polejowska
commited on
Commit
•
ba1f39a
1
Parent(s):
eaaba4b
End of training
Browse files- trainer_state.json +224 -14
trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"best_metric": 1.
|
3 |
-
"best_model_checkpoint": "detr-r50-cd45rb-8ah-6l\\checkpoint-
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -148,18 +148,228 @@
|
|
148 |
"step": 46060
|
149 |
},
|
150 |
{
|
151 |
-
"epoch":
|
152 |
-
"
|
153 |
-
"
|
154 |
-
"
|
155 |
-
|
156 |
-
|
157 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
}
|
159 |
],
|
160 |
-
"max_steps":
|
161 |
-
"num_train_epochs":
|
162 |
-
"total_flos":
|
163 |
"trial_name": null,
|
164 |
"trial_params": null
|
165 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 1.5793741941452026,
|
3 |
+
"best_model_checkpoint": "detr-r50-cd45rb-8ah-6l\\checkpoint-115150",
|
4 |
+
"epoch": 25.0,
|
5 |
+
"global_step": 115150,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
148 |
"step": 46060
|
149 |
},
|
150 |
{
|
151 |
+
"epoch": 11.0,
|
152 |
+
"learning_rate": 5.6022579244463745e-06,
|
153 |
+
"loss": 2.0948,
|
154 |
+
"step": 50666
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 11.0,
|
158 |
+
"eval_loss": 1.65455162525177,
|
159 |
+
"eval_runtime": 203.3509,
|
160 |
+
"eval_samples_per_second": 8.758,
|
161 |
+
"eval_steps_per_second": 1.097,
|
162 |
+
"step": 50666
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 12.0,
|
166 |
+
"learning_rate": 5.202518454190188e-06,
|
167 |
+
"loss": 2.0943,
|
168 |
+
"step": 55272
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 12.0,
|
172 |
+
"eval_loss": 1.6905332803726196,
|
173 |
+
"eval_runtime": 203.6074,
|
174 |
+
"eval_samples_per_second": 8.747,
|
175 |
+
"eval_steps_per_second": 1.095,
|
176 |
+
"step": 55272
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 13.0,
|
180 |
+
"learning_rate": 4.8026921406860624e-06,
|
181 |
+
"loss": 2.0819,
|
182 |
+
"step": 59878
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 13.0,
|
186 |
+
"eval_loss": 1.6430035829544067,
|
187 |
+
"eval_runtime": 203.499,
|
188 |
+
"eval_samples_per_second": 8.752,
|
189 |
+
"eval_steps_per_second": 1.096,
|
190 |
+
"step": 59878
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 14.0,
|
194 |
+
"learning_rate": 4.402865827181937e-06,
|
195 |
+
"loss": 2.0795,
|
196 |
+
"step": 64484
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 14.0,
|
200 |
+
"eval_loss": 1.6439093351364136,
|
201 |
+
"eval_runtime": 203.71,
|
202 |
+
"eval_samples_per_second": 8.743,
|
203 |
+
"eval_steps_per_second": 1.095,
|
204 |
+
"step": 64484
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 15.0,
|
208 |
+
"learning_rate": 4.002952670429874e-06,
|
209 |
+
"loss": 2.0566,
|
210 |
+
"step": 69090
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 15.0,
|
214 |
+
"eval_loss": 1.6448781490325928,
|
215 |
+
"eval_runtime": 203.5479,
|
216 |
+
"eval_samples_per_second": 8.75,
|
217 |
+
"eval_steps_per_second": 1.096,
|
218 |
+
"step": 69090
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 16.0,
|
222 |
+
"learning_rate": 3.603126356925749e-06,
|
223 |
+
"loss": 2.0435,
|
224 |
+
"step": 73696
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 16.0,
|
228 |
+
"eval_loss": 1.6203994750976562,
|
229 |
+
"eval_runtime": 203.4545,
|
230 |
+
"eval_samples_per_second": 8.754,
|
231 |
+
"eval_steps_per_second": 1.096,
|
232 |
+
"step": 73696
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 17.0,
|
236 |
+
"learning_rate": 3.2033000434216244e-06,
|
237 |
+
"loss": 2.0375,
|
238 |
+
"step": 78302
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 17.0,
|
242 |
+
"eval_loss": 1.6195158958435059,
|
243 |
+
"eval_runtime": 202.3983,
|
244 |
+
"eval_samples_per_second": 8.799,
|
245 |
+
"eval_steps_per_second": 1.102,
|
246 |
+
"step": 78302
