Augusto777
commited on
Commit
•
8e7c851
1
Parent(s):
33ad953
End of training
Browse files- README.md +2 -2
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Apr19_07-22-40_DESKTOP-SKBE9FB/events.out.tfevents.1713533467.DESKTOP-SKBE9FB.604.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +208 -208
README.md
CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
-
- Accuracy: 0.
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.8386
|
21 |
+
- Accuracy: 0.65
|
22 |
|
23 |
## Model description
|
24 |
|
all_results.json
CHANGED
@@ -1,12 +1,12 @@
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss":
|
5 |
-
"eval_runtime": 1.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 1.
|
8 |
-
"train_loss": 1.
|
9 |
-
"train_runtime":
|
10 |
-
"train_samples_per_second":
|
11 |
-
"train_steps_per_second": 0.
|
12 |
}
|
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
+
"eval_accuracy": 0.65,
|
4 |
+
"eval_loss": 0.8386151194572449,
|
5 |
+
"eval_runtime": 1.0431,
|
6 |
+
"eval_samples_per_second": 57.519,
|
7 |
+
"eval_steps_per_second": 1.917,
|
8 |
+
"train_loss": 1.4638761638784084,
|
9 |
+
"train_runtime": 475.2898,
|
10 |
+
"train_samples_per_second": 86.07,
|
11 |
+
"train_steps_per_second": 0.619
|
12 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss":
|
5 |
-
"eval_runtime": 1.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 1.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
+
"eval_accuracy": 0.65,
|
4 |
+
"eval_loss": 0.8386151194572449,
|
5 |
+
"eval_runtime": 1.0431,
|
6 |
+
"eval_samples_per_second": 57.519,
|
7 |
+
"eval_steps_per_second": 1.917
|
8 |
}
|
runs/Apr19_07-22-40_DESKTOP-SKBE9FB/events.out.tfevents.1713533467.DESKTOP-SKBE9FB.604.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9abb5481ec5302aeeefad24dfe4bac914b2a5a15e4a2d9b332729db181a9738
|
3 |
+
size 411
|
train_results.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
-
"train_loss": 1.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples_per_second":
|
6 |
-
"train_steps_per_second": 0.
|
7 |
}
|
|
|
1 |
{
|
2 |
"epoch": 37.94,
|
3 |
+
"train_loss": 1.4638761638784084,
|
4 |
+
"train_runtime": 475.2898,
|
5 |
+
"train_samples_per_second": 86.07,
|
6 |
+
"train_steps_per_second": 0.619
|
7 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-
|
4 |
"epoch": 37.935483870967744,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 294,
|
@@ -11,497 +11,497 @@
|
|
11 |
{
|
12 |
"epoch": 0.9,
|
13 |
"eval_accuracy": 0.11666666666666667,
|
14 |
-
"eval_loss":
|
15 |
-
"eval_runtime":
|
16 |
-
"eval_samples_per_second":
|
17 |
-
"eval_steps_per_second":
|
18 |
"step": 7
|
19 |
},
|
20 |
{
|
21 |
"epoch": 1.55,
|
22 |
"learning_rate": 1.6000000000000003e-05,
|
23 |
-
"loss":
|
24 |
"step": 12
|
25 |
},
|
26 |
{
|
27 |
"epoch": 1.94,
|
28 |
"eval_accuracy": 0.11666666666666667,
|
29 |
-
"eval_loss":
|
30 |
-
"eval_runtime": 0.
|
31 |
-
"eval_samples_per_second":
|
32 |
-
"eval_steps_per_second": 2.
|
33 |
"step": 15
|
34 |
},
|
35 |
{
|
36 |
"epoch": 2.97,
|
37 |
"eval_accuracy": 0.11666666666666667,
|
38 |
-
"eval_loss":
|
39 |
-
"eval_runtime":
|
40 |
-
"eval_samples_per_second":
|
41 |
-
"eval_steps_per_second":
|
42 |
"step": 23
|
43 |
},
|
44 |
{
|
45 |
"epoch": 3.1,
|
46 |
"learning_rate": 3.2000000000000005e-05,
|
47 |
-
"loss":
|
48 |
"step": 24
|
49 |
},
|
50 |
{
|
51 |
"epoch": 4.0,
|
52 |
"eval_accuracy": 0.11666666666666667,
|
53 |
-
"eval_loss":
|
54 |
-
"eval_runtime":
|
55 |
-
"eval_samples_per_second":
|
56 |
-
"eval_steps_per_second":
|
57 |
"step": 31
|
58 |
},
|
59 |
{
|
60 |
"epoch": 4.65,
|
61 |
"learning_rate": 3.909090909090909e-05,
|
62 |
-
"loss":
|
63 |
"step": 36
|
64 |
},
|
65 |
{
|
66 |
"epoch": 4.9,
|
67 |
"eval_accuracy": 0.11666666666666667,
|
68 |
-
"eval_loss":
|
69 |
-
"eval_runtime": 0.
|
70 |
-
"eval_samples_per_second":
|
71 |
-
"eval_steps_per_second": 2.
