mansee commited on
Commit
091466e
1 Parent(s): 9936660

End of training

Browse files
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.9420152091254753
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.1766
36
- - Accuracy: 0.9420
37
 
38
  ## Model description
39
 
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.9448669201520913
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.1727
36
+ - Accuracy: 0.9449
37
 
38
  ## Model description
39
 
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9448669201520913,
4
+ "eval_loss": 0.17274008691310883,
5
+ "eval_runtime": 6.7385,
6
+ "eval_samples_per_second": 156.119,
7
+ "eval_steps_per_second": 4.897,
8
+ "total_flos": 2.3513765580619776e+18,
9
+ "train_loss": 0.21299834186966354,
10
+ "train_runtime": 1451.9061,
11
+ "train_samples_per_second": 65.156,
12
+ "train_steps_per_second": 0.51
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9448669201520913,
4
+ "eval_loss": 0.17274008691310883,
5
+ "eval_runtime": 6.7385,
6
+ "eval_samples_per_second": 156.119,
7
+ "eval_steps_per_second": 4.897
8
+ }
runs/Dec14_07-54-56_7d44aa449014/events.out.tfevents.1702541981.7d44aa449014.542.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4590efc5be5558286ddf532ba391158fd1cce12255d5abf5c00fd4378acbb515
3
+ size 411
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 2.3513765580619776e+18,
4
+ "train_loss": 0.21299834186966354,
5
+ "train_runtime": 1451.9061,
6
+ "train_samples_per_second": 65.156,
7
+ "train_steps_per_second": 0.51
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1006 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9448669201520913,
3
+ "best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/checkpoint-222",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 740,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07,
13
+ "learning_rate": 3.3783783783783788e-06,
14
+ "loss": 0.6924,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.14,
19
+ "learning_rate": 6.7567567567567575e-06,
20
+ "loss": 0.6582,
21
+ "step": 10
22
+ },
23
+ {
24
+ "epoch": 0.2,
25
+ "learning_rate": 1.0135135135135136e-05,
26
+ "loss": 0.5654,
27
+ "step": 15
28
+ },
29
+ {
30
+ "epoch": 0.27,
31
+ "learning_rate": 1.3513513513513515e-05,
32
+ "loss": 0.4386,
33
+ "step": 20
34
+ },
35
+ {
36
+ "epoch": 0.34,
37
+ "learning_rate": 1.6891891891891892e-05,
38
+ "loss": 0.3558,
39
+ "step": 25
40
+ },
41
+ {
42
+ "epoch": 0.41,
43
+ "learning_rate": 2.0270270270270273e-05,
44
+ "loss": 0.2918,
45
+ "step": 30
46
+ },
47
+ {
48
+ "epoch": 0.47,
49
+ "learning_rate": 2.364864864864865e-05,
50
+ "loss": 0.3143,
51
+ "step": 35
52
+ },
53
+ {
54
+ "epoch": 0.54,
55
+ "learning_rate": 2.702702702702703e-05,
56
+ "loss": 0.2339,
57
+ "step": 40
58
+ },
59
+ {
60
+ "epoch": 0.61,
61
+ "learning_rate": 3.0405405405405407e-05,
62
+ "loss": 0.2426,
63
+ "step": 45
64
