End of training
Browse files- README.md +3 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +614 -218
README.md
CHANGED
@@ -22,7 +22,7 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
-
value: 0.
|
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.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.7321109123434705
|
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: 0.4505
|
36 |
+
- Accuracy: 0.7321
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 32.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 4.
|
8 |
-
"total_flos":
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second": 61.
|
12 |
-
"train_steps_per_second": 0.
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 4.99,
|
3 |
+
"eval_accuracy": 0.7321109123434705,
|
4 |
+
"eval_loss": 0.45054975152015686,
|
5 |
+
"eval_runtime": 32.1968,
|
6 |
+
"eval_samples_per_second": 138.896,
|
7 |
+
"eval_steps_per_second": 4.348,
|
8 |
+
"total_flos": 4.994244969949348e+18,
|
9 |
+
"train_loss": 0.6026059660941931,
|
10 |
+
"train_runtime": 3290.1425,
|
11 |
+
"train_samples_per_second": 61.16,
|
12 |
+
"train_steps_per_second": 0.477
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 32.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 4.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 4.99,
|
3 |
+
"eval_accuracy": 0.7321109123434705,
|
4 |
+
"eval_loss": 0.45054975152015686,
|
5 |
+
"eval_runtime": 32.1968,
|
6 |
+
"eval_samples_per_second": 138.896,
|
7 |
+
"eval_steps_per_second": 4.348
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second": 61.
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 4.99,
|
3 |
+
"total_flos": 4.994244969949348e+18,
|
4 |
+
"train_loss": 0.6026059660941931,
|
5 |
+
"train_runtime": 3290.1425,
|
6 |
+
"train_samples_per_second": 61.16,
|
7 |
+
"train_steps_per_second": 0.477
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,619 +1,1015 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-img_orientation/checkpoint-
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
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":
|
14 |
-
"loss": 1.
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.06,
|
19 |
-
"learning_rate":
|
20 |
-
"loss": 1.
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.1,
|
25 |
-
"learning_rate":
|
26 |
-
"loss": 1.
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.13,
|
31 |
-
"learning_rate":
|
32 |
-
"loss": 1.
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.16,
|
37 |
-
"learning_rate":
|
38 |
-
"loss": 1.
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.19,
|
43 |
-
"learning_rate":
|
44 |
-
"loss": 1.
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.22,
|
49 |
-
"learning_rate":
|
50 |
-
"loss": 1.
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.25,
|
55 |
-
"learning_rate":
|
56 |
-
"loss":
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 0.29,
|
61 |
-
"learning_rate":
|
62 |
-
"loss": 0.
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 0.32,
|
67 |
-
"learning_rate":
|
68 |
-
"loss": 0.
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.35,
|
73 |
-
"learning_rate":
|
74 |
-
"loss": 0.
|
75 |
"step": 110
|
76 |
},
|
77 |
{
|
78 |
"epoch": 0.38,
|
79 |
-
"learning_rate":
|
80 |
-
"loss": 0.
|
81 |
"step": 120
|
82 |
},
|
83 |
{
|
84 |
"epoch": 0.41,
|
85 |
-
"learning_rate": 4.
|
86 |
-
"loss": 0.
|
87 |
"step": 130
|
88 |
},
|
89 |
{
|
90 |
"epoch": 0.45,
|
91 |
-
"learning_rate": 4.
|
92 |
-
"loss": 0.
|
93 |
"step": 140
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.48,
|
97 |
-
"learning_rate": 4.
|
98 |
-
"loss": 0.
|
99 |
"step": 150
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.51,
|
103 |
-
"learning_rate": 4.
|
104 |
-
"loss": 0.
|
105 |
"step": 160
|
106 |
},
|
107 |
{
|
108 |
"epoch": 0.54,
|
109 |
-
"learning_rate": 4.
|
110 |
-
"loss": 0.
|
111 |
"step": 170
|
112 |
},
|
113 |
{
|
114 |
"epoch": 0.57,
|
115 |
-
"learning_rate": 4.
|
116 |
-
"loss": 0.
|
117 |
"step": 180
|
118 |
},
|
119 |
{
|
120 |
"epoch": 0.6,
|
121 |
-
"learning_rate": 4.
|
122 |
-
"loss": 0.
|
123 |
"step": 190
|
124 |
},
|
125 |
{
|
126 |
"epoch": 0.64,
|
127 |
-
"learning_rate": 4.
|
128 |
-
"loss": 0.
|
129 |
"step": 200
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.67,
|
133 |
-
"learning_rate": 4.
|
134 |
-
"loss": 0.
|
135 |
"step": 210
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.7,
|
139 |
-
"learning_rate": 4.
|
140 |
-
"loss": 0.
|
141 |
"step": 220
|
142 |
},
|
143 |
{
|
144 |
"epoch": 0.73,
|
145 |
-
"learning_rate": 4.
|
146 |
-
"loss": 0.
|
147 |
"step": 230
|
148 |
},
|
149 |
{
|
150 |
"epoch": 0.76,
|
151 |
-
"learning_rate": 4.
|
152 |
-
"loss": 0.
|
153 |
"step": 240
|
154 |
},
|
155 |
{
|
156 |
"epoch": 0.79,
|
157 |
-
"learning_rate": 4.
|
158 |
-
"loss": 0.
|
159 |
"step": 250
|
160 |
},
|
161 |
{
|
162 |
"epoch": 0.83,
|
163 |
-
"learning_rate": 4.
|
164 |
-
"loss": 0.
|
165 |
"step": 260
|
166 |
},
|
167 |
{
|
168 |
"epoch": 0.86,
|
169 |
-
"learning_rate":
|
170 |
-
"loss": 0.
|
171 |
"step": 270
|
172 |
},
|
173 |
{
|
174 |
"epoch": 0.89,
|
175 |
-
"learning_rate":
|
176 |
-
"loss": 0.
|
177 |
"step": 280
|
178 |
},
|
179 |
{
|
180 |
"epoch": 0.92,
|
181 |
-
"learning_rate":
|
182 |
-
"loss": 0.
|
183 |
"step": 290
|
184 |
},
|
185 |
{
|
186 |
"epoch": 0.95,
|
187 |
-
"learning_rate":
|
188 |
-
"loss": 0.
|
189 |
"step": 300
|
190 |
},
|
191 |
{
|
192 |
"epoch": 0.99,
|
193 |
-
"learning_rate":
|
194 |
-
"loss": 0.
|
195 |
"step": 310
|
196 |
},
|
197 |
{
|
198 |
"epoch": 1.0,
|
199 |
-
"eval_accuracy": 0.
|
200 |
-
"eval_loss": 0.
|
201 |
-
"eval_runtime":
|
202 |
-
"eval_samples_per_second":
|
203 |
-
"eval_steps_per_second": 4.
