Commit
•
e2d2f04
1
Parent(s):
608d311
End of training
Browse files- all_results.json +12 -12
- eval_results.json +7 -7
- train_results.json +7 -7
- trainer_state.json +850 -657
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 1.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.982857142857143,
|
| 3 |
+
"eval_accuracy": 0.7499688084840923,
|
| 4 |
+
"eval_loss": 0.8480741381645203,
|
| 5 |
+
"eval_runtime": 72.5483,
|
| 6 |
+
"eval_samples_per_second": 110.478,
|
| 7 |
+
"eval_steps_per_second": 3.46,
|
| 8 |
+
"total_flos": 3.468170115610067e+18,
|
| 9 |
+
"train_loss": 1.2575330436776537,
|
| 10 |
+
"train_runtime": 2631.6311,
|
| 11 |
+
"train_samples_per_second": 53.178,
|
| 12 |
+
"train_steps_per_second": 0.414
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.982857142857143,
|
| 3 |
+
"eval_accuracy": 0.7499688084840923,
|
| 4 |
+
"eval_loss": 0.8480741381645203,
|
| 5 |
+
"eval_runtime": 72.5483,
|
| 6 |
+
"eval_samples_per_second": 110.478,
|
| 7 |
+
"eval_steps_per_second": 3.46
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
-
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.982857142857143,
|
| 3 |
+
"total_flos": 3.468170115610067e+18,
|
| 4 |
+
"train_loss": 1.2575330436776537,
|
| 5 |
+
"train_runtime": 2631.6311,
|
| 6 |
+
"train_samples_per_second": 53.178,
|
| 7 |
+
"train_steps_per_second": 0.414
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,657 +1,850 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
-
"step": 10
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 2.
|
| 23 |
-
"step": 20
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss": 2.
|
| 30 |
-
"step": 30
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 2.
|
| 37 |
-
"step": 40
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 22.
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss": 2.
|
| 44 |
-
"step": 50
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
-
"step": 60
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 1.
|
| 58 |
-
"step": 70
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
-
"step": 80
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 4.
|
| 71 |
-
"loss": 1.
|
| 72 |
-
"step": 90
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 4.
|
| 78 |
-
"loss": 1.
|
| 79 |
-
"step": 100
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 4.
|
| 85 |
-
"loss": 1.
|
| 86 |
-
"step": 110
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss": 1.
|
| 93 |
-
"step": 120
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 4.
|
| 99 |
-
"loss": 1.
|
| 100 |
-
"step": 130
|
| 101 |
-
},
|
| 102 |
-
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 4.
|
| 106 |
-
"loss": 1.
|
| 107 |
-
"step": 140
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 4.
|
| 113 |
-
"loss": 1.
|
| 114 |
-
"step": 150
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 10.
|
| 119 |
-
"learning_rate": 4.
|
| 120 |
-
"loss": 1.
|
| 121 |
-
"step": 160
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 4.
|
| 127 |
-
"loss": 1.
|
| 128 |
-
"step": 170
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss": 1.
|
| 135 |
-
"step": 180
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 4.
|
| 141 |
-
"loss": 1.
|
| 142 |
-
"step": 190
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 4.
|
| 148 |
-
"loss": 1.
|
| 149 |
-
"step": 200
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 4.
|
| 155 |
-
"loss": 1.
