JEdward7777 commited on
Commit
9f4ff4f
1 Parent(s): 3169fac

End of training

Browse files
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
- "epoch": 59.89,
3
- "eval_accuracy": 0.96875,
4
- "eval_loss": 0.18353550136089325,
5
- "eval_runtime": 4.1792,
6
- "eval_samples_per_second": 15.314,
7
- "eval_steps_per_second": 0.479,
8
- "total_flos": 8.516851567051162e+17,
9
- "train_loss": 0.5424989521503448,
10
- "train_runtime": 4233.7913,
11
- "train_samples_per_second": 8.106,
12
  "train_steps_per_second": 0.057
13
  }
 
1
  {
2
+ "epoch": 59.84,
3
+ "eval_accuracy": 0.9846153846153847,
4
+ "eval_loss": 0.1295621246099472,
5
+ "eval_runtime": 4.3947,
6
+ "eval_samples_per_second": 14.791,
7
+ "eval_steps_per_second": 0.683,
8
+ "total_flos": 8.648855308501955e+17,
9
+ "train_loss": 0.577343902985255,
10
+ "train_runtime": 4213.9216,
11
+ "train_samples_per_second": 8.273,
12
  "train_steps_per_second": 0.057
13
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 59.89,
3
- "eval_accuracy": 0.96875,
4
- "eval_loss": 0.18353550136089325,
5
- "eval_runtime": 4.1792,
6
- "eval_samples_per_second": 15.314,
7
- "eval_steps_per_second": 0.479
8
  }
 
1
  {
2
+ "epoch": 59.84,
3
+ "eval_accuracy": 0.9846153846153847,
4
+ "eval_loss": 0.1295621246099472,
5
+ "eval_runtime": 4.3947,
6
+ "eval_samples_per_second": 14.791,
7
+ "eval_steps_per_second": 0.683
8
  }
runs/Jan05_13-26-49_78b464702e4c/events.out.tfevents.1672929710.78b464702e4c.573.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf84a9cee06528ecf79659bde56b701c98d8e8a6a52cf3659e0abcbe033c4b70
3
+ size 363
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 59.89,
3
- "total_flos": 8.516851567051162e+17,
4
- "train_loss": 0.5424989521503448,
5
- "train_runtime": 4233.7913,
6
- "train_samples_per_second": 8.106,
7
  "train_steps_per_second": 0.057
8
  }
 
1
  {
2
+ "epoch": 59.84,
3
+ "total_flos": 8.648855308501955e+17,
4
+ "train_loss": 0.577343902985255,
5
+ "train_runtime": 4213.9216,
6
+ "train_samples_per_second": 8.273,
7
  "train_steps_per_second": 0.057
8
  }
trainer_state.json CHANGED
@@ -1,637 +1,637 @@
1
  {
2
- "best_metric": 0.96875,
3
- "best_model_checkpoint": "delivery_truck_classification/checkpoint-52",
4
- "epoch": 59.888888888888886,
5
  "global_step": 240,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
- "epoch": 0.89,
12
- "eval_accuracy": 0.15625,
13
- "eval_loss": 2.007406234741211,
14
- "eval_runtime": 4.1284,
15
- "eval_samples_per_second": 15.502,
16
- "eval_steps_per_second": 0.484,
17
  "step": 4
18
  },
19
  {
20
- "epoch": 1.89,
21
- "eval_accuracy": 0.25,
22
- "eval_loss": 1.8895925283432007,
23
- "eval_runtime": 4.9483,
24
- "eval_samples_per_second": 12.934,
25
- "eval_steps_per_second": 0.404,
26
  "step": 8
27
  },
28
  {
29
- "epoch": 2.89,
30
- "eval_accuracy": 0.40625,
31
- "eval_loss": 1.7421282529830933,
32
- "eval_runtime": 4.1039,
33
- "eval_samples_per_second": 15.595,
34
- "eval_steps_per_second": 0.487,
35
  "step": 12
36
  },
37
  {
38
- "epoch": 3.89,
39
- "eval_accuracy": 0.4375,
40
- "eval_loss": 1.5891958475112915,
41
- "eval_runtime": 4.5489,
42
- "eval_samples_per_second": 14.069,
43
- "eval_steps_per_second": 0.44,
44
  "step": 16
45
  },
46
  {
47
- "epoch": 4.89,
48
  "learning_rate": 4.166666666666667e-05,
49
- "loss": 1.973,
50
  "step": 20
51
  },
52
  {
53
- "epoch": 4.89,
54
- "eval_accuracy": 0.609375,
55
- "eval_loss": 1.362261176109314,
56
- "eval_runtime": 4.1743,
57
- "eval_samples_per_second": 15.332,
58
- "eval_steps_per_second": 0.479,
59
  "step": 20
60
  },
61
  {
62
- "epoch": 5.89,
63
- "eval_accuracy": 0.609375,
64
- "eval_loss": 1.1093113422393799,
65
- "eval_runtime": 4.4702,
66
- "eval_samples_per_second": 14.317,
67
- "eval_steps_per_second": 0.447,
68
  "step": 24
69
  },
70
  {
71
- "epoch": 6.89,
72
- "eval_accuracy": 0.78125,
73
- "eval_loss": 0.7900974154472351,
74
- "eval_runtime": 4.1973,
75
- "eval_samples_per_second": 15.248,
76
- "eval_steps_per_second": 0.476,
77
  "step": 28
78
  },
79
  {
80
- "epoch": 7.89,
81
- "eval_accuracy": 0.84375,
82
- "eval_loss": 0.5773038268089294,
83
- "eval_runtime": 4.6009,
84
- "eval_samples_per_second": 13.91,
85
- "eval_steps_per_second": 0.435,
86
  "step": 32
87
  },
88
  {
89
- "epoch": 8.89,
90
- "eval_accuracy": 0.890625,
91
- "eval_loss": 0.38568082451820374,
92
- "eval_runtime": 4.1614,
93
- "eval_samples_per_second": 15.379,