|
204 |
"step": 314
|
205 |
},
|
206 |
{
|
207 |
"epoch": 1.02,
|
208 |
-
"learning_rate":
|
209 |
-
"loss": 0.
|
210 |
"step": 320
|
211 |
},
|
212 |
{
|
213 |
"epoch": 1.05,
|
214 |
-
"learning_rate":
|
215 |
-
"loss": 0.
|
216 |
"step": 330
|
217 |
},
|
218 |
{
|
219 |
"epoch": 1.08,
|
220 |
-
"learning_rate":
|
221 |
-
"loss": 0.
|
222 |
"step": 340
|
223 |
},
|
224 |
{
|
225 |
"epoch": 1.11,
|
226 |
-
"learning_rate":
|
227 |
-
"loss": 0.
|
228 |
"step": 350
|
229 |
},
|
230 |
{
|
231 |
"epoch": 1.14,
|
232 |
-
"learning_rate":
|
233 |
-
"loss": 0.
|
234 |
"step": 360
|
235 |
},
|
236 |
{
|
237 |
"epoch": 1.18,
|
238 |
-
"learning_rate":
|
239 |
-
"loss": 0.
|
240 |
"step": 370
|
241 |
},
|
242 |
{
|
243 |
"epoch": 1.21,
|
244 |
-
"learning_rate":
|
245 |
-
"loss": 0.
|
246 |
"step": 380
|
247 |
},
|
248 |
{
|
249 |
"epoch": 1.24,
|
250 |
-
"learning_rate":
|
251 |
-
"loss": 0.
|
252 |
"step": 390
|
253 |
},
|
254 |
{
|
255 |
"epoch": 1.27,
|
256 |
-
"learning_rate":
|
257 |
-
"loss": 0.
|
258 |
"step": 400
|
259 |
},
|
260 |
{
|
261 |
"epoch": 1.3,
|
262 |
-
"learning_rate":
|
263 |
-
"loss": 0.
|
264 |
"step": 410
|
265 |
},
|
266 |
{
|
267 |
"epoch": 1.34,
|
268 |
-
"learning_rate":
|
269 |
-
"loss": 0.
|
270 |
"step": 420
|
271 |
},
|
272 |
{
|
273 |
"epoch": 1.37,
|
274 |
-
"learning_rate":
|
275 |
-
"loss": 0.
|
276 |
"step": 430
|
277 |
},
|
278 |
{
|
279 |
"epoch": 1.4,
|
280 |
-
"learning_rate":
|
281 |
-
"loss": 0.
|
282 |
"step": 440
|
283 |
},
|
284 |
{
|
285 |
"epoch": 1.43,
|
286 |
-
"learning_rate":
|
287 |
-
"loss": 0.
|
288 |
"step": 450
|
289 |
},
|
290 |
{
|
291 |
"epoch": 1.46,
|
292 |
-
"learning_rate":
|
293 |
-
"loss": 0.
|
294 |
"step": 460
|
295 |
},
|
296 |
{
|
297 |
"epoch": 1.49,
|
298 |
-
"learning_rate":
|
299 |
-
"loss": 0.
|
300 |
"step": 470
|
301 |
},
|
302 |
{
|
303 |
"epoch": 1.53,
|
304 |
-
"learning_rate":
|
305 |
-
"loss": 0.
|
306 |
"step": 480
|
307 |
},
|
308 |
{
|
309 |
"epoch": 1.56,
|
310 |
-
"learning_rate":
|
311 |
-
"loss": 0.
|
312 |
"step": 490
|
313 |
},
|
314 |
{
|
315 |
"epoch": 1.59,
|
316 |
-
"learning_rate":
|
317 |
-
"loss": 0.
|
318 |
"step": 500
|
319 |
},
|
320 |
{
|
321 |
"epoch": 1.62,
|
322 |
-
"learning_rate":
|
323 |
-
"loss": 0.
|
324 |
"step": 510
|
325 |
},
|
326 |
{
|
327 |
"epoch": 1.65,
|
328 |
-
"learning_rate":
|
329 |
-
"loss": 0.
|
330 |
"step": 520
|
331 |
},
|
332 |
{
|
333 |
"epoch": 1.69,
|
334 |
-
"learning_rate":
|
335 |
-
"loss": 0.
|
336 |
"step": 530
|
337 |
},
|
338 |
{
|
339 |
"epoch": 1.72,
|
340 |
-
"learning_rate":
|
341 |
-
"loss": 0.
|
342 |
"step": 540
|
343 |
},
|
344 |
{
|
345 |
"epoch": 1.75,
|
346 |
-
"learning_rate":
|
347 |
-
"loss": 0.
|
348 |
"step": 550
|
349 |
},
|
350 |
{
|
351 |
"epoch": 1.78,
|
352 |
-
"learning_rate":
|
353 |
-
"loss": 0.
|
354 |
"step": 560
|
355 |
},
|
356 |
{
|
357 |
"epoch": 1.81,
|
358 |
-
"learning_rate":
|
359 |
-
"loss": 0.
|
360 |
"step": 570
|
361 |
},
|
362 |
{
|
363 |
"epoch": 1.84,
|
364 |
-
"learning_rate":
|
365 |
-
"loss": 0.
|
366 |
"step": 580
|
367 |
},
|
368 |
{
|
369 |
"epoch": 1.88,
|
370 |
-
"learning_rate":
|
371 |
-
"loss": 0.
|
372 |
"step": 590
|
373 |
},
|
374 |
{
|
375 |
"epoch": 1.91,
|
376 |
-
"learning_rate":
|
377 |
-
"loss": 0.
|
378 |
"step": 600
|
379 |
},
|
380 |
{
|
381 |
"epoch": 1.94,
|
382 |
-
"learning_rate":
|
383 |
-
"loss": 0.
|
384 |
"step": 610
|
385 |
},
|
386 |
{
|
387 |
"epoch": 1.97,
|
388 |
-
"learning_rate":
|
389 |
-
"loss": 0.
|
390 |
"step": 620
|
391 |
},
|
392 |
{
|
393 |
"epoch": 2.0,
|
394 |
-
"eval_accuracy": 0.
|
395 |
-
"eval_loss": 0.
|
396 |
-
"eval_runtime": 32.
|
397 |
-
"eval_samples_per_second":
|
398 |
-
"eval_steps_per_second": 4.
|
399 |
"step": 629
|
400 |
},
|
401 |
{
|
402 |
"epoch": 2.0,
|
403 |
-
"learning_rate":
|
404 |
-
"loss": 0.
|
405 |
"step": 630
|
406 |
},
|
407 |
{
|
408 |
"epoch": 2.03,
|
409 |
-
"learning_rate":
|
410 |
-
"loss": 0.
|
411 |
"step": 640
|
412 |
},
|
413 |
{
|
414 |
"epoch": 2.07,
|
415 |
-
"learning_rate":
|
416 |
-
"loss": 0.
