MatrixYao commited on
Commit
fe8619b
·
verified ·
1 Parent(s): dffec9c

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 +469 -0
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9829629629629629,
4
+ "eval_loss": 0.04924270510673523,
5
+ "eval_runtime": 4.4148,
6
+ "eval_samples_per_second": 611.578,
7
+ "eval_steps_per_second": 19.253,
8
+ "total_flos": 1.8124066505760768e+18,
9
+ "train_loss": 0.2828052028229362,
10
+ "train_runtime": 282.6955,
11
+ "train_samples_per_second": 257.875,
12
+ "train_steps_per_second": 2.016
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9829629629629629,
4
+ "eval_loss": 0.04924270510673523,
5
+ "eval_runtime": 4.4148,
6
+ "eval_samples_per_second": 611.578,
7
+ "eval_steps_per_second": 19.253
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 1.8124066505760768e+18,
4
+ "train_loss": 0.2828052028229362,
5
+ "train_runtime": 282.6955,
6
+ "train_samples_per_second": 257.875,
7
+ "train_steps_per_second": 2.016
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,469 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 570,
3
+ "best_metric": 0.9829629629629629,
4
+ "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-570",
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 570,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.05263157894736842,
14
+ "grad_norm": 4.065399646759033,
15
+ "learning_rate": 7.894736842105263e-06,
16
+ "loss": 2.2613,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.10526315789473684,
21
+ "grad_norm": 5.319888114929199,
22
+ "learning_rate": 1.6666666666666667e-05,
23
+ "loss": 2.01,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.15789473684210525,
28
+ "grad_norm": 6.084249496459961,
29
+ "learning_rate": 2.5438596491228074e-05,
30
+ "loss": 1.5179,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.21052631578947367,
35
+ "grad_norm": 7.026301860809326,
36
+ "learning_rate": 3.421052631578947e-05,
37
+ "loss": 0.8649,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.2631578947368421,
42
+ "grad_norm": 11.555365562438965,
43
+ "learning_rate": 4.298245614035088e-05,
44
+ "loss": 0.4874,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.3157894736842105,
49
+ "grad_norm": 17.860612869262695,
50
+ "learning_rate": 4.980506822612086e-05,
51
+ "loss": 0.3923,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.3684210526315789,
56
+ "grad_norm": 26.220460891723633,
57
+ "learning_rate": 4.883040935672515e-05,
58
+ "loss": 0.3607,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.42105263157894735,
63
+ "grad_norm": 11.298727035522461,
64
+ "learning_rate": 4.7855750487329435e-05,
65
+ "loss": 0.2858,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.47368421052631576,
70
+ "grad_norm": 13.609187126159668,
71
+ "learning_rate": 4.6881091617933726e-05,
72
+ "loss": 0.3207,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.5263157894736842,
77
+ "grad_norm": 10.572683334350586,
78
+ "learning_rate": 4.590643274853802e-05,
79
+ "loss": 0.2541,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.5789473684210527,
84
+ "grad_norm": 10.712231636047363,
85
+ "learning_rate": 4.49317738791423e-05,
86