94
- "eval_steps_per_second": 0.481,
95
  "step": 36
96
  },
97
  {
98
- "epoch": 9.89,
99
  "learning_rate": 4.62962962962963e-05,
100
- "loss": 1.0433,
101
  "step": 40
102
  },
103
  {
104
- "epoch": 9.89,
105
- "eval_accuracy": 0.90625,
106
- "eval_loss": 0.32543647289276123,
107
- "eval_runtime": 4.5756,
108
- "eval_samples_per_second": 13.987,
109
- "eval_steps_per_second": 0.437,
110
  "step": 40
111
  },
112
  {
113
- "epoch": 10.89,
114
- "eval_accuracy": 0.921875,
115
- "eval_loss": 0.2461281269788742,
116
- "eval_runtime": 4.2012,
117
- "eval_samples_per_second": 15.234,
118
- "eval_steps_per_second": 0.476,
119
  "step": 44
120
  },
121
  {
122
- "epoch": 11.89,
123
- "eval_accuracy": 0.921875,
124
- "eval_loss": 0.2339743673801422,
125
- "eval_runtime": 4.5321,
126
- "eval_samples_per_second": 14.121,
127
- "eval_steps_per_second": 0.441,
128
  "step": 48
129
  },
130
  {
131
- "epoch": 12.89,
132
- "eval_accuracy": 0.96875,
133
- "eval_loss": 0.18353550136089325,
134
- "eval_runtime": 4.3055,
135
- "eval_samples_per_second": 14.865,
136
- "eval_steps_per_second": 0.465,
137
  "step": 52
138
  },
139
  {
140
- "epoch": 13.89,
141
- "eval_accuracy": 0.9375,
142
- "eval_loss": 0.17793762683868408,
143
- "eval_runtime": 4.4643,
144
- "eval_samples_per_second": 14.336,
145
- "eval_steps_per_second": 0.448,
146
  "step": 56
147
  },
148
  {
149
- "epoch": 14.89,
150
  "learning_rate": 4.166666666666667e-05,
151
- "loss": 0.5842,
152
  "step": 60
153
  },
154
  {
155
- "epoch": 14.89,
156
- "eval_accuracy": 0.953125,
157
- "eval_loss": 0.15445110201835632,
158
- "eval_runtime": 4.1434,
159
- "eval_samples_per_second": 15.446,
160
- "eval_steps_per_second": 0.483,
161
  "step": 60
162
  },
163
  {
164
- "epoch": 15.89,
165
- "eval_accuracy": 0.953125,
166
- "eval_loss": 0.14869830012321472,
167
- "eval_runtime": 4.5456,
168
- "eval_samples_per_second": 14.08,
169
- "eval_steps_per_second": 0.44,
170
  "step": 64
171
  },
172
  {
173
- "epoch": 16.89,
174
- "eval_accuracy": 0.921875,
175
- "eval_loss": 0.19956862926483154,
176
- "eval_runtime": 4.0021,
177
- "eval_samples_per_second": 15.992,
178
- "eval_steps_per_second": 0.5,
179
  "step": 68
180
  },
181
  {
182
- "epoch": 17.89,
183
- "eval_accuracy": 0.90625,
184
- "eval_loss": 0.1619431972503662,
185
- "eval_runtime": 4.478,
186
- "eval_samples_per_second": 14.292,
187
- "eval_steps_per_second": 0.447,
188
  "step": 72
189
  },
190
  {
191
- "epoch": 18.89,
192
- "eval_accuracy": 0.96875,
193
- "eval_loss": 0.13495692610740662,
194
- "eval_runtime": 4.1244,
195
- "eval_samples_per_second": 15.517,
196
- "eval_steps_per_second": 0.485,
197
  "step": 76
198
  },
199
  {
200
- "epoch": 19.89,
201
  "learning_rate": 3.7037037037037037e-05,
202
- "loss": 0.4616,
203
  "step": 80
204
  },
205
  {
206
- "epoch": 19.89,
207
- "eval_accuracy": 0.9375,
208
- "eval_loss": 0.17057114839553833,
209
- "eval_runtime": 4.5586,
210
- "eval_samples_per_second": 14.039,
211
- "eval_steps_per_second": 0.439,
212
  "step": 80
213
  },
214
  {
215
- "epoch": 20.89,
216
- "eval_accuracy": 0.921875,
217
- "eval_loss": 0.15793055295944214,
218
- "eval_runtime": 4.126,
219
- "eval_samples_per_second": 15.511,
220
- "eval_steps_per_second": 0.485,
221
  "step": 84
222
  },
223
  {
224
- "epoch": 21.89,
225
- "eval_accuracy": 0.9375,
226
- "eval_loss": 0.16296246647834778,
227
- "eval_runtime": 4.5173,
228
- "eval_samples_per_second": 14.168,
229
- "eval_steps_per_second": 0.443,
230
  "step": 88
231
  },
232
  {
233
- "epoch": 22.89,
234
- "eval_accuracy": 0.90625,
235
- "eval_loss": 0.20802505314350128,
236
- "eval_runtime": 4.0833,
237
- "eval_samples_per_second": 15.674,
238
- "eval_steps_per_second": 0.49,
239
  "step": 92
240
  },
241
  {
242
- "epoch": 23.89,
243
- "eval_accuracy": 0.9375,
244
- "eval_loss": 0.14634451270103455,
245
- "eval_runtime": 4.6059,
246
- "eval_samples_per_second": 13.895,
247
- "eval_steps_per_second": 0.434,
248
  "step": 96
249
  },
250
  {
251
- "epoch": 24.89,
252
  "learning_rate": 3.240740740740741e-05,
253
- "loss": 0.3898,
254
  "step": 100
255
  },
256
  {
257
- "epoch": 24.89,
258
- "eval_accuracy": 0.96875,
259
- "eval_loss": 0.11846979707479477,
260
- "eval_runtime": 4.1451,
261
- "eval_samples_per_second": 15.44,
262
- "eval_steps_per_second": 0.483,
263
  "step": 100
264
  },
265
  {
266
- "epoch": 25.89,
267
- "eval_accuracy": 0.921875,