|
417 |
"step": 650
|
418 |
},
|
419 |
{
|
420 |
"epoch": 2.1,
|
421 |
-
"learning_rate":
|
422 |
-
"loss": 0.
|
423 |
"step": 660
|
424 |
},
|
425 |
{
|
426 |
"epoch": 2.13,
|
427 |
-
"learning_rate":
|
428 |
-
"loss": 0.
|
429 |
"step": 670
|
430 |
},
|
431 |
{
|
432 |
"epoch": 2.16,
|
433 |
-
"learning_rate":
|
434 |
-
"loss": 0.
|
435 |
"step": 680
|
436 |
},
|
437 |
{
|
438 |
"epoch": 2.19,
|
439 |
-
"learning_rate":
|
440 |
-
"loss": 0.
|
441 |
"step": 690
|
442 |
},
|
443 |
{
|
444 |
"epoch": 2.23,
|
445 |
-
"learning_rate":
|
446 |
-
"loss": 0.
|
447 |
"step": 700
|
448 |
},
|
449 |
{
|
450 |
"epoch": 2.26,
|
451 |
-
"learning_rate":
|
452 |
-
"loss": 0.
|
453 |
"step": 710
|
454 |
},
|
455 |
{
|
456 |
"epoch": 2.29,
|
457 |
-
"learning_rate":
|
458 |
-
"loss": 0.
|
459 |
"step": 720
|
460 |
},
|
461 |
{
|
462 |
"epoch": 2.32,
|
463 |
-
"learning_rate":
|
464 |
-
"loss": 0.
|
465 |
"step": 730
|
466 |
},
|
467 |
{
|
468 |
"epoch": 2.35,
|
469 |
-
"learning_rate":
|
470 |
-
"loss": 0.
|
471 |
"step": 740
|
472 |
},
|
473 |
{
|
474 |
"epoch": 2.38,
|
475 |
-
"learning_rate":
|
476 |
-
"loss": 0.
|
477 |
"step": 750
|
478 |
},
|
479 |
{
|
480 |
"epoch": 2.42,
|
481 |
-
"learning_rate":
|
482 |
-
"loss": 0.
|
483 |
"step": 760
|
484 |
},
|
485 |
{
|
486 |
"epoch": 2.45,
|
487 |
-
"learning_rate":
|
488 |
-
"loss": 0.
|
489 |
"step": 770
|
490 |
},
|
491 |
{
|
492 |
"epoch": 2.48,
|
493 |
-
"learning_rate":
|
494 |
-
"loss": 0.
|
495 |
"step": 780
|
496 |
},
|
497 |
{
|
498 |
"epoch": 2.51,
|
499 |
-
"learning_rate":
|
500 |
-
"loss": 0.
|
501 |
"step": 790
|
502 |
},
|
503 |
{
|
504 |
"epoch": 2.54,
|
505 |
-
"learning_rate":
|
506 |
-
"loss": 0.
|
507 |
"step": 800
|
508 |
},
|
509 |
{
|
510 |
"epoch": 2.58,
|
511 |
-
"learning_rate":
|
512 |
-
"loss": 0.
|
513 |
"step": 810
|
514 |
},
|
515 |
{
|
516 |
"epoch": 2.61,
|
517 |
-
"learning_rate":
|
518 |
-
"loss": 0.
|
519 |
"step": 820
|
520 |
},
|
521 |
{
|
522 |
"epoch": 2.64,
|
523 |
-
"learning_rate":
|
524 |
-
"loss": 0.
|
525 |
"step": 830
|
526 |
},
|
527 |
{
|
528 |
"epoch": 2.67,
|
529 |
-
"learning_rate":
|
530 |
-
"loss": 0.
|
531 |
"step": 840
|
532 |
},
|
533 |
{
|
534 |
"epoch": 2.7,
|
535 |
-
"learning_rate":
|
536 |
-
"loss": 0.
|
537 |
"step": 850
|
538 |
},
|
539 |
{
|
540 |
"epoch": 2.73,
|
541 |
-
"learning_rate":
|
542 |
-
"loss": 0.
|
543 |
"step": 860
|
544 |
},
|
545 |
{
|
546 |
"epoch": 2.77,
|
547 |
-
"learning_rate":
|
548 |
-
"loss": 0.
|
549 |
"step": 870
|
550 |
},
|
551 |
{
|
552 |
"epoch": 2.8,
|
553 |
-
"learning_rate":
|
554 |
-
"loss": 0.
|
555 |
"step": 880
|
556 |
},
|
557 |
{
|
558 |
"epoch": 2.83,
|
559 |
-
"learning_rate":
|
560 |
-
"loss": 0.
|
561 |
"step": 890
|
562 |
},
|
563 |
{
|
564 |
"epoch": 2.86,
|
565 |
-
"learning_rate": 2.
|
566 |
-
"loss": 0.
|
567 |
"step": 900
|
568 |
},
|
569 |
{
|
570 |
"epoch": 2.89,
|
571 |
-
"learning_rate":
|
572 |
-
"loss": 0.
|
573 |
"step": 910
|
574 |
},
|
575 |
{
|
576 |
"epoch": 2.93,
|
577 |
-
"learning_rate":
|
578 |
-
"loss": 0.
|
579 |
"step": 920
|
580 |
},
|
581 |
{
|
582 |
"epoch": 2.96,
|
583 |
-
"learning_rate":
|
584 |
-
"loss": 0.
|
585 |
"step": 930
|
586 |
},
|
587 |
{
|
588 |
"epoch": 2.99,
|
589 |
-
"learning_rate":
|
590 |
-
"loss": 0.
|
591 |
"step": 940
|
592 |
},
|
593 |
{
|
594 |
"epoch": 3.0,
|
595 |
-
"eval_accuracy": 0.
|
596 |
-
"eval_loss": 0.
|
597 |
-
"eval_runtime":
|
598 |
-
"eval_samples_per_second":
|
599 |
-
"eval_steps_per_second": 4.
|
600 |
-
"step":
|
601 |
},
|
602 |
{
|
603 |
-
"epoch": 3.