+ },
65
+ {
66
+ "epoch": 0.68,
67
+ "learning_rate": 3.3783783783783784e-05,
68
+ "loss": 0.2386,
69
+ "step": 50
70
+ },
71
+ {
72
+ "epoch": 0.74,
73
+ "learning_rate": 3.7162162162162165e-05,
74
+ "loss": 0.212,
75
+ "step": 55
76
+ },
77
+ {
78
+ "epoch": 0.81,
79
+ "learning_rate": 4.0540540540540545e-05,
80
+ "loss": 0.2642,
81
+ "step": 60
82
+ },
83
+ {
84
+ "epoch": 0.88,
85
+ "learning_rate": 4.391891891891892e-05,
86
+ "loss": 0.2034,
87
+ "step": 65
88
+ },
89
+ {
90
+ "epoch": 0.95,
91
+ "learning_rate": 4.72972972972973e-05,
92
+ "loss": 0.2646,
93
+ "step": 70
94
+ },
95
+ {
96
+ "epoch": 1.0,
97
+ "eval_accuracy": 0.9391634980988594,
98
+ "eval_loss": 0.19738444685935974,
99
+ "eval_runtime": 6.6287,
100
+ "eval_samples_per_second": 158.704,
101
+ "eval_steps_per_second": 4.978,
102
+ "step": 74
103
+ },
104
+ {
105
+ "epoch": 1.01,
106
+ "learning_rate": 4.9924924924924924e-05,
107
+ "loss": 0.2718,
108
+ "step": 75
109
+ },
110
+ {
111
+ "epoch": 1.08,
112
+ "learning_rate": 4.954954954954955e-05,
113
+ "loss": 0.2249,
114
+ "step": 80
115
+ },
116
+ {
117
+ "epoch": 1.15,
118
+ "learning_rate": 4.9174174174174175e-05,
119
+ "loss": 0.2091,
120
+ "step": 85
121
+ },
122
+ {
123
+ "epoch": 1.22,
124
+ "learning_rate": 4.87987987987988e-05,
125
+ "loss": 0.2049,
126
+ "step": 90
127
+ },
128
+ {
129
+ "epoch": 1.28,
130
+ "learning_rate": 4.8423423423423426e-05,
131
+ "loss": 0.2384,
132
+ "step": 95
133
+ },
134
+ {
135
+ "epoch": 1.35,
136
+ "learning_rate": 4.804804804804805e-05,
137
+ "loss": 0.2021,
138
+ "step": 100
139
+ },
140
+ {
141
+ "epoch": 1.42,
142
+ "learning_rate": 4.767267267267268e-05,
143
+ "loss": 0.1791,
144
+ "step": 105
145
+ },
146
+ {
147
+ "epoch": 1.49,
148
+ "learning_rate": 4.72972972972973e-05,
149
+ "loss": 0.2404,
150
+ "step": 110
151
+ },
152
+ {
153
+ "epoch": 1.55,
154
+ "learning_rate": 4.692192192192192e-05,
155
+ "loss": 0.2547,
156
+ "step": 115
157
+ },
158
+ {
159
+ "epoch": 1.62,
160
+ "learning_rate": 4.654654654654655e-05,
161
+ "loss": 0.2253,
162
+ "step": 120
163
+ },
164
+ {
165
+ "epoch": 1.69,
166
+ "learning_rate": 4.617117117117117e-05,
167
+ "loss": 0.2445,
168
+ "step": 125
169
+ },
170
+ {
171
+ "epoch": 1.76,
172
+ "learning_rate": 4.57957957957958e-05,
173
+ "loss": 0.2609,
174
+ "step": 130
175
+ },
176
+ {
177
+ "epoch": 1.82,
178
+ "learning_rate": 4.5420420420420424e-05,
179
+ "loss": 0.2177,
180
+ "step": 135
181
+ },
182
+ {
183
+ "epoch": 1.89,
184
+ "learning_rate": 4.5045045045045046e-05,
185
+ "loss": 0.229,
186
+ "step": 140
187
+ },
188
+ {
189
+ "epoch": 1.96,
190
+ "learning_rate": 4.4669669669669675e-05,
191
+ "loss": 0.2287,
192
+ "step": 145
193
+ },
194
+ {
195
+ "epoch": 2.0,
196
+ "eval_accuracy": 0.935361216730038,
197
+ "eval_loss": 0.19786785542964935,
198
+ "eval_runtime": 7.1255,
199
+ "eval_samples_per_second": 147.638,
200