|
72 |
"step": 38
|
73 |
},
|
74 |
{
|
75 |
"epoch": 5.94,
|
76 |
"eval_accuracy": 0.11666666666666667,
|
77 |
-
"eval_loss":
|
78 |
-
"eval_runtime":
|
79 |
-
"eval_samples_per_second":
|
80 |
-
"eval_steps_per_second":
|
81 |
"step": 46
|
82 |
},
|
83 |
{
|
84 |
"epoch": 6.19,
|
85 |
"learning_rate": 3.7272727272727276e-05,
|
86 |
-
"loss":
|
87 |
"step": 48
|
88 |
},
|
89 |
{
|
90 |
"epoch": 6.97,
|
91 |
"eval_accuracy": 0.11666666666666667,
|
92 |
-
"eval_loss":
|
93 |
-
"eval_runtime":
|
94 |
-
"eval_samples_per_second":
|
95 |
-
"eval_steps_per_second":
|
96 |
"step": 54
|
97 |
},
|
98 |
{
|
99 |
"epoch": 7.74,
|
100 |
"learning_rate": 3.545454545454546e-05,
|
101 |
-
"loss":
|
102 |
"step": 60
|
103 |
},
|
104 |
{
|
105 |
"epoch": 8.0,
|
106 |
-
"eval_accuracy": 0.
|
107 |
-
"eval_loss": 1.
|
108 |
-
"eval_runtime":
|
109 |
-
"eval_samples_per_second":
|
110 |
-
"eval_steps_per_second":
|
111 |
"step": 62
|
112 |
},
|
113 |
{
|
114 |
"epoch": 8.9,
|
115 |
"eval_accuracy": 0.45,
|
116 |
-
"eval_loss": 1.
|
117 |
-
"eval_runtime":
|
118 |
-
"eval_samples_per_second":
|
119 |
-
"eval_steps_per_second":
|
120 |
"step": 69
|
121 |
},
|
122 |
{
|
123 |
"epoch": 9.29,
|
124 |
"learning_rate": 3.363636363636364e-05,
|
125 |
-
"loss": 1.
|
126 |
"step": 72
|
127 |
},
|
128 |
{
|
129 |
"epoch": 9.94,
|
130 |
-
"eval_accuracy": 0.
|
131 |
-
"eval_loss": 1.
|
132 |
-
"eval_runtime":
|
133 |
-
"eval_samples_per_second":
|
134 |
-
"eval_steps_per_second":
|
135 |
"step": 77
|
136 |
},
|
137 |
{
|
138 |
"epoch": 10.84,
|
139 |
"learning_rate": 3.181818181818182e-05,
|
140 |
-
"loss": 1.
|
141 |
"step": 84
|
142 |
},
|
143 |
{
|
144 |
"epoch": 10.97,
|
145 |
-
"eval_accuracy": 0.
|
146 |
-
"eval_loss": 1.
|
147 |
-
"eval_runtime":
|
148 |
-
"eval_samples_per_second":
|
149 |
-
"eval_steps_per_second":
|
150 |
"step": 85
|
151 |
},
|
152 |
{
|
153 |
"epoch": 12.0,
|
154 |
"eval_accuracy": 0.4666666666666667,
|
155 |
-
"eval_loss": 1.
|
156 |
-
"eval_runtime": 0.
|
157 |
-
"eval_samples_per_second":
|
158 |
-
"eval_steps_per_second": 2.
|
159 |
"step": 93
|
160 |
},
|
161 |
{
|
162 |
"epoch": 12.39,
|
163 |
"learning_rate": 3.0000000000000004e-05,
|
164 |
-
"loss": 1.
|
165 |
"step": 96
|
166 |
},
|
167 |
{
|
168 |
"epoch": 12.9,
|
169 |
-
"eval_accuracy": 0.
|
170 |
-
"eval_loss": 1.
|
171 |
-
"eval_runtime":
|
172 |
-
"eval_samples_per_second":
|
173 |
-
"eval_steps_per_second":
|
174 |
"step": 100
|
175 |
},
|
176 |
{
|
177 |
"epoch": 13.94,
|
178 |
"learning_rate": 2.8181818181818185e-05,
|
179 |
-
"loss": 1.
|
180 |
"step": 108
|
181 |
},
|
182 |
{
|
183 |
"epoch": 13.94,
|
184 |
-
"eval_accuracy": 0.
|
185 |
-
"eval_loss": 1.
|
186 |
-
"eval_runtime": 0.
|
187 |
-
"eval_samples_per_second":
|
188 |
-
"eval_steps_per_second": 2.
|
189 |
"step": 108
|
190 |
},
|
191 |
{
|
192 |
"epoch": 14.97,
|
193 |
-
"eval_accuracy": 0.
|
194 |
-
"eval_loss": 1.
|
195 |
-
"eval_runtime": 0.
|
196 |
-
"eval_samples_per_second":
|
197 |
-
"eval_steps_per_second": 2.
|
198 |
"step": 116
|
199 |
},
|
200 |
{
|
201 |
"epoch": 15.48,
|
202 |
"learning_rate": 2.6363636363636365e-05,
|
203 |
-
"loss": 1.
|
204 |
"step": 120
|
205 |
},
|
206 |
{
|
207 |
"epoch": 16.0,
|
208 |
-
"eval_accuracy": 0.
|
209 |
-
"eval_loss": 1.
|
210 |
-
"eval_runtime":
|
211 |
-
"eval_samples_per_second":
|
212 |
-
"eval_steps_per_second":
|
213 |
"step": 124
|
214 |
},
|
215 |
{
|
216 |
"epoch": 16.9,
|
217 |
"eval_accuracy": 0.5166666666666667,
|
218 |
-
"eval_loss": 1.
|
219 |
-
"eval_runtime": 0.
|
220 |
-
"eval_samples_per_second": 67.
|
221 |
-
"eval_steps_per_second": 2.
|
222 |
"step": 131
|
223 |
},
|
224 |
{
|
225 |
"epoch": 17.03,
|
226 |
"learning_rate": 2.454545454545455e-05,
|
227 |
-
"loss": 1.