|
604 |
-
"
|
605 |
-
"
|
606 |
-
"
|
607 |
-
|
608 |
-
|
609 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
610 |
}
|
611 |
],
|
612 |
"logging_steps": 10,
|
613 |
-
"max_steps":
|
614 |
-
"num_train_epochs":
|
615 |
"save_steps": 500,
|
616 |
-
"total_flos":
|
617 |
"trial_name": null,
|
618 |
"trial_params": null
|
619 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.7321109123434705,
|
3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-img_orientation/checkpoint-1258",
|
4 |
+
"epoch": 4.992050874403816,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 1570,
|
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": 3.1847133757961785e-06,
|
14 |
+
"loss": 1.4185,
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.06,
|
19 |
+
"learning_rate": 6.369426751592357e-06,
|
20 |
+
"loss": 1.41,
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.1,
|
25 |
+
"learning_rate": 9.554140127388536e-06,
|
26 |
+
"loss": 1.3794,
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.13,
|
31 |
+
"learning_rate": 1.2738853503184714e-05,
|
32 |
+
"loss": 1.3324,
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.16,
|
37 |
+
"learning_rate": 1.592356687898089e-05,
|
38 |
+
"loss": 1.3049,
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.19,
|
43 |
+
"learning_rate": 1.910828025477707e-05,
|
44 |
+
"loss": 1.2419,
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.22,
|
49 |
+
"learning_rate": 2.229299363057325e-05,
|
50 |
+
"loss": 1.1568,
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.25,
|
55 |
+
"learning_rate": 2.5477707006369428e-05,
|
56 |
+
"loss": 1.0499,
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 0.29,
|
61 |
+
"learning_rate": 2.8662420382165606e-05,
|
62 |
+
"loss": 0.9764,
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 0.32,
|
67 |
+
"learning_rate": 3.184713375796178e-05,
|
68 |
+
"loss": 0.8664,
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.35,
|
73 |
+
"learning_rate": 3.503184713375796e-05,
|
74 |
+
"loss": 0.814,
|
75 |
"step": 110
|
76 |
},
|
77 |
{
|
78 |
"epoch": 0.38,
|
79 |
+
"learning_rate": 3.821656050955414e-05,
|
80 |
+
"loss": 0.7859,
|
81 |
"step": 120
|
82 |
},
|
83 |
{
|
84 |
"epoch": 0.41,
|
85 |
+
"learning_rate": 4.1401273885350325e-05,
|
86 |
+
"loss": 0.7458,
|
87 |
"step": 130
|
88 |
},
|
89 |
{
|
90 |
"epoch": 0.45,
|
91 |
+
"learning_rate": 4.45859872611465e-05,
|
92 |
+
"loss": 0.782,
|
93 |
"step": 140
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.48,
|
97 |
+
"learning_rate": 4.777070063694268e-05,
|
98 |
+
"loss": 0.7633,
|
99 |
"step": 150
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.51,
|
103 |
+
"learning_rate": 4.9893842887473466e-05,
|
104 |
+
"loss": 0.7261,
|
105 |
"step": 160
|
106 |
},
|
107 |
{
|
108 |
"epoch": 0.54,
|
109 |
+
"learning_rate": 4.9539985845718334e-05,
|
110 |
+
"loss": 0.7286,
|
111 |
"step": 170
|
112 |
},
|
113 |
{
|
114 |
"epoch": 0.57,
|
115 |
+
"learning_rate": 4.91861288039632e-05,
|
116 |
+
"loss": 0.6912,
|
117 |
"step": 180
|
118 |
},
|
119 |
{
|
120 |
"epoch": 0.6,
|
121 |
+
"learning_rate": 4.883227176220807e-05,
|
122 |
+
"loss": 0.7093,
|
123 |
"step": 190
|
124 |
},
|
125 |
{
|
126 |
"epoch": 0.64,
|
127 |
+
"learning_rate": 4.8478414720452945e-05,
|
128 |
+
"loss": 0.6893,
|
129 |
"step": 200
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.67,
|
133 |
+
"learning_rate": 4.8124557678697806e-05,
|
134 |
+
"loss": 0.7246,
|
135 |
"step": 210
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.7,
|
139 |
+
"learning_rate": 4.777070063694268e-05,
|
140 |
+
"loss": 0.7175,
|
141 |
"step": 220
|
142 |
},
|
143 |
{
|
144 |
"epoch": 0.73,
|
145 |
+
"learning_rate": 4.741684359518755e-05,
|
146 |
+
"loss": 0.6927,
|
147 |
"step": 230
|
148 |
},
|
149 |
{
|
150 |
"epoch": 0.76,
|
151 |
+
"learning_rate": 4.706298655343242e-05,
|
152 |
+
"loss": 0.6424,
|
153 |
"step": 240
|
154 |
},
|
155 |
{
|
156 |
"epoch": 0.79,
|
157 |
+
"learning_rate": 4.6709129511677285e-05,
|
158 |
+
"loss": 0.6573,
|
159 |
"step": 250
|
160 |
},
|
161 |
{
|
162 |
"epoch": 0.83,
|
163 |
+
"learning_rate": 4.635527246992216e-05,
|
164 |
+
"loss": 0.6852,
|
165 |
"step": 260
|
166 |
},
|
167 |
{
|
168 |
"epoch": 0.86,
|
169 |
+
"learning_rate": 4.600141542816702e-05,
|
170 |
+
"loss": 0.6829,
|
171 |
"step": 270
|
172 |
},
|
173 |
{
|
174 |
"epoch": 0.89,
|
175 |
+
"learning_rate": 4.5647558386411895e-05,
|
176 |
+
"loss": 0.659,
|
177 |
"step": 280
|
178 |
},
|
179 |
{
|
180 |
"epoch": 0.92,
|
181 |
+
"learning_rate": 4.529370134465676e-05,
|
182 |
+
"loss": 0.6117,
|
183 |
"step": 290
|
184 |
},
|
185 |
{
|
186 |
"epoch": 0.95,
|
187 |
+
"learning_rate": 4.493984430290163e-05,
|
188 |
+
"loss": 0.6627,
|
189 |
"step": 300
|
190 |
},
|
191 |
{
|
192 |
"epoch": 0.99,
|
193 |
+
"learning_rate": 4.45859872611465e-05,
|
194 |
+
"loss": 0.6531,
|
195 |
"step": 310
|
196 |
},
|
197 |
{
|
198 |
"epoch": 1.0,
|
199 |
+
"eval_accuracy": 0.7043828264758497,
|
200 |
+
"eval_loss": 0.5172476768493652,
|
201 |
+
"eval_runtime": 33.3984,
|
202 |
+
"eval_samples_per_second": 133.899,
|
203 |
+
"eval_steps_per_second": 4.192,
|
204 |