+ "loss": 0.2205,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.631578947368421,
91
+ "grad_norm": 8.657620429992676,
92
+ "learning_rate": 4.395711500974659e-05,
93
+ "loss": 0.1989,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.6842105263157895,
98
+ "grad_norm": 8.740941047668457,
99
+ "learning_rate": 4.298245614035088e-05,
100
+ "loss": 0.2312,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.7368421052631579,
105
+ "grad_norm": 9.41290283203125,
106
+ "learning_rate": 4.200779727095517e-05,
107
+ "loss": 0.2229,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.7894736842105263,
112
+ "grad_norm": 15.130921363830566,
113
+ "learning_rate": 4.103313840155946e-05,
114
+ "loss": 0.1919,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.8421052631578947,
119
+ "grad_norm": 10.599325180053711,
120
+ "learning_rate": 4.005847953216375e-05,
121
+ "loss": 0.2044,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.8947368421052632,
126
+ "grad_norm": 6.5861968994140625,
127
+ "learning_rate": 3.908382066276803e-05,
128
+ "loss": 0.2228,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.9473684210526315,
133
+ "grad_norm": 11.306436538696289,
134
+ "learning_rate": 3.8109161793372323e-05,
135
+ "loss": 0.1981,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 1.0,
140
+ "grad_norm": 7.8563456535339355,
141
+ "learning_rate": 3.713450292397661e-05,
142
+ "loss": 0.2157,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 1.0,
147
+ "eval_accuracy": 0.9703703703703703,
148
+ "eval_loss": 0.0875425785779953,
149
+ "eval_runtime": 4.4626,
150
+ "eval_samples_per_second": 605.031,
151
+ "eval_steps_per_second": 19.047,
152
+ "step": 190
153
+ },
154
+ {
155
+ "epoch": 1.0526315789473684,
156
+ "grad_norm": 13.41629409790039,
157
+ "learning_rate": 3.61598440545809e-05,
158
+ "loss": 0.201,
159
+ "step": 200
160
+ },
161
+ {
162
+ "epoch": 1.1052631578947367,
163
+ "grad_norm": 4.426482200622559,
164
+ "learning_rate": 3.518518518518519e-05,
165
+ "loss": 0.1957,
166
+ "step": 210
167
+ },
168
+ {
169
+ "epoch": 1.1578947368421053,
170
+ "grad_norm": 9.049041748046875,
171
+ "learning_rate": 3.421052631578947e-05,
172
+ "loss": 0.2389,
173
+ "step": 220
174
+ },
175
+ {
176
+ "epoch": 1.2105263157894737,
177
+ "grad_norm": 5.128701686859131,
178
+ "learning_rate": 3.3235867446393764e-05,
179
+ "loss": 0.1702,
180
+ "step": 230
181
+ },
182
+ {
183
+ "epoch": 1.263157894736842,
184
+ "grad_norm": 7.488633632659912,
185
+ "learning_rate": 3.2261208576998055e-05,
186
+ "loss": 0.2236,
187
+ "step": 240
188
+ },
189
+ {
190
+ "epoch": 1.3157894736842106,
191
+ "grad_norm": 9.638894081115723,
192
+ "learning_rate": 3.128654970760234e-05,
193
+ "loss": 0.2388,
194
+ "step": 250
195
+ },
196
+ {
197
+ "epoch": 1.368421052631579,
198
+ "grad_norm": 8.79790210723877,
199
+ "learning_rate": 3.031189083820663e-05,
200
+ "loss": 0.1877,
201
+ "step": 260
202
+ },
203
+ {
204
+ "epoch": 1.4210526315789473,
205
+ "grad_norm": 7.657425403594971,
206
+ "learning_rate": 2.9337231968810917e-05,
207
+ "loss": 0.1457,
208
+ "step": 270
209
+ },
210
+ {
211
+ "epoch": 1.4736842105263157,
212
+ "grad_norm": 5.015496253967285,
213
+ "learning_rate": 2.8362573099415208e-05,
214
+ "loss": 0.147,