|
228 |
"step": 132
|
229 |
},
|
230 |
{
|
231 |
"epoch": 17.94,
|
232 |
-
"eval_accuracy": 0.
|
233 |
-
"eval_loss": 1.
|
234 |
-
"eval_runtime":
|
235 |
-
"eval_samples_per_second":
|
236 |
-
"eval_steps_per_second":
|
237 |
"step": 139
|
238 |
},
|
239 |
{
|
240 |
"epoch": 18.58,
|
241 |
"learning_rate": 2.2727272727272733e-05,
|
242 |
-
"loss": 1.
|
243 |
"step": 144
|
244 |
},
|
245 |
{
|
246 |
"epoch": 18.97,
|
247 |
"eval_accuracy": 0.55,
|
248 |
-
"eval_loss": 1.
|
249 |
-
"eval_runtime":
|
250 |
-
"eval_samples_per_second":
|
251 |
-
"eval_steps_per_second":
|
252 |
"step": 147
|
253 |
},
|
254 |
{
|
255 |
"epoch": 20.0,
|
256 |
-
"eval_accuracy": 0.
|
257 |
-
"eval_loss":
|
258 |
-
"eval_runtime": 0.
|
259 |
-
"eval_samples_per_second":
|
260 |
-
"eval_steps_per_second": 2.
|
261 |
"step": 155
|
262 |
},
|
263 |
{
|
264 |
"epoch": 20.13,
|
265 |
"learning_rate": 2.090909090909091e-05,
|
266 |
-
"loss":
|
267 |
"step": 156
|
268 |
},
|
269 |
{
|
270 |
"epoch": 20.9,
|
271 |
-
"eval_accuracy": 0.
|
272 |
-
"eval_loss":
|
273 |
-
"eval_runtime": 0.
|
274 |
-
"eval_samples_per_second":
|
275 |
-
"eval_steps_per_second": 2.
|
276 |
"step": 162
|
277 |
},
|
278 |
{
|
279 |
"epoch": 21.68,
|
280 |
"learning_rate": 1.9090909090909094e-05,
|
281 |
-
"loss":
|
282 |
"step": 168
|
283 |
},
|
284 |
{
|
285 |
"epoch": 21.94,
|
286 |
"eval_accuracy": 0.5833333333333334,
|
287 |
-
"eval_loss":
|
288 |
-
"eval_runtime": 0.
|
289 |
-
"eval_samples_per_second":
|
290 |
-
"eval_steps_per_second": 2.
|
291 |
"step": 170
|
292 |
},
|
293 |
{
|
294 |
"epoch": 22.97,
|
295 |
-
"eval_accuracy": 0.
|
296 |
-
"eval_loss":
|
297 |
-
"eval_runtime": 0.
|
298 |
-
"eval_samples_per_second":
|
299 |
-
"eval_steps_per_second": 2.
|
300 |
"step": 178
|
301 |
},
|
302 |
{
|
303 |
"epoch": 23.23,
|
304 |
"learning_rate": 1.7272727272727274e-05,
|
305 |
-
"loss":
|
306 |
"step": 180
|
307 |
},
|
308 |
{
|
309 |
"epoch": 24.0,
|
310 |
-
"eval_accuracy": 0.
|
311 |
-
"eval_loss":
|
312 |
-
"eval_runtime": 0.
|
313 |
-
"eval_samples_per_second":
|
314 |
-
"eval_steps_per_second": 2.
|
315 |
"step": 186
|
316 |
},
|
317 |
{
|
318 |
"epoch": 24.77,
|
319 |
"learning_rate": 1.5454545454545454e-05,
|
320 |
-
"loss": 0.
|
321 |
"step": 192
|
322 |
},
|
323 |
{
|
324 |
"epoch": 24.9,
|
325 |
-
"eval_accuracy": 0.
|
326 |
-
"eval_loss":
|
327 |
-
"eval_runtime": 0.
|
328 |
-
"eval_samples_per_second":
|
329 |
-
"eval_steps_per_second": 2.
|
330 |
"step": 193
|
331 |
},
|
332 |
{
|
333 |
"epoch": 25.94,
|
334 |
-
"eval_accuracy": 0.
|
335 |
-
"eval_loss":
|
336 |
-
"eval_runtime":
|
337 |
-
"eval_samples_per_second":
|
338 |
-
"eval_steps_per_second":
|
339 |
"step": 201
|
340 |
},
|
341 |
{
|
342 |
"epoch": 26.32,
|
343 |
"learning_rate": 1.3636363636363637e-05,
|
344 |
-
"loss": 0.
|
345 |
"step": 204
|
346 |
},
|
347 |
{
|
348 |
"epoch": 26.97,
|
349 |
-
"eval_accuracy": 0.
|
350 |
-
"eval_loss": 0.
|
351 |
-
"eval_runtime": 0.
|
352 |
-
"eval_samples_per_second":
|
353 |
-
"eval_steps_per_second": 2.
|
354 |
"step": 209
|
355 |
},
|
356 |
{
|
357 |
"epoch": 27.87,
|
358 |
"learning_rate": 1.181818181818182e-05,
|
359 |
-
"loss": 0.
|
360 |
"step": 216
|
361 |
},
|
362 |
{
|
363 |
"epoch": 28.0,
|
364 |
-
"eval_accuracy": 0.
|
365 |
-
"eval_loss": 0.
|
366 |
-
"eval_runtime": 0.