"step": 314
|
205 |
},
|
206 |
{
|
207 |
"epoch": 1.02,
|
208 |
+
"learning_rate": 4.423213021939137e-05,
|
209 |
+
"loss": 0.6315,
|
210 |
"step": 320
|
211 |
},
|
212 |
{
|
213 |
"epoch": 1.05,
|
214 |
+
"learning_rate": 4.3878273177636235e-05,
|
215 |
+
"loss": 0.6119,
|
216 |
"step": 330
|
217 |
},
|
218 |
{
|
219 |
"epoch": 1.08,
|
220 |
+
"learning_rate": 4.352441613588111e-05,
|
221 |
+
"loss": 0.6335,
|
222 |
"step": 340
|
223 |
},
|
224 |
{
|
225 |
"epoch": 1.11,
|
226 |
+
"learning_rate": 4.317055909412597e-05,
|
227 |
+
"loss": 0.6371,
|
228 |
"step": 350
|
229 |
},
|
230 |
{
|
231 |
"epoch": 1.14,
|
232 |
+
"learning_rate": 4.2816702052370846e-05,
|
233 |
+
"loss": 0.5907,
|
234 |
"step": 360
|
235 |
},
|
236 |
{
|
237 |
"epoch": 1.18,
|
238 |
+
"learning_rate": 4.2462845010615714e-05,
|
239 |
+
"loss": 0.6202,
|
240 |
"step": 370
|
241 |
},
|
242 |
{
|
243 |
"epoch": 1.21,
|
244 |
+
"learning_rate": 4.210898796886058e-05,
|
245 |
+
"loss": 0.6195,
|
246 |
"step": 380
|
247 |
},
|
248 |
{
|
249 |
"epoch": 1.24,
|
250 |
+
"learning_rate": 4.175513092710545e-05,
|
251 |
+
"loss": 0.6165,
|
252 |
"step": 390
|
253 |
},
|
254 |
{
|
255 |
"epoch": 1.27,
|
256 |
+
"learning_rate": 4.1401273885350325e-05,
|
257 |
+
"loss": 0.6096,
|
258 |
"step": 400
|
259 |
},
|
260 |
{
|
261 |
"epoch": 1.3,
|
262 |
+
"learning_rate": 4.1047416843595186e-05,
|
263 |
+
"loss": 0.5847,
|
264 |
"step": 410
|
265 |
},
|
266 |
{
|
267 |
"epoch": 1.34,
|
268 |
+
"learning_rate": 4.069355980184006e-05,
|
269 |
+
"loss": 0.5893,
|
270 |
"step": 420
|
271 |
},
|
272 |
{
|
273 |
"epoch": 1.37,
|
274 |
+
"learning_rate": 4.033970276008493e-05,
|
275 |
+
"loss": 0.6183,
|
276 |
"step": 430
|
277 |
},
|
278 |
{
|
279 |
"epoch": 1.4,
|
280 |
+
"learning_rate": 3.99858457183298e-05,
|
281 |
+
"loss": 0.5857,
|
282 |
"step": 440
|
283 |
},
|
284 |
{
|
285 |
"epoch": 1.43,
|
286 |
+
"learning_rate": 3.9631988676574665e-05,
|
287 |
+
"loss": 0.5899,
|
288 |
"step": 450
|
289 |
},
|
290 |
{
|
291 |
"epoch": 1.46,
|
292 |
+
"learning_rate": 3.927813163481954e-05,
|
293 |
+
"loss": 0.6146,
|
294 |
"step": 460
|
295 |
},
|
296 |
{
|
297 |
"epoch": 1.49,
|
298 |
+
"learning_rate": 3.89242745930644e-05,
|
299 |
+
"loss": 0.5822,
|
300 |
"step": 470
|
301 |
},
|
302 |
{
|
303 |
"epoch": 1.53,
|
304 |
+
"learning_rate": 3.8570417551309275e-05,
|
305 |
+
"loss": 0.5783,
|
306 |
"step": 480
|
307 |
},
|
308 |
{
|
309 |
"epoch": 1.56,
|
310 |
+
"learning_rate": 3.821656050955414e-05,
|
311 |
+
"loss": 0.6289,
|
312 |
"step": 490
|
313 |
},
|
314 |
{
|
315 |
"epoch": 1.59,
|
316 |
+
"learning_rate": 3.786270346779901e-05,
|
317 |
+
"loss": 0.5934,
|
318 |
"step": 500
|
319 |
},
|
320 |
{
|
321 |
"epoch": 1.62,
|
322 |
+
"learning_rate": 3.750884642604388e-05,
|
323 |
+
"loss": 0.5836,
|
324 |
"step": 510
|
325 |
},
|
326 |
{
|
327 |
"epoch": 1.65,
|
328 |
+
"learning_rate": 3.715498938428875e-05,
|
329 |
+
"loss": 0.566,
|
330 |
"step": 520
|
331 |
},
|
332 |
{
|
333 |
"epoch": 1.69,
|
334 |
+
"learning_rate": 3.6801132342533615e-05,
|
335 |
+
"loss": 0.5933,
|
336 |
"step": 530
|
337 |
},
|
338 |
{
|
339 |
"epoch": 1.72,
|
340 |
+
"learning_rate": 3.644727530077849e-05,
|
341 |
+
"loss": 0.5699,
|
342 |
"step": 540
|
343 |
},
|
344 |
{
|
345 |
"epoch": 1.75,
|
346 |
+
"learning_rate": 3.609341825902335e-05,
|
347 |
+
"loss": 0.587,
|
348 |
"step": 550
|
349 |
},
|
350 |
{
|
351 |
"epoch": 1.78,
|
352 |
+
"learning_rate": 3.5739561217268226e-05,
|
353 |
+
"loss": 0.5909,
|
354 |
"step": 560
|
355 |
},
|
356 |
{
|
357 |
"epoch": 1.81,
|
358 |
+
"learning_rate": 3.5385704175513094e-05,
|
359 |
+
"loss": 0.5825,
|
360 |
"step": 570
|
361 |
},
|
362 |
{
|
363 |
"epoch": 1.84,
|
364 |
+
"learning_rate": 3.503184713375796e-05,
|
365 |
+
"loss": 0.5725,
|
366 |
"step": 580
|
367 |
},
|
368 |
{
|
369 |
"epoch": 1.88,
|
370 |
+
"learning_rate": 3.467799009200283e-05,
|
371 |
+
"loss": 0.5761,
|
372 |
"step": 590
|
373 |
},
|
374 |
{
|
375 |
"epoch": 1.91,
|
376 |
+
"learning_rate": 3.4324133050247705e-05,
|
377 |
+
"loss": 0.5869,
|
378 |
"step": 600
|
379 |
},
|
380 |
{
|
381 |
"epoch": 1.94,
|
382 |
+
"learning_rate": 3.3970276008492566e-05,
|
383 |
+
"loss": 0.623,
|
384 |
"step": 610
|
385 |
},
|
386 |
{
|
387 |
"epoch": 1.97,
|
388 |
+
"learning_rate": 3.361641896673744e-05,
|
389 |
+
"loss": 0.5774,
|
390 |
"step": 620
|
391 |
},
|
392 |
{
|
393 |
"epoch": 2.0,
|
394 |
+
"eval_accuracy": 0.7240608228980322,
|
395 |
+
"eval_loss": 0.461431086063385,
|
396 |
+
"eval_runtime": 32.7276,
|
397 |
+
"eval_samples_per_second": 136.643,
|
398 |
+
"eval_steps_per_second": 4.278,
|
399 |
"step": 629
|
400 |
},
|
401 |
{
|
402 |
"epoch": 2.0,
|
403 |
+
"learning_rate": 3.326256192498231e-05,
|
404 |
+
"loss": 0.5664,
|
405 |
"step": 630
|
406 |
},
|
407 |
{
|
408 |
"epoch": 2.03,
|
409 |
+
"learning_rate": 3.2908704883227177e-05,
|
410 |
+
"loss": 0.5354,
|
411 |
"step": 640
|
412 |
},
|
413 |
{
|
414 |
"epoch": 2.07,
|
415 |
+
"learning_rate": 3.2554847841472044e-05,
|
416 |
+
"loss": 0.5533,
|
417 |
"step": 650
|
418 |
},
|
419 |
{
|
420 |
"epoch": 2.1,
|
421 |
+
"learning_rate": 3.220099079971692e-05,