|
| 156 |
-
"step": 210
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"
|
| 161 |
-
"
|
| 162 |
-
"
|
| 163 |
-
"
|
| 164 |
-
|
| 165 |
-
|
| 166 |
-
|
| 167 |
-
|
| 168 |
-
"
|
| 169 |
-
"
|
| 170 |
-
"
|
| 171 |
-
|
| 172 |
-
|
| 173 |
-
|
| 174 |
-
|
| 175 |
-
"
|
| 176 |
-
"
|
| 177 |
-
"
|
| 178 |
-
|
| 179 |
-
|
| 180 |
-
|
| 181 |
-
|
| 182 |
-
"
|
| 183 |
-
"
|
| 184 |
-
"
|
| 185 |
-
|
| 186 |
-
|
| 187 |
-
|
| 188 |
-
|
| 189 |
-
"
|
| 190 |
-
"
|
| 191 |
-
"
|
| 192 |
-
|
| 193 |
-
|
| 194 |
-
|
| 195 |
-
|
| 196 |
-
"
|
| 197 |
-
"
|
| 198 |
-
"
|
| 199 |
-
|
| 200 |
-
|
| 201 |
-
|
| 202 |
-
|
| 203 |
-
"
|
| 204 |
-
"
|
| 205 |
-
"
|
| 206 |
-
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
"
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
| 214 |
-
"step":
|
| 215 |
-
},
|
| 216 |
-
{
|
| 217 |
-
"epoch": 1.
|
| 218 |
-
"grad_norm":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 1.
|
| 221 |
-
"step": 290
|
| 222 |
-
},
|
| 223 |
-
{
|
| 224 |
-
"epoch": 1.
|
| 225 |
-
"grad_norm":
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss": 1.
|
| 228 |
-
"step": 300
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 1.
|
| 232 |
-
"grad_norm":
|
| 233 |
-
"learning_rate": 3.
|
| 234 |
-
"loss": 1.
|
| 235 |
-
"step": 310
|
| 236 |
-
},
|
| 237 |
-
{
|
| 238 |
-
"epoch": 1.
|
| 239 |
-
"grad_norm":
|
| 240 |
-
"learning_rate": 3.
|
| 241 |
-
"loss": 1.
|
| 242 |
-
"step": 320
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 1.
|
| 246 |
-
"grad_norm": 6.
|
| 247 |
-
"learning_rate": 3.
|
| 248 |
-
"loss": 1.
|
| 249 |
-
"step": 330
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 1.
|
| 253 |
-
"grad_norm":
|
| 254 |
-
"learning_rate": 3.
|
| 255 |
-
"loss": 1.
|
| 256 |
-
"step": 340
|
| 257 |
-
},
|
| 258 |
-
{
|
| 259 |
-
"epoch": 1.
|
| 260 |
-
"grad_norm": 7.
|
| 261 |
-
"learning_rate": 3.
|
| 262 |
-
"loss": 1.
|
| 263 |
-
"step": 350
|
| 264 |
-
},
|
| 265 |
-
{
|
| 266 |
-
"epoch": 1.
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"learning_rate": 3.
|
| 269 |
-
"loss": 1.
|
| 270 |
-
"step": 360
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 1.
|
| 274 |
-
"grad_norm":
|
| 275 |
-
"learning_rate": 3.
|
| 276 |
-
"loss": 1.
|
| 277 |
-
"step": 370
|
| 278 |
-
},
|
| 279 |
-
{
|
| 280 |
-
"epoch": 1.
|
| 281 |
-
"grad_norm":
|
| 282 |
-
"learning_rate": 3.
|
| 283 |
-
"loss": 1.
|
| 284 |
-
"step": 380
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 1.
|
| 288 |
-
"grad_norm":
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 1.
|
| 291 |
-
"step": 390
|
| 292 |
-
},
|
| 293 |
-
{
|
| 294 |
-
"epoch": 1.
|
| 295 |
-
"grad_norm":
|
| 296 |
-
"learning_rate":
|
| 297 |
-
"loss": 1.
|
| 298 |
-
"step": 400
|
| 299 |
-
},
|
| 300 |
-
{
|
| 301 |
-
"epoch": 1.
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 1.
|
| 305 |
-
"step": 410
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 1.
|
| 309 |
-
"grad_norm":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss": 1.
|
| 312 |
-
"step": 420
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 1.