268
- "eval_loss": 0.14448919892311096,
269
- "eval_runtime": 4.4782,
270
- "eval_samples_per_second": 14.291,
271
- "eval_steps_per_second": 0.447,
272
  "step": 104
273
  },
274
  {
275
- "epoch": 26.89,
276
- "eval_accuracy": 0.921875,
277
- "eval_loss": 0.20514565706253052,
278
- "eval_runtime": 4.0807,
279
- "eval_samples_per_second": 15.683,
280
- "eval_steps_per_second": 0.49,
281
  "step": 108
282
  },
283
  {
284
- "epoch": 27.89,
285
- "eval_accuracy": 0.9375,
286
- "eval_loss": 0.19281229376792908,
287
- "eval_runtime": 4.457,
288
- "eval_samples_per_second": 14.36,
289
- "eval_steps_per_second": 0.449,
290
  "step": 112
291
  },
292
  {
293
- "epoch": 28.89,
294
- "eval_accuracy": 0.9375,
295
- "eval_loss": 0.13649864494800568,
296
- "eval_runtime": 4.038,
297
- "eval_samples_per_second": 15.849,
298
- "eval_steps_per_second": 0.495,
299
  "step": 116
300
  },
301
  {
302
- "epoch": 29.89,
303
  "learning_rate": 2.777777777777778e-05,
304
- "loss": 0.3511,
305
  "step": 120
306
  },
307
  {
308
- "epoch": 29.89,
309
- "eval_accuracy": 0.953125,
310
- "eval_loss": 0.10572843253612518,
311
- "eval_runtime": 4.5443,
312
- "eval_samples_per_second": 14.084,
313
- "eval_steps_per_second": 0.44,
314
  "step": 120
315
  },
316
  {
317
- "epoch": 30.89,
318
- "eval_accuracy": 0.953125,
319
- "eval_loss": 0.10910199582576752,
320
- "eval_runtime": 4.2535,
321
- "eval_samples_per_second": 15.046,
322
- "eval_steps_per_second": 0.47,
323
  "step": 124
324
  },
325
  {
326
- "epoch": 31.89,
327
- "eval_accuracy": 0.9375,
328
- "eval_loss": 0.1893911063671112,
329
- "eval_runtime": 4.6955,
330
- "eval_samples_per_second": 13.63,
331
- "eval_steps_per_second": 0.426,
332
  "step": 128
333
  },
334
  {
335
- "epoch": 32.89,
336
- "eval_accuracy": 0.953125,
337
- "eval_loss": 0.12075338512659073,
338
- "eval_runtime": 4.1214,
339
- "eval_samples_per_second": 15.529,
340
- "eval_steps_per_second": 0.485,
341
  "step": 132
342
  },
343
  {
344
- "epoch": 33.89,
345
- "eval_accuracy": 0.96875,
346
- "eval_loss": 0.11010686308145523,
347
- "eval_runtime": 4.5267,
348
- "eval_samples_per_second": 14.138,
349
- "eval_steps_per_second": 0.442,
350
  "step": 136
351
  },
352
  {
353
- "epoch": 34.89,
354
  "learning_rate": 2.314814814814815e-05,
355
- "loss": 0.3286,
356
  "step": 140
357
  },
358
  {
359
- "epoch": 34.89,
360
- "eval_accuracy": 0.9375,
361
- "eval_loss": 0.14093339443206787,
362
- "eval_runtime": 4.119,
363
- "eval_samples_per_second": 15.538,
364
- "eval_steps_per_second": 0.486,
365
  "step": 140
366
  },
367
  {
368
- "epoch": 35.89,
369
- "eval_accuracy": 0.921875,
370
- "eval_loss": 0.18304279446601868,
371
- "eval_runtime": 4.4932,
372
- "eval_samples_per_second": 14.244,
373
- "eval_steps_per_second": 0.445,
374
  "step": 144
375
  },
376
  {
377
- "epoch": 36.89,
378
- "eval_accuracy": 0.921875,
379
- "eval_loss": 0.1518746018409729,
380
- "eval_runtime": 4.1186,
381
- "eval_samples_per_second": 15.539,
382
- "eval_steps_per_second": 0.486,
383
  "step": 148
384
  },
385
  {
386
- "epoch": 37.89,
387
- "eval_accuracy": 0.953125,
388
- "eval_loss": 0.10309180617332458,
389
- "eval_runtime": 4.5179,
390
- "eval_samples_per_second": 14.166,
391
- "eval_steps_per_second": 0.443,
392
  "step": 152
393
  },
394
  {
395
- "epoch": 38.89,
396
- "eval_accuracy": 0.96875,
397
- "eval_loss": 0.09621511399745941,
398
- "eval_runtime": 4.2076,
399
- "eval_samples_per_second": 15.211,
400
- "eval_steps_per_second": 0.475,
401
  "step": 156
402
  },
403
  {
404
- "epoch": 39.89,
405
  "learning_rate": 1.8518518518518518e-05,
406
- "loss": 0.3095,
407
  "step": 160
408
  },
409
  {
410
- "epoch": 39.89,
411
- "eval_accuracy": 0.96875,
412
- "eval_loss": 0.09031268954277039,
413
- "eval_runtime": 4.5982,
414
- "eval_samples_per_second": 13.919,
415
- "eval_steps_per_second": 0.435,
416
  "step": 160
417
  },
418
  {
419
- "epoch": 40.89,
420
- "eval_accuracy": 0.96875,
421
- "eval_loss": 0.08859497308731079,
422
- "eval_runtime": 4.1937,
423
- "eval_samples_per_second": 15.261,
424
- "eval_steps_per_second": 0.477,
425
  "step": 164
426
  },
427
  {
428
- "epoch": 41.89,
429
- "eval_accuracy": 0.96875,
430
- "eval_loss": 0.10331027209758759,
431
- "eval_runtime": 4.5829,
432
- "eval_samples_per_second": 13.965,
433
- "eval_steps_per_second": 0.436,
434
  "step": 168
435
  },
436
  {
437
- "epoch": 42.89,
438