|
367 |
-
"eval_samples_per_second":
|
368 |
-
"eval_steps_per_second": 2.
|
369 |
"step": 217
|
370 |
},
|
371 |
{
|
372 |
"epoch": 28.9,
|
373 |
-
"eval_accuracy": 0.
|
374 |
-
"eval_loss": 0.
|
375 |
-
"eval_runtime": 0.
|
376 |
-
"eval_samples_per_second":
|
377 |
-
"eval_steps_per_second": 2.
|
378 |
"step": 224
|
379 |
},
|
380 |
{
|
381 |
"epoch": 29.42,
|
382 |
"learning_rate": 1e-05,
|
383 |
-
"loss": 0.
|
384 |
"step": 228
|
385 |
},
|
386 |
{
|
387 |
"epoch": 29.94,
|
388 |
-
"eval_accuracy": 0.
|
389 |
-
"eval_loss": 0.
|
390 |
-
"eval_runtime": 0.
|
391 |
-
"eval_samples_per_second":
|
392 |
-
"eval_steps_per_second": 2.
|
393 |
"step": 232
|
394 |
},
|
395 |
{
|
396 |
"epoch": 30.97,
|
397 |
"learning_rate": 8.181818181818183e-06,
|
398 |
-
"loss": 0.
|
399 |
"step": 240
|
400 |
},
|
401 |
{
|
402 |
"epoch": 30.97,
|
403 |
-
"eval_accuracy": 0.
|
404 |
-
"eval_loss": 0.
|
405 |
-
"eval_runtime": 0.
|
406 |
-
"eval_samples_per_second": 68.
|
407 |
-
"eval_steps_per_second": 2.
|
408 |
"step": 240
|
409 |
},
|
410 |
{
|
411 |
"epoch": 32.0,
|
412 |
-
"eval_accuracy": 0.
|
413 |
-
"eval_loss": 0.
|
414 |
-
"eval_runtime": 0.
|
415 |
-
"eval_samples_per_second":
|
416 |
-
"eval_steps_per_second": 2.
|
417 |
"step": 248
|
418 |
},
|
419 |
{
|
420 |
"epoch": 32.52,
|
421 |
"learning_rate": 6.363636363636364e-06,
|
422 |
-
"loss": 0.
|
423 |
"step": 252
|
424 |
},
|
425 |
{
|
426 |
"epoch": 32.9,
|
427 |
-
"eval_accuracy": 0.
|
428 |
-
"eval_loss": 0.
|
429 |
-
"eval_runtime": 0.
|
430 |
-
"eval_samples_per_second":
|
431 |
-
"eval_steps_per_second": 2.
|
432 |
"step": 255
|
433 |
},
|
434 |
{
|
435 |
"epoch": 33.94,
|
436 |
-
"eval_accuracy": 0.
|
437 |
-
"eval_loss": 0.
|
438 |
-
"eval_runtime": 0.
|
439 |
-
"eval_samples_per_second":
|
440 |
-
"eval_steps_per_second": 2.
|
441 |
"step": 263
|
442 |
},
|
443 |
{
|
444 |
"epoch": 34.06,
|
445 |
"learning_rate": 4.5454545454545455e-06,
|
446 |
-
"loss": 0.
|
447 |
"step": 264
|
448 |
},
|
449 |
{
|
450 |
"epoch": 34.97,
|
451 |
-
"eval_accuracy": 0.
|
452 |
-
"eval_loss": 0.
|
453 |
-
"eval_runtime":
|
454 |
-
"eval_samples_per_second":
|
455 |
-
"eval_steps_per_second":
|
456 |
"step": 271
|
457 |
},
|
458 |
{
|
459 |
"epoch": 35.61,
|
460 |
"learning_rate": 2.7272727272727272e-06,
|
461 |
-
"loss": 0.
|
462 |
"step": 276
|
463 |
},
|
464 |
{
|
465 |
"epoch": 36.0,
|
466 |
-
"eval_accuracy": 0.
|
467 |
-
"eval_loss": 0.
|
468 |
-
"eval_runtime": 0.
|
469 |
-
"eval_samples_per_second":
|
470 |
-
"eval_steps_per_second": 2.
|
471 |
"step": 279
|
472 |
},
|
473 |
{
|
474 |
"epoch": 36.9,
|
475 |
-
"eval_accuracy": 0.
|
476 |
-
"eval_loss": 0.
|
477 |
-
"eval_runtime":
|
478 |
-
"eval_samples_per_second":
|
479 |
-
"eval_steps_per_second":
|
480 |
"step": 286
|
481 |
},
|
482 |
{
|
483 |
"epoch": 37.16,
|
484 |
"learning_rate": 9.090909090909091e-07,
|
485 |
-
"loss": 0.
|
486 |
"step": 288
|
487 |
},
|
488 |
{
|
489 |
"epoch": 37.94,
|
490 |
-
"eval_accuracy": 0.
|
491 |
-
"eval_loss": 0.
|
492 |
-
"eval_runtime": 0.
|
493 |
-
"eval_samples_per_second":
|
494 |
-
"eval_steps_per_second": 2.
|
495 |
"step": 294
|
496 |
},
|
497 |
{
|
498 |
"epoch": 37.94,
|
499 |
"step": 294,
|
500 |
"total_flos": 1.2027310550050406e+18,
|
501 |
-
"train_loss": 1.
|
502 |
-
"train_runtime":
|
503 |
-
"train_samples_per_second":
|
504 |
-
"train_steps_per_second": 0.