|
422 |
+
"loss": 0.5576,
|
423 |
"step": 660
|
424 |
},
|
425 |
{
|
426 |
"epoch": 2.13,
|
427 |
+
"learning_rate": 3.184713375796178e-05,
|
428 |
+
"loss": 0.5637,
|
429 |
"step": 670
|
430 |
},
|
431 |
{
|
432 |
"epoch": 2.16,
|
433 |
+
"learning_rate": 3.1493276716206655e-05,
|
434 |
+
"loss": 0.5502,
|
435 |
"step": 680
|
436 |
},
|
437 |
{
|
438 |
"epoch": 2.19,
|
439 |
+
"learning_rate": 3.113941967445152e-05,
|
440 |
+
"loss": 0.5451,
|
441 |
"step": 690
|
442 |
},
|
443 |
{
|
444 |
"epoch": 2.23,
|
445 |
+
"learning_rate": 3.078556263269639e-05,
|
446 |
+
"loss": 0.5996,
|
447 |
"step": 700
|
448 |
},
|
449 |
{
|
450 |
"epoch": 2.26,
|
451 |
+
"learning_rate": 3.0431705590941263e-05,
|
452 |
+
"loss": 0.5754,
|
453 |
"step": 710
|
454 |
},
|
455 |
{
|
456 |
"epoch": 2.29,
|
457 |
+
"learning_rate": 3.0077848549186134e-05,
|
458 |
+
"loss": 0.5321,
|
459 |
"step": 720
|
460 |
},
|
461 |
{
|
462 |
"epoch": 2.32,
|
463 |
+
"learning_rate": 2.9723991507431e-05,
|
464 |
+
"loss": 0.5443,
|
465 |
"step": 730
|
466 |
},
|
467 |
{
|
468 |
"epoch": 2.35,
|
469 |
+
"learning_rate": 2.937013446567587e-05,
|
470 |
+
"loss": 0.5471,
|
471 |
"step": 740
|
472 |
},
|
473 |
{
|
474 |
"epoch": 2.38,
|
475 |
+
"learning_rate": 2.9016277423920734e-05,
|
476 |
+
"loss": 0.5147,
|
477 |
"step": 750
|
478 |
},
|
479 |
{
|
480 |
"epoch": 2.42,
|
481 |
+
"learning_rate": 2.8662420382165606e-05,
|
482 |
+
"loss": 0.556,
|
483 |
"step": 760
|
484 |
},
|
485 |
{
|
486 |
"epoch": 2.45,
|
487 |
+
"learning_rate": 2.8308563340410477e-05,
|
488 |
+
"loss": 0.5613,
|
489 |
"step": 770
|
490 |
},
|
491 |
{
|
492 |
"epoch": 2.48,
|
493 |
+
"learning_rate": 2.7954706298655342e-05,
|
494 |
+
"loss": 0.5748,
|
495 |
"step": 780
|
496 |
},
|
497 |
{
|
498 |
"epoch": 2.51,
|
499 |
+
"learning_rate": 2.7600849256900213e-05,
|
500 |
+
"loss": 0.5698,
|
501 |
"step": 790
|
502 |
},
|
503 |
{
|
504 |
"epoch": 2.54,
|
505 |
+
"learning_rate": 2.7246992215145084e-05,
|
506 |
+
"loss": 0.561,
|
507 |
"step": 800
|
508 |
},
|
509 |
{
|
510 |
"epoch": 2.58,
|
511 |
+
"learning_rate": 2.689313517338995e-05,
|
512 |
+
"loss": 0.5444,
|
513 |
"step": 810
|
514 |
},
|
515 |
{
|
516 |
"epoch": 2.61,
|
517 |
+
"learning_rate": 2.653927813163482e-05,
|
518 |
+
"loss": 0.5156,
|
519 |
"step": 820
|
520 |
},
|
521 |
{
|
522 |
"epoch": 2.64,
|
523 |
+
"learning_rate": 2.6185421089879692e-05,
|
524 |
+
"loss": 0.5358,
|
525 |
"step": 830
|
526 |
},
|
527 |
{
|
528 |
"epoch": 2.67,
|
529 |
+
"learning_rate": 2.5831564048124556e-05,
|
530 |
+
"loss": 0.5156,
|
531 |
"step": 840
|
532 |
},
|
533 |
{
|
534 |
"epoch": 2.7,
|
535 |
+
"learning_rate": 2.5477707006369428e-05,
|
536 |
+
"loss": 0.5386,
|
537 |
"step": 850
|
538 |
},
|
539 |
{
|
540 |
"epoch": 2.73,
|
541 |
+
"learning_rate": 2.51238499646143e-05,
|
542 |
+
"loss": 0.5513,
|
543 |
"step": 860
|
544 |
},
|
545 |
{
|
546 |
"epoch": 2.77,
|
547 |
+
"learning_rate": 2.4769992922859167e-05,
|
548 |
+
"loss": 0.5217,
|
549 |
"step": 870
|
550 |
},
|
551 |
{
|
552 |
"epoch": 2.8,
|
553 |
+
"learning_rate": 2.4416135881104035e-05,
|
554 |
+
"loss": 0.5621,
|
555 |
"step": 880
|
556 |
},
|
557 |
{
|
558 |
"epoch": 2.83,
|
559 |
+
"learning_rate": 2.4062278839348903e-05,
|
560 |
+
"loss": 0.5457,
|
561 |
"step": 890
|
562 |
},
|
563 |
{
|
564 |
"epoch": 2.86,
|
565 |
+
"learning_rate": 2.3708421797593774e-05,
|
566 |
+
"loss": 0.5172,
|
567 |
"step": 900
|
568 |
},
|
569 |
{
|
570 |
"epoch": 2.89,
|
571 |
+
"learning_rate": 2.3354564755838642e-05,
|
572 |
+
"loss": 0.5598,
|
573 |
"step": 910
|
574 |
},
|
575 |
{
|
576 |
"epoch": 2.93,
|
577 |
+
"learning_rate": 2.300070771408351e-05,
|
578 |
+
"loss": 0.5391,
|
579 |
"step": 920
|
580 |
},
|
581 |
{
|
582 |
"epoch": 2.96,
|
583 |
+
"learning_rate": 2.264685067232838e-05,
|
584 |
+
"loss": 0.5284,
|
585 |
"step": 930
|
586 |
},
|
587 |
{
|
588 |
"epoch": 2.99,
|
589 |
+
"learning_rate": 2.229299363057325e-05,
|
590 |
+
"loss": 0.5496,
|
591 |
"step": 940
|
592 |
},
|
593 |
{
|
594 |
"epoch": 3.0,
|
595 |
+
"eval_accuracy": 0.7298747763864043,
|
596 |
+
"eval_loss": 0.4499094486236572,
|
597 |
+
"eval_runtime": 32.2417,
|
598 |
+
"eval_samples_per_second": 138.702,
|
599 |
+
"eval_steps_per_second": 4.342,
|
600 |
+
"step": 943
|
601 |
},
|
602 |
{
|
603 |
+
"epoch": 3.02,
|
604 |
+
"learning_rate": 2.1939136588818118e-05,
|
605 |
+
"loss": 0.5027,
|
606 |
+
"step": 950
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 3.05,
|
610 |
+
"learning_rate": 2.1585279547062986e-05,
|
611 |
+
"loss": 0.5285,
|
612 |
+
"step": 960
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 3.08,
|
616 |
+
"learning_rate": 2.1231422505307857e-05,
|
617 |
+
"loss": 0.4934,
|
618 |
+
"step": 970
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 3.12,
|
622 |
+
"learning_rate": 2.0877565463552725e-05,
|
623 |
+
"loss": 0.5001,
|
624 |
+
"step": 980
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 3.15,
|
628 |
+
"learning_rate": 2.0523708421797593e-05,
|
629 |
+
"loss": 0.5378,
|
630 |
+
"step": 990
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 3.18,
|
634 |
+
"learning_rate": 2.0169851380042464e-05,