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate":
|
| 318 |
-
"loss": 1.
|
| 319 |
-
"step": 430
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 1.
|
| 323 |
-
"
|
| 324 |
-
"
|
| 325 |
-
"
|
| 326 |
-
"
|
| 327 |
-
|
| 328 |
-
|
| 329 |
-
|
| 330 |
-
|
| 331 |
-
"
|
| 332 |
-
"
|
| 333 |
-
"
|
| 334 |
-
|
| 335 |
-
|
| 336 |
-
|
| 337 |
-
|
| 338 |
-
"
|
| 339 |
-
"
|
| 340 |
-
"
|
| 341 |
-
|
| 342 |
-
|
| 343 |
-
|
| 344 |
-
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"
|
| 348 |
-
|
| 349 |
-
|
| 350 |
-
|
| 351 |
-
|
| 352 |
-
"
|
| 353 |
-
"
|
| 354 |
-
"
|
| 355 |
-
|
| 356 |
-
|
| 357 |
-
|
| 358 |
-
|
| 359 |
-
"
|
| 360 |
-
"
|
| 361 |
-
"
|
| 362 |
-
|
| 363 |
-
|
| 364 |
-
|
| 365 |
-
|
| 366 |
-
"
|
| 367 |
-
"
|
| 368 |
-
"
|
| 369 |
-
|
| 370 |
-
|
| 371 |
-
|
| 372 |
-
|
| 373 |
-
"
|
| 374 |
-
"
|
| 375 |
-
"
|
| 376 |
-
|
| 377 |
-
|
| 378 |
-
|
| 379 |
-
|
| 380 |
-
"
|
| 381 |
-
"
|
| 382 |
-
"
|
| 383 |
-
|
| 384 |
-
|
| 385 |
-
|
| 386 |
-
|
| 387 |
-
"
|
| 388 |
-
"
|
| 389 |
-
"
|
| 390 |
-
|
| 391 |
-
|
| 392 |
-
|
| 393 |
-
|
| 394 |
-
"
|
| 395 |
-
"
|
| 396 |
-
"
|
| 397 |
-
|
| 398 |
-
|
| 399 |
-
|
| 400 |
-
|
| 401 |
-
"
|
| 402 |
-
"
|
| 403 |
-
"
|
| 404 |
-
|
| 405 |
-
|
| 406 |
-
|
| 407 |
-
|
| 408 |
-
"
|
| 409 |
-
"
|
| 410 |
-
"
|
| 411 |
-
|
| 412 |
-
|
| 413 |
-
|
| 414 |
-
|
| 415 |
-
"
|
| 416 |
-
"
|
| 417 |
-
"
|
| 418 |
-
"
|
| 419 |
-
"step":
|
| 420 |
-
},
|
| 421 |
-
{
|
| 422 |
-
"epoch": 2.
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss": 1.
|
| 426 |
-
"step": 570
|
| 427 |
-
},
|
| 428 |
-
{
|
| 429 |
-
"epoch": 2.
|
| 430 |
-
"grad_norm":
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss": 1.
|
| 433 |
-
"step": 580
|
| 434 |
-
},
|
| 435 |
-
{
|
| 436 |
-
"epoch": 2.
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 1.
|
| 440 |
-
"step": 590
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 2.
|
| 444 |
-
"grad_norm":
|
| 445 |
-
"learning_rate":
|
| 446 |
-
"loss": 1.
|
| 447 |
-
"step": 600
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 2.
|
| 451 |
-
"grad_norm":
|
| 452 |
-
"learning_rate":
|
| 453 |
-
"loss": 1.
|
| 454 |
-
"step": 610
|
| 455 |
-
},
|
| 456 |
-
{
|
| 457 |
-
"epoch": 2.
|
| 458 |
-
"grad_norm":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 1.
|
| 461 |
-
"step": 620
|
| 462 |
-
},
|
| 463 |
-
{
|
| 464 |
-
"epoch": 2.