|
505 |
}
|
506 |
],
|
507 |
"logging_steps": 12,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.65,
|
3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-209",
|
4 |
"epoch": 37.935483870967744,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 294,
|
|
|
11 |
{
|
12 |
"epoch": 0.9,
|
13 |
"eval_accuracy": 0.11666666666666667,
|
14 |
+
"eval_loss": 6.174824237823486,
|
15 |
+
"eval_runtime": 0.845,
|
16 |
+
"eval_samples_per_second": 71.01,
|
17 |
+
"eval_steps_per_second": 2.367,
|
18 |
"step": 7
|
19 |
},
|
20 |
{
|
21 |
"epoch": 1.55,
|
22 |
"learning_rate": 1.6000000000000003e-05,
|
23 |
+
"loss": 5.327,
|
24 |
"step": 12
|
25 |
},
|
26 |
{
|
27 |
"epoch": 1.94,
|
28 |
"eval_accuracy": 0.11666666666666667,
|
29 |
+
"eval_loss": 6.066026210784912,
|
30 |
+
"eval_runtime": 0.8511,
|
31 |
+
"eval_samples_per_second": 70.499,
|
32 |
+
"eval_steps_per_second": 2.35,
|
33 |
"step": 15
|
34 |
},
|
35 |
{
|
36 |
"epoch": 2.97,
|
37 |
"eval_accuracy": 0.11666666666666667,
|
38 |
+
"eval_loss": 5.4902424812316895,
|
39 |
+
"eval_runtime": 0.8648,
|
40 |
+
"eval_samples_per_second": 69.381,
|
41 |
+
"eval_steps_per_second": 2.313,
|
42 |
"step": 23
|
43 |
},
|
44 |
{
|
45 |
"epoch": 3.1,
|
46 |
"learning_rate": 3.2000000000000005e-05,
|
47 |
+
"loss": 5.0963,
|
48 |
"step": 24
|
49 |
},
|
50 |
{
|
51 |
"epoch": 4.0,
|
52 |
"eval_accuracy": 0.11666666666666667,
|
53 |
+
"eval_loss": 4.276790142059326,
|
54 |
+
"eval_runtime": 0.8095,
|
55 |
+
"eval_samples_per_second": 74.116,
|
56 |
+
"eval_steps_per_second": 2.471,
|
57 |
"step": 31
|
58 |
},
|
59 |
{
|
60 |
"epoch": 4.65,
|
61 |
"learning_rate": 3.909090909090909e-05,
|
62 |
+
"loss": 3.9193,
|
63 |
"step": 36
|
64 |
},
|
65 |
{
|
66 |
"epoch": 4.9,
|
67 |
"eval_accuracy": 0.11666666666666667,
|
68 |
+
"eval_loss": 3.0012993812561035,
|
69 |
+
"eval_runtime": 0.8424,
|
70 |
+
"eval_samples_per_second": 71.228,
|
71 |
+
"eval_steps_per_second": 2.374,
|
72 |
"step": 38
|
73 |
},
|
74 |
{
|
75 |
"epoch": 5.94,
|
76 |
"eval_accuracy": 0.11666666666666667,
|
77 |
+
"eval_loss": 1.9289205074310303,
|
78 |
+
"eval_runtime": 0.8816,
|
79 |
+
"eval_samples_per_second": 68.055,
|
80 |
+
"eval_steps_per_second": 2.269,
|
81 |
"step": 46
|
82 |
},
|
83 |
{
|
84 |
"epoch": 6.19,
|
85 |
"learning_rate": 3.7272727272727276e-05,
|
86 |
+
"loss": 2.2222,
|
87 |
"step": 48
|
88 |
},
|
89 |
{
|
90 |
"epoch": 6.97,
|
91 |
"eval_accuracy": 0.11666666666666667,
|
92 |
+
"eval_loss": 1.385749101638794,
|
93 |
+
"eval_runtime": 0.8792,
|
94 |
+
"eval_samples_per_second": 68.244,
|
95 |
+
"eval_steps_per_second": 2.275,
|
96 |
"step": 54
|
97 |
},
|
98 |
{
|
99 |
"epoch": 7.74,
|
100 |
"learning_rate": 3.545454545454546e-05,
|
101 |
+
"loss": 1.4465,
|
102 |
"step": 60
|
103 |
},
|
104 |
{
|
105 |
"epoch": 8.0,
|
106 |
+
"eval_accuracy": 0.43333333333333335,
|
107 |
+
"eval_loss": 1.342329740524292,
|
108 |
+
"eval_runtime": 0.8314,
|
109 |
+
"eval_samples_per_second": 72.165,
|
110 |
+
"eval_steps_per_second": 2.405,
|
111 |
"step": 62
|
112 |
},
|
113 |
{
|
114 |
"epoch": 8.9,
|
115 |
"eval_accuracy": 0.45,
|
116 |
+
"eval_loss": 1.2786400318145752,
|
117 |
+
"eval_runtime": 0.8842,
|
118 |
+
"eval_samples_per_second": 67.857,
|
119 |
+
"eval_steps_per_second": 2.262,
|
120 |
"step": 69
|
121 |
},
|
122 |
{
|
123 |
"epoch": 9.29,
|
124 |
"learning_rate": 3.363636363636364e-05,
|
125 |
+
"loss": 1.3709,
|
126 |
"step": 72
|
127 |
},
|
128 |
{
|
129 |
"epoch": 9.94,
|
130 |
+
"eval_accuracy": 0.4666666666666667,
|