215
+ "step": 280
216
+ },
217
+ {
218
+ "epoch": 1.526315789473684,
219
+ "grad_norm": 3.1375253200531006,
220
+ "learning_rate": 2.7387914230019495e-05,
221
+ "loss": 0.1461,
222
+ "step": 290
223
+ },
224
+ {
225
+ "epoch": 1.5789473684210527,
226
+ "grad_norm": 9.31838321685791,
227
+ "learning_rate": 2.6413255360623783e-05,
228
+ "loss": 0.1599,
229
+ "step": 300
230
+ },
231
+ {
232
+ "epoch": 1.631578947368421,
233
+ "grad_norm": 6.391341686248779,
234
+ "learning_rate": 2.5438596491228074e-05,
235
+ "loss": 0.1492,
236
+ "step": 310
237
+ },
238
+ {
239
+ "epoch": 1.6842105263157894,
240
+ "grad_norm": 5.891468524932861,
241
+ "learning_rate": 2.4463937621832358e-05,
242
+ "loss": 0.1662,
243
+ "step": 320
244
+ },
245
+ {
246
+ "epoch": 1.736842105263158,
247
+ "grad_norm": 6.506641864776611,
248
+ "learning_rate": 2.348927875243665e-05,
249
+ "loss": 0.1448,
250
+ "step": 330
251
+ },
252
+ {
253
+ "epoch": 1.7894736842105263,
254
+ "grad_norm": 5.7794108390808105,
255
+ "learning_rate": 2.2514619883040936e-05,
256
+ "loss": 0.112,
257
+ "step": 340
258
+ },
259
+ {
260
+ "epoch": 1.8421052631578947,
261
+ "grad_norm": 8.3933744430542,
262
+ "learning_rate": 2.1539961013645223e-05,
263
+ "loss": 0.1292,
264
+ "step": 350
265
+ },
266
+ {
267
+ "epoch": 1.8947368421052633,
268
+ "grad_norm": 6.196931838989258,
269
+ "learning_rate": 2.056530214424951e-05,
270
+ "loss": 0.1418,
271
+ "step": 360
272
+ },
273
+ {
274
+ "epoch": 1.9473684210526314,
275
+ "grad_norm": 7.739428997039795,
276
+ "learning_rate": 1.9590643274853802e-05,
277
+ "loss": 0.1129,
278
+ "step": 370
279
+ },
280
+ {
281
+ "epoch": 2.0,
282
+ "grad_norm": 7.849891185760498,
283
+ "learning_rate": 1.861598440545809e-05,
284
+ "loss": 0.1528,
285
+ "step": 380
286
+ },
287
+ {
288
+ "epoch": 2.0,
289
+ "eval_accuracy": 0.9748148148148148,
290
+ "eval_loss": 0.0713169053196907,
291
+ "eval_runtime": 4.3996,
292
+ "eval_samples_per_second": 613.688,
293
+ "eval_steps_per_second": 19.32,
294
+ "step": 380
295
+ },
296
+ {
297
+ "epoch": 2.0526315789473686,
298
+ "grad_norm": 5.565265655517578,
299
+ "learning_rate": 1.7641325536062377e-05,
300
+ "loss": 0.1234,
301
+ "step": 390
302
+ },
303
+ {
304
+ "epoch": 2.1052631578947367,
305
+ "grad_norm": 6.100377082824707,
306
+ "learning_rate": 1.6666666666666667e-05,
307
+ "loss": 0.1248,
308
+ "step": 400
309
+ },
310
+ {
311
+ "epoch": 2.1578947368421053,
312
+ "grad_norm": 2.68096661567688,
313
+ "learning_rate": 1.5692007797270955e-05,
314
+ "loss": 0.0989,
315
+ "step": 410
316
+ },
317
+ {
318
+ "epoch": 2.2105263157894735,
319
+ "grad_norm": 2.9436898231506348,
320
+ "learning_rate": 1.4717348927875244e-05,
321
+ "loss": 0.1111,
322
+ "step": 420
323
+ },
324
+ {
325
+ "epoch": 2.263157894736842,
326
+ "grad_norm": 6.701323986053467,
327
+ "learning_rate": 1.3742690058479531e-05,
328
+ "loss": 0.1659,
329
+ "step": 430
330
+ },
331
+ {
332
+ "epoch": 2.3157894736842106,
333
+ "grad_norm": 6.343139171600342,
334
+ "learning_rate": 1.276803118908382e-05,
335
+ "loss": 0.103,
336
+ "step": 440
337
+ },
338
+ {
339
+ "epoch": 2.3684210526315788,
340
+ "grad_norm": 5.917227268218994,
341
+ "learning_rate": 1.179337231968811e-05,