|
| 465 |
-
"grad_norm":
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 1.
|
| 468 |
-
"step": 630
|
| 469 |
-
},
|
| 470 |
-
{
|
| 471 |
-
"epoch": 2.
|
| 472 |
-
"grad_norm": 8.
|
| 473 |
-
"learning_rate":
|
| 474 |
-
"loss": 1.
|
| 475 |
-
"step": 640
|
| 476 |
-
},
|
| 477 |
-
{
|
| 478 |
-
"epoch": 2.
|
| 479 |
-
"grad_norm":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 1.
|
| 482 |
-
"step": 650
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 2.
|
| 486 |
-
"
|
| 487 |
-
"
|
| 488 |
-
"
|
| 489 |
-
"
|
| 490 |
-
|
| 491 |
-
|
| 492 |
-
|
| 493 |
-
|
| 494 |
-
"
|
| 495 |
-
"
|
| 496 |
-
"
|
| 497 |
-
|
| 498 |
-
|
| 499 |
-
|
| 500 |
-
|
| 501 |
-
"
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
|
| 505 |
-
|
| 506 |
-
|
| 507 |
-
|
| 508 |
-
"
|
| 509 |
-
"
|
| 510 |
-
"
|
| 511 |
-
|
| 512 |
-
|
| 513 |
-
|
| 514 |
-
|
| 515 |
-
"
|
| 516 |
-
"
|
| 517 |
-
"
|
| 518 |
-
|
| 519 |
-
|
| 520 |
-
|
| 521 |
-
|
| 522 |
-
"
|
| 523 |
-
"
|
| 524 |
-
"
|
| 525 |
-
|
| 526 |
-
|
| 527 |
-
|
| 528 |
-
|
| 529 |
-
"
|
| 530 |
-
"
|
| 531 |
-
"
|
| 532 |
-
|
| 533 |
-
|
| 534 |
-
|
| 535 |
-
|
| 536 |
-
"
|
| 537 |
-
"
|
| 538 |
-
"
|
| 539 |
-
|
| 540 |
-
|
| 541 |
-
|
| 542 |
-
|
| 543 |
-
"
|
| 544 |
-
"
|
| 545 |
-
"
|
| 546 |
-
|
| 547 |
-
|
| 548 |
-
|
| 549 |
-
|
| 550 |
-
"
|
| 551 |
-
"
|
| 552 |
-
"
|
| 553 |
-
|
| 554 |
-
|
| 555 |
-
|
| 556 |
-
|
| 557 |
-
"
|
| 558 |
-
"
|
| 559 |
-
"
|
| 560 |
-
|
| 561 |
-
|
| 562 |
-
|
| 563 |
-
|
| 564 |
-
"
|
| 565 |
-
"
|
| 566 |
-
"
|
| 567 |
-
|
| 568 |
-
|
| 569 |
-
|
| 570 |
-
|
| 571 |
-
"
|
| 572 |
-
"
|
| 573 |
-
"
|
| 574 |
-
|
| 575 |
-
|
| 576 |
-
|
| 577 |
-
|
| 578 |
-
"
|
| 579 |
-
"
|
| 580 |
-
"
|
| 581 |
-
|
| 582 |
-
|
| 583 |
-
|
| 584 |
-
|
| 585 |
-
"
|
| 586 |
-
"
|
| 587 |
-
"
|
| 588 |
-
|
| 589 |
-
|
| 590 |
-
|
| 591 |
-
|
| 592 |
-
"
|
| 593 |
-
"
|
| 594 |
-
"
|
| 595 |
-
|
| 596 |
-
|
| 597 |
-
|
| 598 |
-
|
| 599 |
-
"
|
| 600 |
-
"
|
| 601 |
-
"
|
| 602 |
-
|
| 603 |
-
|
| 604 |
-
|
| 605 |
-
|
| 606 |
-
"
|
| 607 |
-
"
|
| 608 |
-
"
|
| 609 |
-
|
| 610 |
-
|
| 611 |
-
|
| 612 |
-
|
| 613 |
-
"
|
| 614 |
-
"
|
| 615 |
-
"
|
| 616 |
-
|
| 617 |
-
|
| 618 |
-
|
| 619 |
-
|
| 620 |
-
"
|
| 621 |
-
"
|
| 622 |
-
"
|
| 623 |
-
"
|
| 624 |
-
"step":
|
| 625 |
-
},
|
| 626 |
-
{
|
| 627 |
-
"epoch":
|
| 628 |
-
"
|
| 629 |
-
"
|
| 630 |
-
"
|
| 631 |
-
"
|
| 632 |
-
|
| 633 |
-
|
| 634 |
-
|
| 635 |
-
|
| 636 |
-
|
| 637 |
-
|
| 638 |
-
|
| 639 |
-
|
| 640 |
-
|
| 641 |
-
|
| 642 |
-
|
| 643 |
-
"
|
| 644 |
-
|
| 645 |
-
|
| 646 |
-
|
| 647 |
-
|
| 648 |
-
|
| 649 |
-
|
| 650 |
-
"
|
| 651 |
-
|
| 652 |
-
|
| 653 |
-
|
| 654 |
-
|
| 655 |
-
|
| 656 |
-
|
| 657 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7499688084840923,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat\\checkpoint-1090",
|
| 4 |
+