131 |
+
"eval_loss": 1.2653944492340088,
|
132 |
+
"eval_runtime": 0.8365,
|
133 |
+
"eval_samples_per_second": 71.724,
|
134 |
+
"eval_steps_per_second": 2.391,
|
135 |
"step": 77
|
136 |
},
|
137 |
{
|
138 |
"epoch": 10.84,
|
139 |
"learning_rate": 3.181818181818182e-05,
|
140 |
+
"loss": 1.3511,
|
141 |
"step": 84
|
142 |
},
|
143 |
{
|
144 |
"epoch": 10.97,
|
145 |
+
"eval_accuracy": 0.4666666666666667,
|
146 |
+
"eval_loss": 1.2605175971984863,
|
147 |
+
"eval_runtime": 0.8152,
|
148 |
+
"eval_samples_per_second": 73.599,
|
149 |
+
"eval_steps_per_second": 2.453,
|
150 |
"step": 85
|
151 |
},
|
152 |
{
|
153 |
"epoch": 12.0,
|
154 |
"eval_accuracy": 0.4666666666666667,
|
155 |
+
"eval_loss": 1.2183691263198853,
|
156 |
+
"eval_runtime": 0.8329,
|
157 |
+
"eval_samples_per_second": 72.04,
|
158 |
+
"eval_steps_per_second": 2.401,
|
159 |
"step": 93
|
160 |
},
|
161 |
{
|
162 |
"epoch": 12.39,
|
163 |
"learning_rate": 3.0000000000000004e-05,
|
164 |
+
"loss": 1.2749,
|
165 |
"step": 96
|
166 |
},
|
167 |
{
|
168 |
"epoch": 12.9,
|
169 |
+
"eval_accuracy": 0.5,
|
170 |
+
"eval_loss": 1.289415955543518,
|
171 |
+
"eval_runtime": 0.8348,
|
172 |
+
"eval_samples_per_second": 71.87,
|
173 |
+
"eval_steps_per_second": 2.396,
|
174 |
"step": 100
|
175 |
},
|
176 |
{
|
177 |
"epoch": 13.94,
|
178 |
"learning_rate": 2.8181818181818185e-05,
|
179 |
+
"loss": 1.222,
|
180 |
"step": 108
|
181 |
},
|
182 |
{
|
183 |
"epoch": 13.94,
|
184 |
+
"eval_accuracy": 0.5166666666666667,
|
185 |
+
"eval_loss": 1.2071518898010254,
|
186 |
+
"eval_runtime": 0.8805,
|
187 |
+
"eval_samples_per_second": 68.142,
|
188 |
+
"eval_steps_per_second": 2.271,
|
189 |
"step": 108
|
190 |
},
|
191 |
{
|
192 |
"epoch": 14.97,
|
193 |
+
"eval_accuracy": 0.5166666666666667,
|
194 |
+
"eval_loss": 1.1748946905136108,
|
195 |
+
"eval_runtime": 0.8381,
|
196 |
+
"eval_samples_per_second": 71.592,
|
197 |
+
"eval_steps_per_second": 2.386,
|
198 |
"step": 116
|
199 |
},
|
200 |
{
|
201 |
"epoch": 15.48,
|
202 |
"learning_rate": 2.6363636363636365e-05,
|
203 |
+
"loss": 1.1668,
|
204 |
"step": 120
|
205 |
},
|
206 |
{
|
207 |
"epoch": 16.0,
|
208 |
+
"eval_accuracy": 0.5166666666666667,
|
209 |
+
"eval_loss": 1.198820948600769,
|
210 |
+
"eval_runtime": 0.8663,
|
211 |
+
"eval_samples_per_second": 69.26,
|
212 |
+
"eval_steps_per_second": 2.309,
|
213 |
"step": 124
|
214 |
},
|
215 |
{
|
216 |
"epoch": 16.9,
|
217 |
"eval_accuracy": 0.5166666666666667,
|
218 |
+
"eval_loss": 1.2305818796157837,
|
219 |
+
"eval_runtime": 0.8839,
|
220 |
+
"eval_samples_per_second": 67.884,
|
221 |
+
"eval_steps_per_second": 2.263,
|
222 |
"step": 131
|
223 |
},
|
224 |
{
|
225 |
"epoch": 17.03,
|
226 |
"learning_rate": 2.454545454545455e-05,
|
227 |
+
"loss": 1.101,
|
228 |
"step": 132
|
229 |
},
|
230 |
{
|
231 |
"epoch": 17.94,
|
232 |
+
"eval_accuracy": 0.5333333333333333,
|
233 |
+
"eval_loss": 1.1431663036346436,
|
234 |
+
"eval_runtime": 0.8778,
|
235 |
+
"eval_samples_per_second": 68.351,
|
236 |
+
"eval_steps_per_second": 2.278,
|
237 |
"step": 139
|
238 |
},
|
239 |
{
|
240 |
"epoch": 18.58,
|
241 |
"learning_rate": 2.2727272727272733e-05,
|
242 |
+
"loss": 1.029,
|
243 |
"step": 144
|
244 |
},
|
245 |
{
|
246 |
"epoch": 18.97,
|
247 |
"eval_accuracy": 0.55,
|
248 |
+
"eval_loss": 1.0208261013031006,
|
249 |
+
"eval_runtime": 0.8723,
|
250 |
+
"eval_samples_per_second": 68.786,
|
251 |
+
"eval_steps_per_second": 2.293,
|
252 |
"step": 147
|
253 |
},
|
254 |
{
|
255 |
"epoch": 20.0,
|
256 |
+