+ "eval_steps_per_second": 4.631,
201
+ "step": 148
202
+ },
203
+ {
204
+ "epoch": 2.03,
205
+ "learning_rate": 4.42942942942943e-05,
206
+ "loss": 0.2053,
207
+ "step": 150
208
+ },
209
+ {
210
+ "epoch": 2.09,
211
+ "learning_rate": 4.391891891891892e-05,
212
+ "loss": 0.1898,
213
+ "step": 155
214
+ },
215
+ {
216
+ "epoch": 2.16,
217
+ "learning_rate": 4.354354354354355e-05,
218
+ "loss": 0.1946,
219
+ "step": 160
220
+ },
221
+ {
222
+ "epoch": 2.23,
223
+ "learning_rate": 4.316816816816817e-05,
224
+ "loss": 0.2265,
225
+ "step": 165
226
+ },
227
+ {
228
+ "epoch": 2.3,
229
+ "learning_rate": 4.27927927927928e-05,
230
+ "loss": 0.2404,
231
+ "step": 170
232
+ },
233
+ {
234
+ "epoch": 2.36,
235
+ "learning_rate": 4.241741741741742e-05,
236
+ "loss": 0.2073,
237
+ "step": 175
238
+ },
239
+ {
240
+ "epoch": 2.43,
241
+ "learning_rate": 4.204204204204204e-05,
242
+ "loss": 0.2608,
243
+ "step": 180
244
+ },
245
+ {
246
+ "epoch": 2.5,
247
+ "learning_rate": 4.166666666666667e-05,
248
+ "loss": 0.2168,
249
+ "step": 185
250
+ },
251
+ {
252
+ "epoch": 2.57,
253
+ "learning_rate": 4.1291291291291294e-05,
254
+ "loss": 0.2091,
255
+ "step": 190
256
+ },
257
+ {
258
+ "epoch": 2.64,
259
+ "learning_rate": 4.091591591591592e-05,
260
+ "loss": 0.2468,
261
+ "step": 195
262
+ },
263
+ {
264
+ "epoch": 2.7,
265
+ "learning_rate": 4.0540540540540545e-05,
266
+ "loss": 0.2181,
267
+ "step": 200
268
+ },
269
+ {
270
+ "epoch": 2.77,
271
+ "learning_rate": 4.016516516516517e-05,
272
+ "loss": 0.1717,
273
+ "step": 205
274
+ },
275
+ {
276
+ "epoch": 2.84,
277
+ "learning_rate": 3.9789789789789796e-05,
278
+ "loss": 0.205,
279
+ "step": 210
280
+ },
281
+ {
282
+ "epoch": 2.91,
283
+ "learning_rate": 3.941441441441442e-05,
284
+ "loss": 0.1944,
285
+ "step": 215
286
+ },
287
+ {
288
+ "epoch": 2.97,
289
+ "learning_rate": 3.903903903903904e-05,
290
+ "loss": 0.198,
291
+ "step": 220
292
+ },
293
+ {
294
+ "epoch": 3.0,
295
+ "eval_accuracy": 0.9448669201520913,
296
+ "eval_loss": 0.17274008691310883,
297
+ "eval_runtime": 7.4744,
298
+ "eval_samples_per_second": 140.747,
299
+ "eval_steps_per_second": 4.415,
300
+ "step": 222
301
+ },
302
+ {
303
+ "epoch": 3.04,
304
+ "learning_rate": 3.866366366366366e-05,
305
+ "loss": 0.2083,
306
+ "step": 225
307
+ },
308
+ {
309
+ "epoch": 3.11,
310
+ "learning_rate": 3.8288288288288285e-05,
311
+ "loss": 0.1868,
312
+ "step": 230
313
+ },
314
+ {
315
+ "epoch": 3.18,
316
+ "learning_rate": 3.7912912912912914e-05,
317
+ "loss": 0.23,
318
+ "step": 235
319
+ },
320
+ {
321
+ "epoch": 3.24,
322
+ "learning_rate": 3.7537537537537536e-05,
323
+ "loss": 0.2083,
324
+ "step": 240
325
+ },
326
+ {
327
+ "epoch": 3.31,
328
+ "learning_rate": 3.7162162162162165e-05,
329
+ "loss": 0.1851,
330
+ "step": 245
331
+ },
332
+ {
333
+ "epoch": 3.38,
334
+ "learning_rate": 3.678678678678679e-05,
335
+ "loss": 0.2355,