"epoch": 4.982857142857143,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1090,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.045714285714285714,
|
| 13 |
+
"grad_norm": 6.963473320007324,
|
| 14 |
+
"learning_rate": 4.587155963302753e-06,
|
| 15 |
+
"loss": 2.8055,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09142857142857143,
|
| 20 |
+
"grad_norm": 11.695841789245605,
|
| 21 |
+
"learning_rate": 9.174311926605506e-06,
|
| 22 |
+
"loss": 2.7626,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.13714285714285715,
|
| 27 |
+
"grad_norm": 10.859987258911133,
|
| 28 |
+
"learning_rate": 1.3761467889908258e-05,
|
| 29 |
+
"loss": 2.6863,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.18285714285714286,
|
| 34 |
+
"grad_norm": 14.530256271362305,
|
| 35 |
+
"learning_rate": 1.834862385321101e-05,
|
| 36 |
+
"loss": 2.5932,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.22857142857142856,
|
| 41 |
+
"grad_norm": 22.234731674194336,
|
| 42 |
+
"learning_rate": 2.2935779816513765e-05,
|
| 43 |
+
"loss": 2.4006,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2742857142857143,
|
| 48 |
+
"grad_norm": 21.946706771850586,
|
| 49 |
+
"learning_rate": 2.7522935779816515e-05,
|
| 50 |
+
"loss": 2.1545,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"grad_norm": 31.88918685913086,
|
| 56 |
+
"learning_rate": 3.211009174311927e-05,
|
| 57 |
+
"loss": 1.9868,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3657142857142857,
|
| 62 |
+
"grad_norm": 17.8656063079834,
|
| 63 |
+
"learning_rate": 3.669724770642202e-05,
|
| 64 |
+
"loss": 1.8264,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.4114285714285714,
|
| 69 |
+
"grad_norm": 15.480560302734375,
|
| 70 |
+
"learning_rate": 4.1284403669724776e-05,
|
| 71 |
+
"loss": 1.8301,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.45714285714285713,
|
| 76 |
+
"grad_norm": 13.637527465820312,
|
| 77 |
+
"learning_rate": 4.587155963302753e-05,
|
| 78 |
+
"loss": 1.7708,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5028571428571429,
|
| 83 |
+
"grad_norm": 16.25617218017578,
|
| 84 |
+
"learning_rate": 4.994903160040775e-05,
|
| 85 |
+
"loss": 1.7133,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.5485714285714286,
|
| 90 |
+
"grad_norm": 10.56847095489502,
|
| 91 |
+
"learning_rate": 4.943934760448522e-05,
|
| 92 |
+
"loss": 1.5611,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5942857142857143,
|
| 97 |
+
"grad_norm": 16.05897331237793,
|
| 98 |
+
"learning_rate": 4.892966360856269e-05,
|
| 99 |
+
"loss": 1.6387,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.64,
|
| 104 |
+
"grad_norm": 17.286190032958984,
|
| 105 |
+
"learning_rate": 4.8419979612640164e-05,
|
| 106 |
+
"loss": 1.519,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6857142857142857,
|
| 111 |
+
"grad_norm": 13.946508407592773,
|
| 112 |
+
"learning_rate": 4.7910295616717635e-05,
|
| 113 |
+
"loss": 1.4708,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7314285714285714,