"eval_accuracy": 0.6166666666666667,
|
257 |
+
"eval_loss": 0.9576696157455444,
|
258 |
+
"eval_runtime": 0.849,
|
259 |
+
"eval_samples_per_second": 70.671,
|
260 |
+
"eval_steps_per_second": 2.356,
|
261 |
"step": 155
|
262 |
},
|
263 |
{
|
264 |
"epoch": 20.13,
|
265 |
"learning_rate": 2.090909090909091e-05,
|
266 |
+
"loss": 0.9403,
|
267 |
"step": 156
|
268 |
},
|
269 |
{
|
270 |
"epoch": 20.9,
|
271 |
+
"eval_accuracy": 0.5,
|
272 |
+
"eval_loss": 0.9478870034217834,
|
273 |
+
"eval_runtime": 0.8337,
|
274 |
+
"eval_samples_per_second": 71.971,
|
275 |
+
"eval_steps_per_second": 2.399,
|
276 |
"step": 162
|
277 |
},
|
278 |
{
|
279 |
"epoch": 21.68,
|
280 |
"learning_rate": 1.9090909090909094e-05,
|
281 |
+
"loss": 0.8887,
|
282 |
"step": 168
|
283 |
},
|
284 |
{
|
285 |
"epoch": 21.94,
|
286 |
"eval_accuracy": 0.5833333333333334,
|
287 |
+
"eval_loss": 0.8909610509872437,
|
288 |
+
"eval_runtime": 0.8389,
|
289 |
+
"eval_samples_per_second": 71.521,
|
290 |
+
"eval_steps_per_second": 2.384,
|
291 |
"step": 170
|
292 |
},
|
293 |
{
|
294 |
"epoch": 22.97,
|
295 |
+
"eval_accuracy": 0.5333333333333333,
|
296 |
+
"eval_loss": 0.9441585540771484,
|
297 |
+
"eval_runtime": 0.8343,
|
298 |
+
"eval_samples_per_second": 71.914,
|
299 |
+
"eval_steps_per_second": 2.397,
|
300 |
"step": 178
|
301 |
},
|
302 |
{
|
303 |
"epoch": 23.23,
|
304 |
"learning_rate": 1.7272727272727274e-05,
|
305 |
+
"loss": 0.8506,
|
306 |
"step": 180
|
307 |
},
|
308 |
{
|
309 |
"epoch": 24.0,
|
310 |
+
"eval_accuracy": 0.6,
|
311 |
+
"eval_loss": 0.8923367261886597,
|
312 |
+
"eval_runtime": 0.8341,
|
313 |
+
"eval_samples_per_second": 71.937,
|
314 |
+
"eval_steps_per_second": 2.398,
|
315 |
"step": 186
|
316 |
},
|
317 |
{
|
318 |
"epoch": 24.77,
|
319 |
"learning_rate": 1.5454545454545454e-05,
|
320 |
+
"loss": 0.8064,
|
321 |
"step": 192
|
322 |
},
|
323 |
{
|
324 |
"epoch": 24.9,
|
325 |
+
"eval_accuracy": 0.6,
|
326 |
+
"eval_loss": 0.8972867131233215,
|
327 |
+
"eval_runtime": 0.8817,
|
328 |
+
"eval_samples_per_second": 68.052,
|
329 |
+
"eval_steps_per_second": 2.268,
|
330 |
"step": 193
|
331 |
},
|
332 |
{
|
333 |
"epoch": 25.94,
|
334 |
+
"eval_accuracy": 0.55,
|
335 |
+
"eval_loss": 0.907895028591156,
|
336 |
+
"eval_runtime": 0.87,
|
337 |
+
"eval_samples_per_second": 68.962,
|
338 |
+
"eval_steps_per_second": 2.299,
|
339 |
"step": 201
|
340 |
},
|
341 |
{
|
342 |
"epoch": 26.32,
|
343 |
"learning_rate": 1.3636363636363637e-05,
|
344 |
+
"loss": 0.7434,
|
345 |
"step": 204
|
346 |
},
|
347 |
{
|
348 |
"epoch": 26.97,
|
349 |
+
"eval_accuracy": 0.65,
|
350 |
+
"eval_loss": 0.8386151194572449,
|
351 |
+
"eval_runtime": 0.9123,
|
352 |
+
"eval_samples_per_second": 65.768,
|
353 |
+
"eval_steps_per_second": 2.192,
|
354 |
"step": 209
|
355 |
},
|
356 |
{
|
357 |
"epoch": 27.87,
|
358 |
"learning_rate": 1.181818181818182e-05,
|
359 |
+
"loss": 0.7404,
|
360 |
"step": 216
|
361 |
},
|
362 |
{
|
363 |
"epoch": 28.0,
|
364 |
+
"eval_accuracy": 0.6166666666666667,
|
365 |
+
"eval_loss": 0.8644894361495972,
|
366 |
+
"eval_runtime": 0.8305,
|
367 |
+
"eval_samples_per_second": 72.242,
|
368 |
+
"eval_steps_per_second": 2.408,
|
369 |
"step": 217
|
370 |
},
|
371 |
{
|
372 |
"epoch": 28.9,
|
373 |
+
"eval_accuracy": 0.5666666666666667,
|
374 |
+
"eval_loss": 0.8599310517311096,
|
375 |
+
"eval_runtime": 0.9003,
|
376 |
+
"eval_samples_per_second": 66.644,
|
377 |
+
"eval_steps_per_second": 2.221,
|
378 |
"step": 224
|
379 |
},
|
380 |
{
|
381 |
"epoch": 29.42,
|
382 |
"learning_rate": 1e-05,