336
+ "step": 250
337
+ },
338
+ {
339
+ "epoch": 3.45,
340
+ "learning_rate": 3.641141141141141e-05,
341
+ "loss": 0.2357,
342
+ "step": 255
343
+ },
344
+ {
345
+ "epoch": 3.51,
346
+ "learning_rate": 3.603603603603604e-05,
347
+ "loss": 0.1995,
348
+ "step": 260
349
+ },
350
+ {
351
+ "epoch": 3.58,
352
+ "learning_rate": 3.566066066066066e-05,
353
+ "loss": 0.205,
354
+ "step": 265
355
+ },
356
+ {
357
+ "epoch": 3.65,
358
+ "learning_rate": 3.528528528528528e-05,
359
+ "loss": 0.247,
360
+ "step": 270
361
+ },
362
+ {
363
+ "epoch": 3.72,
364
+ "learning_rate": 3.490990990990991e-05,
365
+ "loss": 0.1991,
366
+ "step": 275
367
+ },
368
+ {
369
+ "epoch": 3.78,
370
+ "learning_rate": 3.453453453453453e-05,
371
+ "loss": 0.1895,
372
+ "step": 280
373
+ },
374
+ {
375
+ "epoch": 3.85,
376
+ "learning_rate": 3.415915915915916e-05,
377
+ "loss": 0.2351,
378
+ "step": 285
379
+ },
380
+ {
381
+ "epoch": 3.92,
382
+ "learning_rate": 3.3783783783783784e-05,
383
+ "loss": 0.2073,
384
+ "step": 290
385
+ },
386
+ {
387
+ "epoch": 3.99,
388
+ "learning_rate": 3.3408408408408406e-05,
389
+ "loss": 0.1889,
390
+ "step": 295
391
+ },
392
+ {
393
+ "epoch": 4.0,
394
+ "eval_accuracy": 0.9429657794676806,
395
+ "eval_loss": 0.174727663397789,
396
+ "eval_runtime": 7.3243,
397
+ "eval_samples_per_second": 143.632,
398
+ "eval_steps_per_second": 4.506,
399
+ "step": 296
400
+ },
401
+ {
402
+ "epoch": 4.05,
403
+ "learning_rate": 3.3033033033033035e-05,
404
+ "loss": 0.1688,
405
+ "step": 300
406
+ },
407
+ {
408
+ "epoch": 4.12,
409
+ "learning_rate": 3.265765765765766e-05,
410
+ "loss": 0.1977,
411
+ "step": 305
412
+ },
413
+ {
414
+ "epoch": 4.19,
415
+ "learning_rate": 3.2282282282282286e-05,
416
+ "loss": 0.198,
417
+ "step": 310
418
+ },
419
+ {
420
+ "epoch": 4.26,
421
+ "learning_rate": 3.190690690690691e-05,
422
+ "loss": 0.2217,
423
+ "step": 315
424
+ },
425
+ {
426
+ "epoch": 4.32,
427
+ "learning_rate": 3.153153153153153e-05,
428
+ "loss": 0.1959,
429
+ "step": 320
430
+ },
431
+ {
432
+ "epoch": 4.39,
433
+ "learning_rate": 3.115615615615616e-05,
434
+ "loss": 0.2229,
435
+ "step": 325
436
+ },
437
+ {
438
+ "epoch": 4.46,
439
+ "learning_rate": 3.078078078078078e-05,
440
+ "loss": 0.1919,
441
+ "step": 330
442
+ },
443
+ {
444
+ "epoch": 4.53,
445
+ "learning_rate": 3.0405405405405407e-05,
446
+ "loss": 0.2318,
447
+ "step": 335
448
+ },
449
+ {
450
+ "epoch": 4.59,
451
+ "learning_rate": 3.0030030030030033e-05,
452
+ "loss": 0.214,
453
+ "step": 340
454
+ },
455
+ {
456
+ "epoch": 4.66,
457
+ "learning_rate": 2.9654654654654658e-05,
458
+ "loss": 0.1875,
459
+ "step": 345
460
+ },
461
+ {
462
+ "epoch": 4.73,
463
+ "learning_rate": 2.927927927927928e-05,
464
+ "loss": 0.2289,
465
+ "step": 350
466
+ },
467
+ {
468
+ "epoch": 4.8,
469
+ "learning_rate": 2.8903903903903906e-05,
470
+ "loss": 0.1887,
471
+ "step": 355
472