- "eval_accuracy": 0.953125,
439
- "eval_loss": 0.11172451823949814,
440
- "eval_runtime": 4.2291,
441
- "eval_samples_per_second": 15.133,
442
- "eval_steps_per_second": 0.473,
443
  "step": 172
444
  },
445
  {
446
- "epoch": 43.89,
447
- "eval_accuracy": 0.9375,
448
- "eval_loss": 0.11921519041061401,
449
- "eval_runtime": 4.595,
450
- "eval_samples_per_second": 13.928,
451
- "eval_steps_per_second": 0.435,
452
  "step": 176
453
  },
454
  {
455
- "epoch": 44.89,
456
  "learning_rate": 1.388888888888889e-05,
457
- "loss": 0.3056,
458
  "step": 180
459
  },
460
  {
461
- "epoch": 44.89,
462
- "eval_accuracy": 0.953125,
463
- "eval_loss": 0.09840899705886841,
464
- "eval_runtime": 4.2775,
465
- "eval_samples_per_second": 14.962,
466
- "eval_steps_per_second": 0.468,
467
  "step": 180
468
  },
469
  {
470
- "epoch": 45.89,
471
- "eval_accuracy": 0.953125,
472
- "eval_loss": 0.08203314244747162,
473
- "eval_runtime": 4.5543,
474
- "eval_samples_per_second": 14.053,
475
- "eval_steps_per_second": 0.439,
476
  "step": 184
477
  },
478
  {
479
- "epoch": 46.89,
480
- "eval_accuracy": 0.953125,
481
- "eval_loss": 0.08574767410755157,
482
- "eval_runtime": 4.1553,
483
- "eval_samples_per_second": 15.402,
484
- "eval_steps_per_second": 0.481,
485
  "step": 188
486
  },
487
  {
488
- "epoch": 47.89,
489
- "eval_accuracy": 0.953125,
490
- "eval_loss": 0.10580457001924515,
491
- "eval_runtime": 4.6204,
492
- "eval_samples_per_second": 13.852,
493
- "eval_steps_per_second": 0.433,
494
  "step": 192
495
  },
496
  {
497
- "epoch": 48.89,
498
- "eval_accuracy": 0.9375,
499
- "eval_loss": 0.11630271375179291,
500
- "eval_runtime": 4.1519,
501
- "eval_samples_per_second": 15.415,
502
- "eval_steps_per_second": 0.482,
503
  "step": 196
504
  },
505
  {
506
- "epoch": 49.89,
507
  "learning_rate": 9.259259259259259e-06,
508
- "loss": 0.255,
509
  "step": 200
510
  },
511
  {
512
- "epoch": 49.89,
513
- "eval_accuracy": 0.953125,
514
- "eval_loss": 0.11214150488376617,
515
- "eval_runtime": 4.4888,
516
- "eval_samples_per_second": 14.258,
517
- "eval_steps_per_second": 0.446,
518
  "step": 200
519
  },
520
  {
521
- "epoch": 50.89,
522
- "eval_accuracy": 0.96875,
523
- "eval_loss": 0.10038581490516663,
524
- "eval_runtime": 4.1624,
525
- "eval_samples_per_second": 15.376,
526
- "eval_steps_per_second": 0.48,
527
  "step": 204
528
  },
529
  {
530
- "epoch": 51.89,
531
- "eval_accuracy": 0.96875,
532
- "eval_loss": 0.09537401050329208,
533
- "eval_runtime": 4.5154,
534
- "eval_samples_per_second": 14.174,
535
- "eval_steps_per_second": 0.443,
536
  "step": 208
537
  },
538
  {
539
- "epoch": 52.89,
540
- "eval_accuracy": 0.96875,
541
- "eval_loss": 0.0925481989979744,
542
- "eval_runtime": 4.141,
543
- "eval_samples_per_second": 15.455,
544
- "eval_steps_per_second": 0.483,
545
  "step": 212
546
  },
547
  {
548
- "epoch": 53.89,
549
- "eval_accuracy": 0.96875,
550
- "eval_loss": 0.08920943737030029,
551
- "eval_runtime": 4.5736,
552
- "eval_samples_per_second": 13.993,
553
- "eval_steps_per_second": 0.437,
554
  "step": 216
555
  },
556
  {
557
- "epoch": 54.89,
558
  "learning_rate": 4.6296296296296296e-06,
559
- "loss": 0.2494,
560
  "step": 220
561
  },
562
  {
563
- "epoch": 54.89,
564
- "eval_accuracy": 0.96875,
565
- "eval_loss": 0.08926128596067429,
566
- "eval_runtime": 4.0979,
567
- "eval_samples_per_second": 15.618,
568
- "eval_steps_per_second": 0.488,
569
  "step": 220
570
  },
571
  {
572
- "epoch": 55.89,
573
- "eval_accuracy": 0.96875,
574
- "eval_loss": 0.0900685042142868,
575
- "eval_runtime": 4.5713,
576
- "eval_samples_per_second": 14.001,
577
- "eval_steps_per_second": 0.438,
578
  "step": 224
579
  },
580
  {
581
- "epoch": 56.89,
582
- "eval_accuracy": 0.96875,
583
- "eval_loss": 0.08957040309906006,
584
- "eval_runtime": 4.2031,
585
- "eval_samples_per_second": 15.227,
586
- "eval_steps_per_second": 0.476,
587
  "step": 228
588
  },
589
  {
590
- "epoch": 57.89,
591
- "eval_accuracy": 0.96875,
592
- "eval_loss": 0.09029233455657959,
593
- "eval_runtime": 4.5507,
594
- "eval_samples_per_second": 14.064,
595
- "eval_steps_per_second": 0.439,
596
  "step": 232
597
  },
598
  {
599
- "epoch": 58.89,
600
- "eval_accuracy": 0.96875,
601
- "eval_loss": 0.09133943915367126,
602
- "eval_runtime": 4.5211,
603
- "eval_samples_per_second": 14.156,
604
- "eval_steps_per_second": 0.442,
605
  "step": 236
606
  },
607
  {
608
- "epoch": 59.89,