|
635 |
+
"loss": 0.4971,
|
636 |
+
"step": 1000
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 3.21,
|
640 |
+
"learning_rate": 1.9815994338287332e-05,
|
641 |
+
"loss": 0.5028,
|
642 |
+
"step": 1010
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 3.24,
|
646 |
+
"learning_rate": 1.94621372965322e-05,
|
647 |
+
"loss": 0.5353,
|
648 |
+
"step": 1020
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 3.28,
|
652 |
+
"learning_rate": 1.910828025477707e-05,
|
653 |
+
"loss": 0.5243,
|
654 |
+
"step": 1030
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 3.31,
|
658 |
+
"learning_rate": 1.875442321302194e-05,
|
659 |
+
"loss": 0.5302,
|
660 |
+
"step": 1040
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 3.34,
|
664 |
+
"learning_rate": 1.8400566171266808e-05,
|
665 |
+
"loss": 0.5101,
|
666 |
+
"step": 1050
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 3.37,
|
670 |
+
"learning_rate": 1.8046709129511676e-05,
|
671 |
+
"loss": 0.5032,
|
672 |
+
"step": 1060
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 3.4,
|
676 |
+
"learning_rate": 1.7692852087756547e-05,
|
677 |
+
"loss": 0.5075,
|
678 |
+
"step": 1070
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 3.43,
|
682 |
+
"learning_rate": 1.7338995046001415e-05,
|
683 |
+
"loss": 0.5148,
|
684 |
+
"step": 1080
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 3.47,
|
688 |
+
"learning_rate": 1.6985138004246283e-05,
|
689 |
+
"loss": 0.5152,
|
690 |
+
"step": 1090
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 3.5,
|
694 |
+
"learning_rate": 1.6631280962491154e-05,
|
695 |
+
"loss": 0.5222,
|
696 |
+
"step": 1100
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 3.53,
|
700 |
+
"learning_rate": 1.6277423920736022e-05,
|
701 |
+
"loss": 0.4955,
|
702 |
+
"step": 1110
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 3.56,
|
706 |
+
"learning_rate": 1.592356687898089e-05,
|
707 |
+
"loss": 0.5029,
|
708 |
+
"step": 1120
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 3.59,
|
712 |
+
"learning_rate": 1.556970983722576e-05,
|
713 |
+
"loss": 0.516,
|
714 |
+
"step": 1130
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 3.62,
|
718 |
+
"learning_rate": 1.5215852795470631e-05,
|
719 |
+
"loss": 0.5105,
|
720 |
+
"step": 1140
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 3.66,
|
724 |
+
"learning_rate": 1.48619957537155e-05,
|
725 |
+
"loss": 0.5064,
|
726 |
+
"step": 1150
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 3.69,
|
730 |
+
"learning_rate": 1.4508138711960367e-05,
|
731 |
+
"loss": 0.5082,
|
732 |
+
"step": 1160
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 3.72,
|
736 |
+
"learning_rate": 1.4154281670205239e-05,
|
737 |
+
"loss": 0.4877,
|
738 |
+
"step": 1170
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 3.75,
|
742 |
+
"learning_rate": 1.3800424628450107e-05,
|
743 |
+
"loss": 0.5252,
|
744 |
+
"step": 1180
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 3.78,
|
748 |
+
"learning_rate": 1.3446567586694975e-05,
|
749 |
+
"loss": 0.5415,
|
750 |
+
"step": 1190
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 3.82,
|
754 |
+
"learning_rate": 1.3092710544939846e-05,
|
755 |
+
"loss": 0.4911,
|
756 |
+
"step": 1200
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 3.85,
|
760 |
+
"learning_rate": 1.2738853503184714e-05,
|
761 |
+
"loss": 0.5024,
|
762 |
+
"step": 1210
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 3.88,
|
766 |
+
"learning_rate": 1.2384996461429584e-05,
|
767 |
+
"loss": 0.5201,
|
768 |
+
"step": 1220
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 3.91,
|
772 |
+
"learning_rate": 1.2031139419674452e-05,
|
773 |
+
"loss": 0.5291,
|
774 |
+
"step": 1230
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 3.94,
|
778 |
+
"learning_rate": 1.1677282377919321e-05,
|
779 |
+
"loss": 0.5271,
|
780 |
+
"step": 1240
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 3.97,
|
784 |
+
"learning_rate": 1.132342533616419e-05,
|
785 |
+
"loss": 0.5014,
|
786 |
+
"step": 1250
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 4.0,
|
790 |
+
"eval_accuracy": 0.7321109123434705,
|
791 |
+
"eval_loss": 0.45054975152015686,
|
792 |
+
"eval_runtime": 31.9873,
|
793 |
+
"eval_samples_per_second": 139.805,
|
794 |
+
"eval_steps_per_second": 4.377,
|
795 |
+
"step": 1258
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 4.01,
|
799 |
+
"learning_rate": 1.0969568294409059e-05,
|
800 |
+
"loss": 0.5164,
|
801 |
+
"step": 1260
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 4.04,
|
805 |
+
"learning_rate": 1.0615711252653929e-05,
|
806 |
+
"loss": 0.4968,
|
807 |
+
"step": 1270
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 4.07,
|
811 |
+
"learning_rate": 1.0261854210898796e-05,
|
812 |
+
"loss": 0.4781,
|
813 |
+
"step": 1280
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 4.1,
|
817 |
+
"learning_rate": 9.907997169143666e-06,
|
818 |
+
"loss": 0.4935,
|
819 |
+
"step": 1290
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 4.13,
|
823 |
+
"learning_rate": 9.554140127388536e-06,
|
824 |
+
"loss": 0.455,
|
825 |
+
"step": 1300