+ },
473
+ {
474
+ "epoch": 4.86,
475
+ "learning_rate": 2.852852852852853e-05,
476
+ "loss": 0.1918,
477
+ "step": 360
478
+ },
479
+ {
480
+ "epoch": 4.93,
481
+ "learning_rate": 2.8153153153153157e-05,
482
+ "loss": 0.1691,
483
+ "step": 365
484
+ },
485
+ {
486
+ "epoch": 5.0,
487
+ "learning_rate": 2.777777777777778e-05,
488
+ "loss": 0.223,
489
+ "step": 370
490
+ },
491
+ {
492
+ "epoch": 5.0,
493
+ "eval_accuracy": 0.9448669201520913,
494
+ "eval_loss": 0.17105169594287872,
495
+ "eval_runtime": 6.7978,
496
+ "eval_samples_per_second": 154.755,
497
+ "eval_steps_per_second": 4.854,
498
+ "step": 370
499
+ },
500
+ {
501
+ "epoch": 5.07,
502
+ "learning_rate": 2.7402402402402405e-05,
503
+ "loss": 0.1872,
504
+ "step": 375
505
+ },
506
+ {
507
+ "epoch": 5.14,
508
+ "learning_rate": 2.702702702702703e-05,
509
+ "loss": 0.2184,
510
+ "step": 380
511
+ },
512
+ {
513
+ "epoch": 5.2,
514
+ "learning_rate": 2.6651651651651656e-05,
515
+ "loss": 0.2066,
516
+ "step": 385
517
+ },
518
+ {
519
+ "epoch": 5.27,
520
+ "learning_rate": 2.6276276276276278e-05,
521
+ "loss": 0.181,
522
+ "step": 390
523
+ },
524
+ {
525
+ "epoch": 5.34,
526
+ "learning_rate": 2.5900900900900903e-05,
527
+ "loss": 0.1743,
528
+ "step": 395
529
+ },
530
+ {
531
+ "epoch": 5.41,
532
+ "learning_rate": 2.552552552552553e-05,
533
+ "loss": 0.188,
534
+ "step": 400
535
+ },
536
+ {
537
+ "epoch": 5.47,
538
+ "learning_rate": 2.5150150150150154e-05,
539
+ "loss": 0.2181,
540
+ "step": 405
541
+ },
542
+ {
543
+ "epoch": 5.54,
544
+ "learning_rate": 2.4774774774774777e-05,
545
+ "loss": 0.2173,
546
+ "step": 410
547
+ },
548
+ {
549
+ "epoch": 5.61,
550
+ "learning_rate": 2.43993993993994e-05,
551
+ "loss": 0.2207,
552
+ "step": 415
553
+ },
554
+ {
555
+ "epoch": 5.68,
556
+ "learning_rate": 2.4024024024024024e-05,
557
+ "loss": 0.2025,
558
+ "step": 420
559
+ },
560
+ {
561
+ "epoch": 5.74,
562
+ "learning_rate": 2.364864864864865e-05,
563
+ "loss": 0.1718,
564
+ "step": 425
565
+ },
566
+ {
567
+ "epoch": 5.81,
568
+ "learning_rate": 2.3273273273273275e-05,
569
+ "loss": 0.189,
570
+ "step": 430
571
+ },
572
+ {
573
+ "epoch": 5.88,
574
+ "learning_rate": 2.28978978978979e-05,
575
+ "loss": 0.1891,
576
+ "step": 435
577
+ },
578
+ {
579
+ "epoch": 5.95,
580
+ "learning_rate": 2.2522522522522523e-05,
581
+ "loss": 0.1771,
582
+ "step": 440
583
+ },
584
+ {
585
+ "epoch": 6.0,
586
+ "eval_accuracy": 0.938212927756654,
587
+ "eval_loss": 0.16966551542282104,
588
+ "eval_runtime": 6.7174,
589
+ "eval_samples_per_second": 156.609,
590
+ "eval_steps_per_second": 4.913,
591
+ "step": 444
592
+ },
593
+ {
594
+ "epoch": 6.01,
595
+ "learning_rate": 2.214714714714715e-05,
596
+ "loss": 0.212,
597
+ "step": 445
598
+ },
599
+ {
600
+ "epoch": 6.08,
601
+ "learning_rate": 2.1771771771771774e-05,
602
+ "loss": 0.193,
603
+ "step": 450
604
+ },
605
+ {
606
+ "epoch": 6.15,