609
  "learning_rate": 0.0,
610
- "loss": 0.2588,
611
  "step": 240
612
  },
613
  {
614
- "epoch": 59.89,
615
- "eval_accuracy": 0.96875,
616
- "eval_loss": 0.09184324741363525,
617
- "eval_runtime": 4.5891,
618
- "eval_samples_per_second": 13.946,
619
- "eval_steps_per_second": 0.436,
620
  "step": 240
621
  },
622
  {
623
- "epoch": 59.89,
624
  "step": 240,
625
- "total_flos": 8.516851567051162e+17,
626
- "train_loss": 0.5424989521503448,
627
- "train_runtime": 4233.7913,
628
- "train_samples_per_second": 8.106,
629
  "train_steps_per_second": 0.057
630
  }
631
  ],
632
  "max_steps": 240,
633
  "num_train_epochs": 60,
634
- "total_flos": 8.516851567051162e+17,
635
  "trial_name": null,
636
  "trial_params": null
637
  }
 
1
  {
2
+ "best_metric": 0.9846153846153847,
3
+ "best_model_checkpoint": "delivery_truck_classification/checkpoint-104",
4
+ "epoch": 59.8421052631579,
5
  "global_step": 240,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
+ "epoch": 0.84,
12
+ "eval_accuracy": 0.3230769230769231,
13
+ "eval_loss": 1.819914698600769,
14
+ "eval_runtime": 4.4893,
15
+ "eval_samples_per_second": 14.479,
16
+ "eval_steps_per_second": 0.668,
17
  "step": 4
18
  },
19
  {
20
+ "epoch": 1.84,
21
+ "eval_accuracy": 0.4153846153846154,
22
+ "eval_loss": 1.7274595499038696,
23
+ "eval_runtime": 5.1038,
24
+ "eval_samples_per_second": 12.736,
25
+ "eval_steps_per_second": 0.588,
26
  "step": 8
27
  },
28
  {
29
+ "epoch": 2.84,
30
+ "eval_accuracy": 0.46153846153846156,
31
+ "eval_loss": 1.628129005432129,
32
+ "eval_runtime": 4.4017,
33
+ "eval_samples_per_second": 14.767,
34
+ "eval_steps_per_second": 0.682,
35
  "step": 12
36
  },
37
  {
38
+ "epoch": 3.84,
39
+ "eval_accuracy": 0.46153846153846156,
40
+ "eval_loss": 1.5271515846252441,
41
+ "eval_runtime": 4.7528,
42
+ "eval_samples_per_second": 13.676,
43
+ "eval_steps_per_second": 0.631,
44
  "step": 16
45
  },
46
  {
47
+ "epoch": 4.84,
48
  "learning_rate": 4.166666666666667e-05,
49
+ "loss": 1.9537,
50
  "step": 20
51
  },
52
  {
53
+ "epoch": 4.84,
54
+ "eval_accuracy": 0.5076923076923077,
55
+ "eval_loss": 1.3667770624160767,
56
+ "eval_runtime": 4.3866,
57
+ "eval_samples_per_second": 14.818,
58
+ "eval_steps_per_second": 0.684,
59
  "step": 20
60
  },
61
  {
62
+ "epoch": 5.84,
63
+ "eval_accuracy": 0.6,
64
+ "eval_loss": 1.0963521003723145,
65
+ "eval_runtime": 4.7064,
66
+ "eval_samples_per_second": 13.811,
67
+ "eval_steps_per_second": 0.637,
68
  "step": 24
69
  },
70
  {
71
+ "epoch": 6.84,
72
+ "eval_accuracy": 0.7846153846153846,
73
+ "eval_loss": 0.7691364884376526,
74
+ "eval_runtime": 4.2153,
75
+ "eval_samples_per_second": 15.42,
76
+ "eval_steps_per_second": 0.712,
77
  "step": 28
78
  },
79
  {
80
+ "epoch": 7.84,
81
+ "eval_accuracy": 0.8307692307692308,
82
+ "eval_loss": 0.6369762420654297,
83
+ "eval_runtime": 4.7258,
84
+ "eval_samples_per_second": 13.754,
85
+ "eval_steps_per_second": 0.635,
86
  "step": 32
87
  },
88
  {
89
+ "epoch": 8.84,
90
+ "eval_accuracy": 0.9076923076923077,
91
+ "eval_loss": 0.4328678548336029,
92
+ "eval_runtime": 4.7157,
93
+ "eval_samples_per_second": 13.784,
94
+ "eval_steps_per_second": 0.636,
95
  "step": 36
96
  },
97
  {
98
+ "epoch": 9.84,
99
  "learning_rate": 4.62962962962963e-05,
100
+ "loss": 1.0682,
101
  "step": 40
102
  },
103
  {
104
+ "epoch": 9.84,
105
+ "eval_accuracy": 0.9076923076923077,
106
+ "eval_loss": 0.3517535626888275,
107
+ "eval_runtime": 4.7127,
108
+ "eval_samples_per_second": 13.792,
109
+ "eval_steps_per_second": 0.637,
110
  "step": 40
111
  },
112
  {
113
+ "epoch": 10.84,
114
+ "eval_accuracy": 0.8923076923076924,
115
+ "eval_loss": 0.3229323625564575,
116
+ "eval_runtime": 4.2434,
117
+ "eval_samples_per_second": 15.318,
118
+ "eval_steps_per_second": 0.707,
119
  "step": 44
120
  },
121
  {
122
+ "epoch": 11.84,
123
+ "eval_accuracy": 0.9384615384615385,
124
+ "eval_loss": 0.2324085831642151,
125
+ "eval_runtime": 4.734,
126
+ "eval_samples_per_second": 13.731,
127
+ "eval_steps_per_second": 0.634,
128
  "step": 48
129
  },
130
  {
131
+ "epoch": 12.84,
132
+ "eval_accuracy": 0.9384615384615385,
133
+ "eval_loss": 0.23686641454696655,
134
+ "eval_runtime": 4.3354,
135
+ "eval_samples_per_second": 14.993,
136
+ "eval_steps_per_second": 0.692,
137
  "step": 52
138
  },
139
  {
140
+ "epoch": 13.84,
141
+ "eval_accuracy": 0.9384615384615385,
142
+ "eval_loss": 0.21190524101257324,