|
383 |
+
"loss": 0.7215,
|
384 |
"step": 228
|
385 |
},
|
386 |
{
|
387 |
"epoch": 29.94,
|
388 |
+
"eval_accuracy": 0.65,
|
389 |
+
"eval_loss": 0.8420030474662781,
|
390 |
+
"eval_runtime": 0.8701,
|
391 |
+
"eval_samples_per_second": 68.954,
|
392 |
+
"eval_steps_per_second": 2.298,
|
393 |
"step": 232
|
394 |
},
|
395 |
{
|
396 |
"epoch": 30.97,
|
397 |
"learning_rate": 8.181818181818183e-06,
|
398 |
+
"loss": 0.6743,
|
399 |
"step": 240
|
400 |
},
|
401 |
{
|
402 |
"epoch": 30.97,
|
403 |
+
"eval_accuracy": 0.5666666666666667,
|
404 |
+
"eval_loss": 0.855254590511322,
|
405 |
+
"eval_runtime": 0.8716,
|
406 |
+
"eval_samples_per_second": 68.841,
|
407 |
+
"eval_steps_per_second": 2.295,
|
408 |
"step": 240
|
409 |
},
|
410 |
{
|
411 |
"epoch": 32.0,
|
412 |
+
"eval_accuracy": 0.6166666666666667,
|
413 |
+
"eval_loss": 0.8355186581611633,
|
414 |
+
"eval_runtime": 0.8931,
|
415 |
+
"eval_samples_per_second": 67.182,
|
416 |
+
"eval_steps_per_second": 2.239,
|
417 |
"step": 248
|
418 |
},
|
419 |
{
|
420 |
"epoch": 32.52,
|
421 |
"learning_rate": 6.363636363636364e-06,
|
422 |
+
"loss": 0.6767,
|
423 |
"step": 252
|
424 |
},
|
425 |
{
|
426 |
"epoch": 32.9,
|
427 |
+
"eval_accuracy": 0.5833333333333334,
|
428 |
+
"eval_loss": 0.8694211840629578,
|
429 |
+
"eval_runtime": 0.87,
|
430 |
+
"eval_samples_per_second": 68.966,
|
431 |
+
"eval_steps_per_second": 2.299,
|
432 |
"step": 255
|
433 |
},
|
434 |
{
|
435 |
"epoch": 33.94,
|
436 |
+
"eval_accuracy": 0.65,
|
437 |
+
"eval_loss": 0.8559350371360779,
|
438 |
+
"eval_runtime": 0.8505,
|
439 |
+
"eval_samples_per_second": 70.545,
|
440 |
+
"eval_steps_per_second": 2.351,
|
441 |
"step": 263
|
442 |
},
|
443 |
{
|
444 |
"epoch": 34.06,
|
445 |
"learning_rate": 4.5454545454545455e-06,
|
446 |
+
"loss": 0.6606,
|
447 |
"step": 264
|
448 |
},
|
449 |
{
|
450 |
"epoch": 34.97,
|
451 |
+
"eval_accuracy": 0.6166666666666667,
|
452 |
+
"eval_loss": 0.8350656032562256,
|
453 |
+
"eval_runtime": 0.8389,
|
454 |
+
"eval_samples_per_second": 71.519,
|
455 |
+
"eval_steps_per_second": 2.384,
|
456 |
"step": 271
|
457 |
},
|
458 |
{
|
459 |
"epoch": 35.61,
|
460 |
"learning_rate": 2.7272727272727272e-06,
|
461 |
+
"loss": 0.6488,
|
462 |
"step": 276
|
463 |
},
|
464 |
{
|
465 |
"epoch": 36.0,
|
466 |
+
"eval_accuracy": 0.6333333333333333,
|
467 |
+
"eval_loss": 0.8287425637245178,
|
468 |
+
"eval_runtime": 0.8966,
|
469 |
+
"eval_samples_per_second": 66.919,
|
470 |
+
"eval_steps_per_second": 2.231,
|
471 |
"step": 279
|
472 |
},
|
473 |
{
|
474 |
"epoch": 36.9,
|
475 |
+
"eval_accuracy": 0.6166666666666667,
|
476 |
+
"eval_loss": 0.8377331495285034,
|
477 |
+
"eval_runtime": 0.8452,
|
478 |
+
"eval_samples_per_second": 70.987,
|
479 |
+
"eval_steps_per_second": 2.366,
|
480 |
"step": 286
|
481 |
},
|
482 |
{
|
483 |
"epoch": 37.16,
|
484 |
"learning_rate": 9.090909090909091e-07,
|
485 |
+
"loss": 0.6544,
|
486 |
"step": 288
|
487 |
},
|
488 |
{
|
489 |
"epoch": 37.94,
|
490 |
+
"eval_accuracy": 0.6,
|
491 |
+
"eval_loss": 0.8406462073326111,
|
492 |
+
"eval_runtime": 0.8978,
|
493 |
+
"eval_samples_per_second": 66.829,
|
494 |
+
"eval_steps_per_second": 2.228,
|
495 |
"step": 294
|
496 |
},
|
497 |
{
|
498 |
"epoch": 37.94,
|
499 |
"step": 294,
|
500 |
"total_flos": 1.2027310550050406e+18,
|
501 |
+
"train_loss": 1.4638761638784084,
|
502 |
+
"train_runtime": 475.2898,
|
503 |
+
"train_samples_per_second": 86.07,
|
504 |
+
"train_steps_per_second": 0.619
|
505 |
}
|
506 |
],
|
507 |
"logging_steps": 12,
|