607
+ "learning_rate": 2.13963963963964e-05,
608
+ "loss": 0.2115,
609
+ "step": 455
610
+ },
611
+ {
612
+ "epoch": 6.22,
613
+ "learning_rate": 2.102102102102102e-05,
614
+ "loss": 0.2092,
615
+ "step": 460
616
+ },
617
+ {
618
+ "epoch": 6.28,
619
+ "learning_rate": 2.0645645645645647e-05,
620
+ "loss": 0.1678,
621
+ "step": 465
622
+ },
623
+ {
624
+ "epoch": 6.35,
625
+ "learning_rate": 2.0270270270270273e-05,
626
+ "loss": 0.1898,
627
+ "step": 470
628
+ },
629
+ {
630
+ "epoch": 6.42,
631
+ "learning_rate": 1.9894894894894898e-05,
632
+ "loss": 0.1948,
633
+ "step": 475
634
+ },
635
+ {
636
+ "epoch": 6.49,
637
+ "learning_rate": 1.951951951951952e-05,
638
+ "loss": 0.2115,
639
+ "step": 480
640
+ },
641
+ {
642
+ "epoch": 6.55,
643
+ "learning_rate": 1.9144144144144142e-05,
644
+ "loss": 0.1735,
645
+ "step": 485
646
+ },
647
+ {
648
+ "epoch": 6.62,
649
+ "learning_rate": 1.8768768768768768e-05,
650
+ "loss": 0.1791,
651
+ "step": 490
652
+ },
653
+ {
654
+ "epoch": 6.69,
655
+ "learning_rate": 1.8393393393393393e-05,
656
+ "loss": 0.1728,
657
+ "step": 495
658
+ },
659
+ {
660
+ "epoch": 6.76,
661
+ "learning_rate": 1.801801801801802e-05,
662
+ "loss": 0.1969,
663
+ "step": 500
664
+ },
665
+ {
666
+ "epoch": 6.82,
667
+ "learning_rate": 1.764264264264264e-05,
668
+ "loss": 0.2058,
669
+ "step": 505
670
+ },
671
+ {
672
+ "epoch": 6.89,
673
+ "learning_rate": 1.7267267267267267e-05,
674
+ "loss": 0.1762,
675
+ "step": 510
676
+ },
677
+ {
678
+ "epoch": 6.96,
679
+ "learning_rate": 1.6891891891891892e-05,
680
+ "loss": 0.1864,
681
+ "step": 515
682
+ },
683
+ {
684
+ "epoch": 7.0,
685
+ "eval_accuracy": 0.9391634980988594,
686
+ "eval_loss": 0.16719643771648407,
687
+ "eval_runtime": 7.3392,
688
+ "eval_samples_per_second": 143.339,
689
+ "eval_steps_per_second": 4.496,
690
+ "step": 518
691
+ },
692
+ {
693
+ "epoch": 7.03,
694
+ "learning_rate": 1.6516516516516518e-05,
695
+ "loss": 0.2046,
696
+ "step": 520
697
+ },
698
+ {
699
+ "epoch": 7.09,
700
+ "learning_rate": 1.6141141141141143e-05,
701
+ "loss": 0.1794,
702
+ "step": 525
703
+ },
704
+ {
705
+ "epoch": 7.16,
706
+ "learning_rate": 1.5765765765765765e-05,
707
+ "loss": 0.1919,
708
+ "step": 530
709
+ },
710
+ {
711
+ "epoch": 7.23,
712
+ "learning_rate": 1.539039039039039e-05,
713
+ "loss": 0.187,
714
+ "step": 535
715
+ },
716
+ {
717
+ "epoch": 7.3,
718
+ "learning_rate": 1.5015015015015016e-05,
719
+ "loss": 0.1637,
720
+ "step": 540
721
+ },
722
+ {
723
+ "epoch": 7.36,
724
+ "learning_rate": 1.463963963963964e-05,
725
+ "loss": 0.1572,
726
+ "step": 545
727
+ },
728
+ {
729
+ "epoch": 7.43,
730
+ "learning_rate": 1.4264264264264266e-05,
731
+ "loss": 0.1708,
732
+ "step": 550
733
+ },
734
+ {
735
+ "epoch": 7.5,
736
+ "learning_rate": 1.388888888888889e-05,
737
+ "loss": 0.1963,
738
+ "step": 555
739
+ },
740
+ {
741
+ "epoch": 7.57,
742