|
| 118 |
+
"grad_norm": 10.632002830505371,
|
| 119 |
+
"learning_rate": 4.740061162079511e-05,
|
| 120 |
+
"loss": 1.5119,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7771428571428571,
|
| 125 |
+
"grad_norm": 11.511331558227539,
|
| 126 |
+
"learning_rate": 4.6890927624872586e-05,
|
| 127 |
+
"loss": 1.4476,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8228571428571428,
|
| 132 |
+
"grad_norm": 9.265079498291016,
|
| 133 |
+
"learning_rate": 4.638124362895006e-05,
|
| 134 |
+
"loss": 1.4864,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.8685714285714285,
|
| 139 |
+
"grad_norm": 12.077649116516113,
|
| 140 |
+
"learning_rate": 4.587155963302753e-05,
|
| 141 |
+
"loss": 1.4542,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9142857142857143,
|
| 146 |
+
"grad_norm": 8.974185943603516,
|
| 147 |
+
"learning_rate": 4.5361875637104995e-05,
|
| 148 |
+
"loss": 1.4505,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.96,
|
| 153 |
+
"grad_norm": 7.196322441101074,
|
| 154 |
+
"learning_rate": 4.4852191641182466e-05,
|
| 155 |
+
"loss": 1.4736,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9965714285714286,
|
| 160 |
+
"eval_accuracy": 0.6177167810355584,
|
| 161 |
+
"eval_loss": 1.2365009784698486,
|
| 162 |
+
"eval_runtime": 74.6739,
|
| 163 |
+
"eval_samples_per_second": 107.333,
|
| 164 |
+
"eval_steps_per_second": 3.361,
|
| 165 |
+
"step": 218
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.0057142857142858,
|
| 169 |
+
"grad_norm": 8.575163841247559,
|
| 170 |
+
"learning_rate": 4.434250764525994e-05,
|
| 171 |
+
"loss": 1.2932,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.0514285714285714,
|
| 176 |
+
"grad_norm": 9.491937637329102,
|
| 177 |
+
"learning_rate": 4.383282364933741e-05,
|
| 178 |
+
"loss": 1.3322,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.0971428571428572,
|
| 183 |
+
"grad_norm": 7.442806720733643,
|
| 184 |
+
"learning_rate": 4.332313965341488e-05,
|
| 185 |
+
"loss": 1.3634,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.1428571428571428,
|
| 190 |
+
"grad_norm": 11.594585418701172,
|
| 191 |
+
"learning_rate": 4.281345565749236e-05,
|
| 192 |
+
"loss": 1.3838,
|
| 193 |
+
"step": 250
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.1885714285714286,
|
| 197 |
+
"grad_norm": 9.72411823272705,
|
| 198 |
+
"learning_rate": 4.230377166156983e-05,
|
| 199 |
+
"loss": 1.3695,
|
| 200 |
+
"step": 260
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.2342857142857142,
|
| 204 |
+
"grad_norm": 8.765079498291016,
|
| 205 |
+
"learning_rate": 4.1794087665647304e-05,
|
| 206 |
+
"loss": 1.241,
|
| 207 |
+
"step": 270
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.28,
|
| 211 |
+
"grad_norm": 7.949846267700195,
|
| 212 |
+
"learning_rate": 4.1284403669724776e-05,
|
| 213 |
+
"loss": 1.3099,
|
| 214 |
+
"step": 280
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.3257142857142856,