143
+ "eval_runtime": 4.793,
144
+ "eval_samples_per_second": 13.561,
145
+ "eval_steps_per_second": 0.626,
146
  "step": 56
147
  },
148
  {
149
+ "epoch": 14.84,
150
  "learning_rate": 4.166666666666667e-05,
151
+ "loss": 0.6335,
152
  "step": 60
153
  },
154
  {
155
+ "epoch": 14.84,
156
+ "eval_accuracy": 0.9384615384615385,
157
+ "eval_loss": 0.18046662211418152,
158
+ "eval_runtime": 4.2602,
159
+ "eval_samples_per_second": 15.257,
160
+ "eval_steps_per_second": 0.704,
161
  "step": 60
162
  },
163
  {
164
+ "epoch": 15.84,
165
+ "eval_accuracy": 0.9076923076923077,
166
+ "eval_loss": 0.21350829303264618,
167
+ "eval_runtime": 4.7427,
168
+ "eval_samples_per_second": 13.705,
169
+ "eval_steps_per_second": 0.633,
170
  "step": 64
171
  },
172
  {
173
+ "epoch": 16.84,
174
+ "eval_accuracy": 0.9230769230769231,
175
+ "eval_loss": 0.1889328807592392,
176
+ "eval_runtime": 4.2616,
177
+ "eval_samples_per_second": 15.253,
178
+ "eval_steps_per_second": 0.704,
179
  "step": 68
180
  },
181
  {
182
+ "epoch": 17.84,
183
+ "eval_accuracy": 0.9538461538461539,
184
+ "eval_loss": 0.16009008884429932,
185
+ "eval_runtime": 4.7629,
186
+ "eval_samples_per_second": 13.647,
187
+ "eval_steps_per_second": 0.63,
188
  "step": 72
189
  },
190
  {
191
+ "epoch": 18.84,
192
+ "eval_accuracy": 0.9692307692307692,
193
+ "eval_loss": 0.14124387502670288,
194
+ "eval_runtime": 4.3251,
195
+ "eval_samples_per_second": 15.029,
196
+ "eval_steps_per_second": 0.694,
197
  "step": 76
198
  },
199
  {
200
+ "epoch": 19.84,
201
  "learning_rate": 3.7037037037037037e-05,
202
+ "loss": 0.5133,
203
  "step": 80
204
  },
205
  {
206
+ "epoch": 19.84,
207
+ "eval_accuracy": 0.9538461538461539,
208
+ "eval_loss": 0.14968368411064148,
209
+ "eval_runtime": 4.7085,
210
+ "eval_samples_per_second": 13.805,
211
+ "eval_steps_per_second": 0.637,
212
  "step": 80
213
  },
214
  {
215
+ "epoch": 20.84,
216
+ "eval_accuracy": 0.9538461538461539,
217
+ "eval_loss": 0.15449711680412292,
218
+ "eval_runtime": 4.2803,
219
+ "eval_samples_per_second": 15.186,
220
+ "eval_steps_per_second": 0.701,
221
  "step": 84
222
  },
223
  {
224
+ "epoch": 21.84,
225
+ "eval_accuracy": 0.9538461538461539,
226
+ "eval_loss": 0.12975920736789703,
227
+ "eval_runtime": 6.2107,
228
+ "eval_samples_per_second": 10.466,
229
+ "eval_steps_per_second": 0.483,
230
  "step": 88
231
  },
232
  {
233
+ "epoch": 22.84,
234
+ "eval_accuracy": 0.9538461538461539,
235
+ "eval_loss": 0.14153267443180084,
236
+ "eval_runtime": 4.2538,
237
+ "eval_samples_per_second": 15.28,
238
+ "eval_steps_per_second": 0.705,
239
  "step": 92
240
  },
241
  {
242
+ "epoch": 23.84,
243
+ "eval_accuracy": 0.9230769230769231,
244
+ "eval_loss": 0.16845214366912842,
245
+ "eval_runtime": 4.6805,
246
+ "eval_samples_per_second": 13.888,
247
+ "eval_steps_per_second": 0.641,
248
  "step": 96
249
  },
250
  {
251
+ "epoch": 24.84,
252
  "learning_rate": 3.240740740740741e-05,
253
+ "loss": 0.4383,
254
  "step": 100
255
  },
256
  {
257
+ "epoch": 24.84,
258
+ "eval_accuracy": 0.9384615384615385,
259
+ "eval_loss": 0.13808518648147583,
260
+ "eval_runtime": 4.2829,
261
+ "eval_samples_per_second": 15.177,
262
+ "eval_steps_per_second": 0.7,
263
  "step": 100
264
  },
265
  {
266
+ "epoch": 25.84,
267
+ "eval_accuracy": 0.9846153846153847,
268
+ "eval_loss": 0.1295621246099472,
269
+ "eval_runtime": 4.717,
270
+ "eval_samples_per_second": 13.78,
271
+ "eval_steps_per_second": 0.636,
272
  "step": 104
273
  },
274
  {
275
+ "epoch": 26.84,
276
+ "eval_accuracy": 0.9538461538461539,
277
+ "eval_loss": 0.1106828972697258,
278
+ "eval_runtime": 4.4246,
279
+ "eval_samples_per_second": 14.691,
280
+ "eval_steps_per_second": 0.678,
281
  "step": 108
282
  },
283
  {
284
+ "epoch": 27.84,
285
+ "eval_accuracy": 0.9384615384615385,
286
+ "eval_loss": 0.12369797378778458,
287
+ "eval_runtime": 4.7653,
288
+ "eval_samples_per_second": 13.64,
289
+ "eval_steps_per_second": 0.63,
290
  "step": 112
291
  },
292
  {
293
+ "epoch": 28.84,
294
+ "eval_accuracy": 0.9538461538461539,
295
+ "eval_loss": 0.1366400122642517,
296
+ "eval_runtime": 4.3203,
297
+ "eval_samples_per_second": 15.045,
298
+ "eval_steps_per_second": 0.694,
299
  "step": 116
300
  },
301
  {
302
+ "epoch": 29.84,
303
  "learning_rate": 2.777777777777778e-05,
304
+ "loss": 0.4149,
305
  "step": 120
306
  },
307
  {
308
+ "epoch": 29.84,
309