+ "learning_rate": 1.3513513513513515e-05,
743
+ "loss": 0.1977,
744
+ "step": 560
745
+ },
746
+ {
747
+ "epoch": 7.64,
748
+ "learning_rate": 1.3138138138138139e-05,
749
+ "loss": 0.1751,
750
+ "step": 565
751
+ },
752
+ {
753
+ "epoch": 7.7,
754
+ "learning_rate": 1.2762762762762764e-05,
755
+ "loss": 0.1707,
756
+ "step": 570
757
+ },
758
+ {
759
+ "epoch": 7.77,
760
+ "learning_rate": 1.2387387387387388e-05,
761
+ "loss": 0.1561,
762
+ "step": 575
763
+ },
764
+ {
765
+ "epoch": 7.84,
766
+ "learning_rate": 1.2012012012012012e-05,
767
+ "loss": 0.1918,
768
+ "step": 580
769
+ },
770
+ {
771
+ "epoch": 7.91,
772
+ "learning_rate": 1.1636636636636638e-05,
773
+ "loss": 0.2219,
774
+ "step": 585
775
+ },
776
+ {
777
+ "epoch": 7.97,
778
+ "learning_rate": 1.1261261261261261e-05,
779
+ "loss": 0.1716,
780
+ "step": 590
781
+ },
782
+ {
783
+ "epoch": 8.0,
784
+ "eval_accuracy": 0.9429657794676806,
785
+ "eval_loss": 0.18012098968029022,
786
+ "eval_runtime": 7.5217,
787
+ "eval_samples_per_second": 139.861,
788
+ "eval_steps_per_second": 4.387,
789
+ "step": 592
790
+ },
791
+ {
792
+ "epoch": 8.04,
793
+ "learning_rate": 1.0885885885885887e-05,
794
+ "loss": 0.1603,
795
+ "step": 595
796
+ },
797
+ {
798
+ "epoch": 8.11,
799
+ "learning_rate": 1.051051051051051e-05,
800
+ "loss": 0.1871,
801
+ "step": 600
802
+ },
803
+ {
804
+ "epoch": 8.18,
805
+ "learning_rate": 1.0135135135135136e-05,
806
+ "loss": 0.1972,
807
+ "step": 605
808
+ },
809
+ {
810
+ "epoch": 8.24,
811
+ "learning_rate": 9.75975975975976e-06,
812
+ "loss": 0.223,
813
+ "step": 610
814
+ },
815
+ {
816
+ "epoch": 8.31,
817
+ "learning_rate": 9.384384384384384e-06,
818
+ "loss": 0.1672,
819
+ "step": 615
820
+ },
821
+ {
822
+ "epoch": 8.38,
823
+ "learning_rate": 9.00900900900901e-06,
824
+ "loss": 0.1529,
825
+ "step": 620
826
+ },
827
+ {
828
+ "epoch": 8.45,
829
+ "learning_rate": 8.633633633633633e-06,
830
+ "loss": 0.1701,
831
+ "step": 625
832
+ },
833
+ {
834
+ "epoch": 8.51,
835
+ "learning_rate": 8.258258258258259e-06,
836
+ "loss": 0.1925,
837
+ "step": 630
838
+ },
839
+ {
840
+ "epoch": 8.58,
841
+ "learning_rate": 7.882882882882883e-06,
842
+ "loss": 0.1448,
843
+ "step": 635
844
+ },
845
+ {
846
+ "epoch": 8.65,
847
+ "learning_rate": 7.507507507507508e-06,
848
+ "loss": 0.181,
849
+ "step": 640
850
+ },
851
+ {
852
+ "epoch": 8.72,
853
+ "learning_rate": 7.132132132132133e-06,
854
+ "loss": 0.1984,
855
+ "step": 645
856
+ },
857
+ {
858
+ "epoch": 8.78,
859
+ "learning_rate": 6.7567567567567575e-06,
860
+ "loss": 0.203,
861
+ "step": 650
862
+ },
863
+ {
864
+ "epoch": 8.85,
865
+ "learning_rate": 6.381381381381382e-06,
866
+ "loss": 0.1554,
867
+ "step": 655
868
+ },
869
+ {
870
+ "epoch": 8.92,
871
+ "learning_rate": 6.006006006006006e-06,
872
+ "loss": 0.1743,
873
+ "step": 660
874
+ },
875
+ {
876
+ "epoch": 8.99,
877
+ "learning_rate": 5.630630630630631e-06,