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 4.17,
|
829 |
+
"learning_rate": 9.200283085633404e-06,
|
830 |
+
"loss": 0.5263,
|
831 |
+
"step": 1310
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 4.2,
|
835 |
+
"learning_rate": 8.846426043878273e-06,
|
836 |
+
"loss": 0.4888,
|
837 |
+
"step": 1320
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 4.23,
|
841 |
+
"learning_rate": 8.492569002123141e-06,
|
842 |
+
"loss": 0.4853,
|
843 |
+
"step": 1330
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 4.26,
|
847 |
+
"learning_rate": 8.138711960368011e-06,
|
848 |
+
"loss": 0.4744,
|
849 |
+
"step": 1340
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 4.29,
|
853 |
+
"learning_rate": 7.78485491861288e-06,
|
854 |
+
"loss": 0.4537,
|
855 |
+
"step": 1350
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 4.32,
|
859 |
+
"learning_rate": 7.43099787685775e-06,
|
860 |
+
"loss": 0.5045,
|
861 |
+
"step": 1360
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 4.36,
|
865 |
+
"learning_rate": 7.077140835102619e-06,
|
866 |
+
"loss": 0.4707,
|
867 |
+
"step": 1370
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 4.39,
|
871 |
+
"learning_rate": 6.723283793347487e-06,
|
872 |
+
"loss": 0.4776,
|
873 |
+
"step": 1380
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 4.42,
|
877 |
+
"learning_rate": 6.369426751592357e-06,
|
878 |
+
"loss": 0.5119,
|
879 |
+
"step": 1390
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 4.45,
|
883 |
+
"learning_rate": 6.015569709837226e-06,
|
884 |
+
"loss": 0.5521,
|
885 |
+
"step": 1400
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 4.48,
|
889 |
+
"learning_rate": 5.661712668082095e-06,
|
890 |
+
"loss": 0.4871,
|
891 |
+
"step": 1410
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 4.52,
|
895 |
+
"learning_rate": 5.307855626326964e-06,
|
896 |
+
"loss": 0.5006,
|
897 |
+
"step": 1420
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 4.55,
|
901 |
+
"learning_rate": 4.953998584571833e-06,
|
902 |
+
"loss": 0.4733,
|
903 |
+
"step": 1430
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 4.58,
|
907 |
+
"learning_rate": 4.600141542816702e-06,
|
908 |
+
"loss": 0.4942,
|
909 |
+
"step": 1440
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 4.61,
|
913 |
+
"learning_rate": 4.246284501061571e-06,
|
914 |
+
"loss": 0.4806,
|
915 |
+
"step": 1450
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 4.64,
|
919 |
+
"learning_rate": 3.89242745930644e-06,
|
920 |
+
"loss": 0.5122,
|
921 |
+
"step": 1460
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 4.67,
|
925 |
+
"learning_rate": 3.5385704175513096e-06,
|
926 |
+
"loss": 0.4843,
|
927 |
+
"step": 1470
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 4.71,
|
931 |
+
"learning_rate": 3.1847133757961785e-06,
|
932 |
+
"loss": 0.4757,
|
933 |
+
"step": 1480
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 4.74,
|
937 |
+
"learning_rate": 2.8308563340410473e-06,
|
938 |
+
"loss": 0.4717,
|
939 |
+
"step": 1490
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 4.77,
|
943 |
+
"learning_rate": 2.4769992922859165e-06,
|
944 |
+
"loss": 0.5049,
|
945 |
+
"step": 1500
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 4.8,
|
949 |
+
"learning_rate": 2.1231422505307854e-06,
|
950 |
+
"loss": 0.4666,
|
951 |
+
"step": 1510
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 4.83,
|
955 |
+
"learning_rate": 1.7692852087756548e-06,
|
956 |
+
"loss": 0.4781,
|
957 |
+
"step": 1520
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 4.86,
|
961 |
+
"learning_rate": 1.4154281670205236e-06,
|
962 |
+
"loss": 0.485,
|
963 |
+
"step": 1530
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 4.9,
|
967 |
+
"learning_rate": 1.0615711252653927e-06,
|
968 |
+
"loss": 0.5166,
|
969 |
+
"step": 1540
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 4.93,
|
973 |
+
"learning_rate": 7.077140835102618e-07,
|
974 |
+
"loss": 0.5056,
|
975 |
+
"step": 1550
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 4.96,
|
979 |
+
"learning_rate": 3.538570417551309e-07,
|
980 |
+
"loss": 0.4843,
|
981 |
+
"step": 1560
|
982 |
+
},
|
983 |
+
{
|
984 |
+
"epoch": 4.99,
|
985 |
+
"learning_rate": 0.0,
|
986 |
+
"loss": 0.4746,
|
987 |
+
"step": 1570
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 4.99,
|
991 |
+
"eval_accuracy": 0.7294275491949911,
|
992 |
+
"eval_loss": 0.45043760538101196,
|
993 |
+
"eval_runtime": 33.0157,
|
994 |
+
"eval_samples_per_second": 135.451,
|
995 |
+
"eval_steps_per_second": 4.24,
|
996 |
+
"step": 1570
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 4.99,
|
1000 |
+
"step": 1570,
|
1001 |
+
"total_flos": 4.994244969949348e+18,
|
1002 |
+
"train_loss": 0.6026059660941931,
|
1003 |
+
"train_runtime": 3290.1425,
|
1004 |
+
"train_samples_per_second": 61.16,
|
1005 |
+
"train_steps_per_second": 0.477
|
1006 |
}
|
1007 |
],
|
1008 |
"logging_steps": 10,
|
1009 |
+
"max_steps": 1570,
|
1010 |
+
"num_train_epochs": 5,
|
1011 |
"save_steps": 500,
|
1012 |
+
"total_flos": 4.994244969949348e+18,
|
1013 |
"trial_name": null,
|
1014 |
"trial_params": null
|
1015 |
}
|