+ "eval_accuracy": 0.9692307692307692,
310
+ "eval_loss": 0.13494354486465454,
311
+ "eval_runtime": 4.6686,
312
+ "eval_samples_per_second": 13.923,
313
+ "eval_steps_per_second": 0.643,
314
  "step": 120
315
  },
316
  {
317
+ "epoch": 30.84,
318
+ "eval_accuracy": 0.9846153846153847,
319
+ "eval_loss": 0.10458508133888245,
320
+ "eval_runtime": 4.2843,
321
+ "eval_samples_per_second": 15.172,
322
+ "eval_steps_per_second": 0.7,
323
  "step": 124
324
  },
325
  {
326
+ "epoch": 31.84,
327
+ "eval_accuracy": 0.9846153846153847,
328
+ "eval_loss": 0.0881725624203682,
329
+ "eval_runtime": 4.7053,
330
+ "eval_samples_per_second": 13.814,
331
+ "eval_steps_per_second": 0.638,
332
  "step": 128
333
  },
334
  {
335
+ "epoch": 32.84,
336
+ "eval_accuracy": 0.9846153846153847,
337
+ "eval_loss": 0.10219049453735352,
338
+ "eval_runtime": 4.3211,
339
+ "eval_samples_per_second": 15.042,
340
+ "eval_steps_per_second": 0.694,
341
  "step": 132
342
  },
343
  {
344
+ "epoch": 33.84,
345
+ "eval_accuracy": 0.9692307692307692,
346
+ "eval_loss": 0.12066025286912918,
347
+ "eval_runtime": 4.8023,
348
+ "eval_samples_per_second": 13.535,
349
+ "eval_steps_per_second": 0.625,
350
  "step": 136
351
  },
352
  {
353
+ "epoch": 34.84,
354
  "learning_rate": 2.314814814814815e-05,
355
+ "loss": 0.3657,
356
  "step": 140
357
  },
358
  {
359
+ "epoch": 34.84,
360
+ "eval_accuracy": 0.9538461538461539,
361
+ "eval_loss": 0.11678420007228851,
362
+ "eval_runtime": 4.4027,
363
+ "eval_samples_per_second": 14.764,
364
+ "eval_steps_per_second": 0.681,
365
  "step": 140
366
  },
367
  {
368
+ "epoch": 35.84,
369
+ "eval_accuracy": 0.9846153846153847,
370
+ "eval_loss": 0.09219826757907867,
371
+ "eval_runtime": 6.9593,
372
+ "eval_samples_per_second": 9.34,
373
+ "eval_steps_per_second": 0.431,
374
  "step": 144
375
  },
376
  {
377
+ "epoch": 36.84,
378
+ "eval_accuracy": 0.9846153846153847,
379
+ "eval_loss": 0.09314019978046417,
380
+ "eval_runtime": 4.3,
381
+ "eval_samples_per_second": 15.116,
382
+ "eval_steps_per_second": 0.698,
383
  "step": 148
384
  },
385
  {
386
+ "epoch": 37.84,
387
+ "eval_accuracy": 0.9692307692307692,
388
+ "eval_loss": 0.10055378824472427,
389
+ "eval_runtime": 4.7227,
390
+ "eval_samples_per_second": 13.763,
391
+ "eval_steps_per_second": 0.635,
392
  "step": 152
393
  },
394
  {
395
+ "epoch": 38.84,
396
+ "eval_accuracy": 0.9692307692307692,
397
+ "eval_loss": 0.09866410493850708,
398
+ "eval_runtime": 4.3725,
399
+ "eval_samples_per_second": 14.866,
400
+ "eval_steps_per_second": 0.686,
401
  "step": 156
402
  },
403
  {
404
+ "epoch": 39.84,
405
  "learning_rate": 1.8518518518518518e-05,
406
+ "loss": 0.3294,
407
  "step": 160
408
  },
409
  {
410
+ "epoch": 39.84,
411
+ "eval_accuracy": 0.9692307692307692,
412
+ "eval_loss": 0.11275947839021683,
413
+ "eval_runtime": 4.6903,
414
+ "eval_samples_per_second": 13.858,
415
+ "eval_steps_per_second": 0.64,
416
  "step": 160
417
  },
418
  {
419
+ "epoch": 40.84,
420
+ "eval_accuracy": 0.9538461538461539,
421
+ "eval_loss": 0.11520560085773468,
422
+ "eval_runtime": 4.3303,
423
+ "eval_samples_per_second": 15.01,
424
+ "eval_steps_per_second": 0.693,
425
  "step": 164
426
  },
427
  {
428
+ "epoch": 41.84,
429
+ "eval_accuracy": 0.9538461538461539,
430
+ "eval_loss": 0.09974207729101181,
431
+ "eval_runtime": 4.7259,
432
+ "eval_samples_per_second": 13.754,
433
+ "eval_steps_per_second": 0.635,
434
  "step": 168
435
  },
436
  {
437
+ "epoch": 42.84,
438
+ "eval_accuracy": 0.9692307692307692,
439
+ "eval_loss": 0.09684933722019196,
440
+ "eval_runtime": 4.3525,
441
+ "eval_samples_per_second": 14.934,
442
+ "eval_steps_per_second": 0.689,
443
  "step": 172
444
  },
445
  {
446
+ "epoch": 43.84,
447
+ "eval_accuracy": 0.9846153846153847,
448
+ "eval_loss": 0.08187751471996307,
449
+ "eval_runtime": 4.6538,
450
+ "eval_samples_per_second": 13.967,
451
+ "eval_steps_per_second": 0.645,
452
  "step": 176
453
  },
454
  {
455
+ "epoch": 44.84,
456
  "learning_rate": 1.388888888888889e-05,
457
+ "loss": 0.3198,
458
  "step": 180
459
  },
460
  {
461
+ "epoch": 44.84,
462
+ "eval_accuracy": 0.9846153846153847,
463
+ "eval_loss": 0.07292691618204117,
464
+ "eval_runtime": 4.3483,
465
+ "eval_samples_per_second": 14.948,
466
+ "eval_steps_per_second": 0.69,
467
  "step": 180
468
  },
469
  {
470
+ "epoch": 45.84,
471
+ "eval_accuracy": 0.9846153846153847,
472
+ "eval_loss": 0.07438476383686066,