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 18.0,
|
250 |
+
"learning_rate": 2.803473729917499e-06,
|
251 |
+
"loss": 2.032,
|
252 |
+
"step": 82908
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 18.0,
|
256 |
+
"eval_loss": 1.6127674579620361,
|
257 |
+
"eval_runtime": 206.9631,
|
258 |
+
"eval_samples_per_second": 8.605,
|
259 |
+
"eval_steps_per_second": 1.077,
|
260 |
+
"step": 82908
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 19.0,
|
264 |
+
"learning_rate": 2.4035605731654367e-06,
|
265 |
+
"loss": 2.0079,
|
266 |
+
"step": 87514
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 19.0,
|
270 |
+
"eval_loss": 1.608217716217041,
|
271 |
+
"eval_runtime": 202.2977,
|
272 |
+
"eval_samples_per_second": 8.804,
|
273 |
+
"eval_steps_per_second": 1.102,
|
274 |
+
"step": 87514
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 20.0,
|
278 |
+
"learning_rate": 2.0037342596613115e-06,
|
279 |
+
"loss": 1.9985,
|
280 |
+
"step": 92120
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 20.0,
|
284 |
+
"eval_loss": 1.6037356853485107,
|
285 |
+
"eval_runtime": 202.9099,
|
286 |
+
"eval_samples_per_second": 8.777,
|
287 |
+
"eval_steps_per_second": 1.099,
|
288 |
+
"step": 92120
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 21.0,
|
292 |
+
"learning_rate": 1.6039947894051239e-06,
|
293 |
+
"loss": 1.9976,
|
294 |
+
"step": 96726
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 21.0,
|
298 |
+
"eval_loss": 1.6005302667617798,
|
299 |
+
"eval_runtime": 202.3563,
|
300 |
+
"eval_samples_per_second": 8.801,
|
301 |
+
"eval_steps_per_second": 1.102,
|
302 |
+
"step": 96726
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 22.0,
|
306 |
+
"learning_rate": 1.2040816326530612e-06,
|
307 |
+
"loss": 1.9887,
|
308 |
+
"step": 101332
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 22.0,
|
312 |
+
"eval_loss": 1.5969282388687134,
|
313 |
+
"eval_runtime": 202.3053,
|
314 |
+
"eval_samples_per_second": 8.804,
|
315 |
+
"eval_steps_per_second": 1.102,
|
316 |
+
"step": 101332
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 23.0,
|
320 |
+
"learning_rate": 8.042553191489362e-07,
|
321 |
+
"loss": 1.9841,
|
322 |
+
"step": 105938
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 23.0,
|
326 |
+
"eval_loss": 1.5841056108474731,
|
327 |
+
"eval_runtime": 204.4797,
|
328 |
+
"eval_samples_per_second": 8.71,
|
329 |
+
"eval_steps_per_second": 1.091,
|
330 |
+
"step": 105938
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 24.0,
|
334 |
+
"learning_rate": 4.043421623968737e-07,
|
335 |
+
"loss": 1.9763,
|
336 |
+
"step": 110544
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 24.0,
|
340 |
+
"eval_loss": 1.582602620124817,
|
341 |
+
"eval_runtime": 204.7835,
|
342 |
+
"eval_samples_per_second": 8.697,
|
343 |
+
"eval_steps_per_second": 1.089,
|
344 |
+
"step": 110544
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 25.0,
|
348 |
+
"learning_rate": 4.515848892748589e-09,
|
349 |
+
"loss": 1.9686,
|
350 |
+
"step": 115150
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 25.0,
|
354 |
+
"eval_loss": 1.5793741941452026,
|
355 |
+
"eval_runtime": 204.7422,
|
356 |
+
"eval_samples_per_second": 8.699,
|
357 |
+
"eval_steps_per_second": 1.089,
|
358 |
+
"step": 115150
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 25.0,
|
362 |
+
"step": 115150,
|
363 |
+
"total_flos": 2.2008277540080006e+20,
|
364 |
+
"train_loss": 1.217662907416956,
|
365 |
+
"train_runtime": 44441.0276,
|
366 |
+
"train_samples_per_second": 10.363,
|
367 |
+
"train_steps_per_second": 2.591
|
368 |
}
|
369 |
],
|
370 |
+
"max_steps": 115150,
|
371 |
+
"num_train_epochs": 25,
|
372 |
+
"total_flos": 2.2008277540080006e+20,
|
373 |
"trial_name": null,
|
374 |
"trial_params": null
|
375 |
}
|