342
+ "loss": 0.1059,
343
+ "step": 450
344
+ },
345
+ {
346
+ "epoch": 2.4210526315789473,
347
+ "grad_norm": 6.181001663208008,
348
+ "learning_rate": 1.0818713450292397e-05,
349
+ "loss": 0.116,
350
+ "step": 460
351
+ },
352
+ {
353
+ "epoch": 2.473684210526316,
354
+ "grad_norm": 7.878546714782715,
355
+ "learning_rate": 9.844054580896686e-06,
356
+ "loss": 0.1418,
357
+ "step": 470
358
+ },
359
+ {
360
+ "epoch": 2.526315789473684,
361
+ "grad_norm": 8.960369110107422,
362
+ "learning_rate": 8.869395711500974e-06,
363
+ "loss": 0.1609,
364
+ "step": 480
365
+ },
366
+ {
367
+ "epoch": 2.5789473684210527,
368
+ "grad_norm": 10.864946365356445,
369
+ "learning_rate": 7.894736842105263e-06,
370
+ "loss": 0.1314,
371
+ "step": 490
372
+ },
373
+ {
374
+ "epoch": 2.6315789473684212,
375
+ "grad_norm": 6.69527530670166,
376
+ "learning_rate": 6.920077972709551e-06,
377
+ "loss": 0.1472,
378
+ "step": 500
379
+ },
380
+ {
381
+ "epoch": 2.6842105263157894,
382
+ "grad_norm": 8.905725479125977,
383
+ "learning_rate": 5.94541910331384e-06,
384
+ "loss": 0.1382,
385
+ "step": 510
386
+ },
387
+ {
388
+ "epoch": 2.736842105263158,
389
+ "grad_norm": 4.663083076477051,
390
+ "learning_rate": 4.970760233918129e-06,
391
+ "loss": 0.1509,
392
+ "step": 520
393
+ },
394
+ {
395
+ "epoch": 2.7894736842105265,
396
+ "grad_norm": 6.759707927703857,
397
+ "learning_rate": 3.996101364522417e-06,
398
+ "loss": 0.115,
399
+ "step": 530
400
+ },
401
+ {
402
+ "epoch": 2.8421052631578947,
403
+ "grad_norm": 4.101389408111572,
404
+ "learning_rate": 3.0214424951267056e-06,
405
+ "loss": 0.0855,
406
+ "step": 540
407
+ },
408
+ {
409
+ "epoch": 2.8947368421052633,
410
+ "grad_norm": 5.965505123138428,
411
+ "learning_rate": 2.0467836257309943e-06,
412
+ "loss": 0.1056,
413
+ "step": 550
414
+ },
415
+ {
416
+ "epoch": 2.9473684210526314,
417
+ "grad_norm": 3.79522442817688,
418
+ "learning_rate": 1.0721247563352826e-06,
419
+ "loss": 0.0831,
420
+ "step": 560
421
+ },
422
+ {
423
+ "epoch": 3.0,
424
+ "grad_norm": 2.4792110919952393,
425
+ "learning_rate": 9.746588693957116e-08,
426
+ "loss": 0.0861,
427
+ "step": 570
428
+ },
429
+ {
430
+ "epoch": 3.0,
431
+ "eval_accuracy": 0.9829629629629629,
432
+ "eval_loss": 0.04924270510673523,
433
+ "eval_runtime": 4.4303,
434
+ "eval_samples_per_second": 609.434,
435
+ "eval_steps_per_second": 19.186,
436
+ "step": 570
437
+ },
438
+ {
439
+ "epoch": 3.0,
440
+ "step": 570,
441
+ "total_flos": 1.8124066505760768e+18,
442
+ "train_loss": 0.2828052028229362,
443
+ "train_runtime": 282.6955,
444
+ "train_samples_per_second": 257.875,
445
+ "train_steps_per_second": 2.016
446
+ }
447
+ ],
448
+ "logging_steps": 10,
449
+ "max_steps": 570,
450
+ "num_input_tokens_seen": 0,
451
+ "num_train_epochs": 3,
452
+ "save_steps": 500,
453
+ "stateful_callbacks": {
454
+ "TrainerControl": {
455
+ "args": {
456
+ "should_epoch_stop": false,
457
+ "should_evaluate": false,
458
+ "should_log": false,
459
+ "should_save": true,
460
+ "should_training_stop": true
461
+ },
462
+ "attributes": {}
463
+ }
464
+ },
465
+ "total_flos": 1.8124066505760768e+18,
466
+ "train_batch_size": 32,
467
+ "trial_name": null,
468
+ "trial_params": null
469
+ }