473
+ "eval_runtime": 4.8323,
474
+ "eval_samples_per_second": 13.451,
475
+ "eval_steps_per_second": 0.621,
476
  "step": 184
477
  },
478
  {
479
+ "epoch": 46.84,
480
+ "eval_accuracy": 0.9692307692307692,
481
+ "eval_loss": 0.09510225057601929,
482
+ "eval_runtime": 4.3498,
483
+ "eval_samples_per_second": 14.943,
484
+ "eval_steps_per_second": 0.69,
485
  "step": 188
486
  },
487
  {
488
+ "epoch": 47.84,
489
+ "eval_accuracy": 0.9692307692307692,
490
+ "eval_loss": 0.09657642990350723,
491
+ "eval_runtime": 4.7293,
492
+ "eval_samples_per_second": 13.744,
493
+ "eval_steps_per_second": 0.634,
494
  "step": 192
495
  },
496
  {
497
+ "epoch": 48.84,
498
+ "eval_accuracy": 0.9846153846153847,
499
+ "eval_loss": 0.08329986035823822,
500
+ "eval_runtime": 4.2989,
501
+ "eval_samples_per_second": 15.12,
502
+ "eval_steps_per_second": 0.698,
503
  "step": 196
504
  },
505
  {
506
+ "epoch": 49.84,
507
  "learning_rate": 9.259259259259259e-06,
508
+ "loss": 0.2936,
509
  "step": 200
510
  },
511
  {
512
+ "epoch": 49.84,
513
+ "eval_accuracy": 0.9846153846153847,
514
+ "eval_loss": 0.06941184401512146,
515
+ "eval_runtime": 4.7066,
516
+ "eval_samples_per_second": 13.811,
517
+ "eval_steps_per_second": 0.637,
518
  "step": 200
519
  },
520
  {
521
+ "epoch": 50.84,
522
+ "eval_accuracy": 0.9846153846153847,
523
+ "eval_loss": 0.06908843666315079,
524
+ "eval_runtime": 4.3338,
525
+ "eval_samples_per_second": 14.999,
526
+ "eval_steps_per_second": 0.692,
527
  "step": 204
528
  },
529
  {
530
+ "epoch": 51.84,
531
+ "eval_accuracy": 0.9846153846153847,
532
+ "eval_loss": 0.0735919326543808,
533
+ "eval_runtime": 4.7155,
534
+ "eval_samples_per_second": 13.784,
535
+ "eval_steps_per_second": 0.636,
536
  "step": 208
537
  },
538
  {
539
+ "epoch": 52.84,
540
+ "eval_accuracy": 0.9692307692307692,
541
+ "eval_loss": 0.08048100024461746,
542
+ "eval_runtime": 4.337,
543
+ "eval_samples_per_second": 14.987,
544
+ "eval_steps_per_second": 0.692,
545
  "step": 212
546
  },
547
  {
548
+ "epoch": 53.84,
549
+ "eval_accuracy": 0.9846153846153847,
550
+ "eval_loss": 0.08007017523050308,
551
+ "eval_runtime": 4.765,
552
+ "eval_samples_per_second": 13.641,
553
+ "eval_steps_per_second": 0.63,
554
  "step": 216
555
  },
556
  {
557
+ "epoch": 54.84,
558
  "learning_rate": 4.6296296296296296e-06,
559
+ "loss": 0.3127,
560
  "step": 220
561
  },
562
  {
563
+ "epoch": 54.84,
564
+ "eval_accuracy": 0.9846153846153847,
565
+ "eval_loss": 0.0826273187994957,
566
+ "eval_runtime": 4.2934,
567
+ "eval_samples_per_second": 15.139,
568
+ "eval_steps_per_second": 0.699,
569
  "step": 220
570
  },
571
  {
572
+ "epoch": 55.84,
573
+ "eval_accuracy": 0.9692307692307692,
574
+ "eval_loss": 0.08565446734428406,
575
+ "eval_runtime": 4.7099,
576
+ "eval_samples_per_second": 13.801,
577
+ "eval_steps_per_second": 0.637,
578
  "step": 224
579
  },
580
  {
581
+ "epoch": 56.84,
582
+ "eval_accuracy": 0.9846153846153847,
583
+ "eval_loss": 0.0864274650812149,
584
+ "eval_runtime": 4.246,
585
+ "eval_samples_per_second": 15.309,
586
+ "eval_steps_per_second": 0.707,
587
  "step": 228
588
  },
589
  {
590
+ "epoch": 57.84,
591
+ "eval_accuracy": 0.9846153846153847,
592
+ "eval_loss": 0.08776658028364182,
593
+ "eval_runtime": 4.758,
594
+ "eval_samples_per_second": 13.661,
595
+ "eval_steps_per_second": 0.631,
596
  "step": 232
597
  },
598
  {
599
+ "epoch": 58.84,
600
+ "eval_accuracy": 0.9846153846153847,
601
+ "eval_loss": 0.08772400766611099,
602
+ "eval_runtime": 4.2874,
603
+ "eval_samples_per_second": 15.161,
604
+ "eval_steps_per_second": 0.7,
605
  "step": 236
606
  },
607
  {
608
+ "epoch": 59.84,
609
  "learning_rate": 0.0,
610
+ "loss": 0.285,
611
  "step": 240
612
  },
613
  {
614
+ "epoch": 59.84,
615
+ "eval_accuracy": 0.9846153846153847,
616
+ "eval_loss": 0.08735792338848114,
617
+ "eval_runtime": 4.7501,
618
+ "eval_samples_per_second": 13.684,
619
+ "eval_steps_per_second": 0.632,
620
  "step": 240
621
  },
622
  {
623
+ "epoch": 59.84,
624
  "step": 240,
625
+ "total_flos": 8.648855308501955e+17,
626
+ "train_loss": 0.577343902985255,
627
+ "train_runtime": 4213.9216,
628
+ "train_samples_per_second": 8.273,
629
  "train_steps_per_second": 0.057
630
  }
631
  ],
632
  "max_steps": 240,
633
  "num_train_epochs": 60,
634
+ "total_flos": 8.648855308501955e+17,
635
  "trial_name": null,
636
  "trial_params": null
637
  }