878
+ "loss": 0.192,
879
+ "step": 665
880
+ },
881
+ {
882
+ "epoch": 9.0,
883
+ "eval_accuracy": 0.94106463878327,
884
+ "eval_loss": 0.17536889016628265,
885
+ "eval_runtime": 6.7103,
886
+ "eval_samples_per_second": 156.774,
887
+ "eval_steps_per_second": 4.918,
888
+ "step": 666
889
+ },
890
+ {
891
+ "epoch": 9.05,
892
+ "learning_rate": 5.255255255255255e-06,
893
+ "loss": 0.1613,
894
+ "step": 670
895
+ },
896
+ {
897
+ "epoch": 9.12,
898
+ "learning_rate": 4.87987987987988e-06,
899
+ "loss": 0.1746,
900
+ "step": 675
901
+ },
902
+ {
903
+ "epoch": 9.19,
904
+ "learning_rate": 4.504504504504505e-06,
905
+ "loss": 0.1584,
906
+ "step": 680
907
+ },
908
+ {
909
+ "epoch": 9.26,
910
+ "learning_rate": 4.129129129129129e-06,
911
+ "loss": 0.1849,
912
+ "step": 685
913
+ },
914
+ {
915
+ "epoch": 9.32,
916
+ "learning_rate": 3.753753753753754e-06,
917
+ "loss": 0.1871,
918
+ "step": 690
919
+ },
920
+ {
921
+ "epoch": 9.39,
922
+ "learning_rate": 3.3783783783783788e-06,
923
+ "loss": 0.1302,
924
+ "step": 695
925
+ },
926
+ {
927
+ "epoch": 9.46,
928
+ "learning_rate": 3.003003003003003e-06,
929
+ "loss": 0.1586,
930
+ "step": 700
931
+ },
932
+ {
933
+ "epoch": 9.53,
934
+ "learning_rate": 2.6276276276276277e-06,
935
+ "loss": 0.2135,
936
+ "step": 705
937
+ },
938
+ {
939
+ "epoch": 9.59,
940
+ "learning_rate": 2.2522522522522524e-06,
941
+ "loss": 0.1994,
942
+ "step": 710
943
+ },
944
+ {
945
+ "epoch": 9.66,
946
+ "learning_rate": 1.876876876876877e-06,
947
+ "loss": 0.1693,
948
+ "step": 715
949
+ },
950
+ {
951
+ "epoch": 9.73,
952
+ "learning_rate": 1.5015015015015015e-06,
953
+ "loss": 0.2072,
954
+ "step": 720
955
+ },
956
+ {
957
+ "epoch": 9.8,
958
+ "learning_rate": 1.1261261261261262e-06,
959
+ "loss": 0.1806,
960
+ "step": 725
961
+ },
962
+ {
963
+ "epoch": 9.86,
964
+ "learning_rate": 7.507507507507508e-07,
965
+ "loss": 0.1692,
966
+ "step": 730
967
+ },
968
+ {
969
+ "epoch": 9.93,
970
+ "learning_rate": 3.753753753753754e-07,
971
+ "loss": 0.1998,
972
+ "step": 735
973
+ },
974
+ {
975
+ "epoch": 10.0,
976
+ "learning_rate": 0.0,
977
+ "loss": 0.1886,
978
+ "step": 740
979
+ },
980
+ {
981
+ "epoch": 10.0,
982
+ "eval_accuracy": 0.9420152091254753,
983
+ "eval_loss": 0.17660604417324066,
984
+ "eval_runtime": 6.7015,
985
+ "eval_samples_per_second": 156.979,
986
+ "eval_steps_per_second": 4.924,
987
+ "step": 740
988
+ },
989
+ {
990
+ "epoch": 10.0,
991
+ "step": 740,
992
+ "total_flos": 2.3513765580619776e+18,
993
+ "train_loss": 0.21299834186966354,
994
+ "train_runtime": 1451.9061,
995
+ "train_samples_per_second": 65.156,
996
+ "train_steps_per_second": 0.51
997
+ }
998
+ ],
999
+ "logging_steps": 5,
1000
+ "max_steps": 740,
1001
+ "num_train_epochs": 10,
1002
+ "save_steps": 500,
1003
+ "total_flos": 2.3513765580619776e+18,
1004
+ "trial_name": null,
1005
+ "trial_params": null
1006
+ }