step": 440
574
+ },
575
+ {
576
+ "epoch": 4.75,
577
+ "learning_rate": 4.310344827586207e-05,
578
+ "loss": 0.127,
579
+ "step": 445
580
+ },
581
+ {
582
+ "epoch": 4.8,
583
+ "learning_rate": 3.232758620689655e-05,
584
+ "loss": 0.1219,
585
+ "step": 450
586
+ },
587
+ {
588
+ "epoch": 4.85,
589
+ "learning_rate": 2.1551724137931036e-05,
590
+ "loss": 0.1023,
591
+ "step": 455
592
+ },
593
+ {
594
+ "epoch": 4.91,
595
+ "learning_rate": 1.0775862068965518e-05,
596
+ "loss": 0.1329,
597
+ "step": 460
598
+ },
599
+ {
600
+ "epoch": 4.96,
601
+ "learning_rate": 0.0,
602
+ "loss": 0.0949,
603
+ "step": 465
604
+ },
605
+ {
606
+ "epoch": 4.96,
607
+ "eval_accuracy": 0.9716500553709856,
608
+ "eval_loss": 0.0727071464061737,
609
+ "eval_runtime": 54.0022,
610
+ "eval_samples_per_second": 83.608,
611
+ "eval_steps_per_second": 5.241,
612
+ "step": 465
613
+ },
614
+ {
615
+ "epoch": 4.96,
616
+ "step": 465,
617
+ "total_flos": 8.138660625246413e+18,
618
+ "train_loss": 0.23672226468722027,
619
+ "train_runtime": 4894.1951,
620
+ "train_samples_per_second": 6.114,
621
+ "train_steps_per_second": 0.095
622
+ }
623
+ ],
624
+ "logging_steps": 5,
625
+ "max_steps": 465,
626
+ "num_train_epochs": 5,
627
+ "save_steps": 500,
628
+ "total_flos": 8.138660625246413e+18,
629
+ "trial_name": null,
630
+ "trial_params": null
631
+ }