|
| 218 |
+
"grad_norm": 7.360624313354492,
|
| 219 |
+
"learning_rate": 4.077471967380224e-05,
|
| 220 |
+
"loss": 1.3464,
|
| 221 |
+
"step": 290
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.3714285714285714,
|
| 225 |
+
"grad_norm": 9.947721481323242,
|
| 226 |
+
"learning_rate": 4.026503567787971e-05,
|
| 227 |
+
"loss": 1.269,
|
| 228 |
+
"step": 300
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.4171428571428573,
|
| 232 |
+
"grad_norm": 7.578132629394531,
|
| 233 |
+
"learning_rate": 3.9755351681957185e-05,
|
| 234 |
+
"loss": 1.2697,
|
| 235 |
+
"step": 310
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.4628571428571429,
|
| 239 |
+
"grad_norm": 7.614137172698975,
|
| 240 |
+
"learning_rate": 3.9245667686034656e-05,
|
| 241 |
+
"loss": 1.2017,
|
| 242 |
+
"step": 320
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.5085714285714285,
|
| 246 |
+
"grad_norm": 6.3952107429504395,
|
| 247 |
+
"learning_rate": 3.8735983690112135e-05,
|
| 248 |
+
"loss": 1.3131,
|
| 249 |
+
"step": 330
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.5542857142857143,
|
| 253 |
+
"grad_norm": 11.049560546875,
|
| 254 |
+
"learning_rate": 3.822629969418961e-05,
|
| 255 |
+
"loss": 1.2547,
|
| 256 |
+
"step": 340
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.6,
|
| 260 |
+
"grad_norm": 7.0264129638671875,
|
| 261 |
+
"learning_rate": 3.771661569826708e-05,
|
| 262 |
+
"loss": 1.2384,
|
| 263 |
+
"step": 350
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.6457142857142857,
|
| 267 |
+
"grad_norm": 8.316610336303711,
|
| 268 |
+
"learning_rate": 3.720693170234455e-05,
|
| 269 |
+
"loss": 1.2679,
|
| 270 |
+
"step": 360
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.6914285714285713,
|
| 274 |
+
"grad_norm": 7.154911041259766,
|
| 275 |
+
"learning_rate": 3.669724770642202e-05,
|
| 276 |
+
"loss": 1.3185,
|
| 277 |
+
"step": 370
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.737142857142857,
|
| 281 |
+
"grad_norm": 7.319551944732666,
|
| 282 |
+
"learning_rate": 3.6187563710499494e-05,
|
| 283 |
+
"loss": 1.1742,
|
| 284 |
+
"step": 380
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.782857142857143,
|
| 288 |
+
"grad_norm": 7.423411846160889,
|
| 289 |
+
"learning_rate": 3.567787971457696e-05,
|
| 290 |
+
"loss": 1.1981,
|
| 291 |
+
"step": 390
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.8285714285714287,
|
| 295 |
+
"grad_norm": 9.163586616516113,
|
| 296 |
+
"learning_rate": 3.516819571865443e-05,
|
| 297 |
+
"loss": 1.2371,
|
| 298 |
+
"step": 400
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.8742857142857143,
|
| 302 |
+
"grad_norm": 7.521251678466797,
|
| 303 |
+
"learning_rate": 3.465851172273191e-05,
|
| 304 |
+
"loss": 1.2717,
|
| 305 |
+
"step": 410
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.92,
|
| 309 |
+
"grad_norm": 10.229044914245605,
|
| 310 |
+
"learning_rate": 3.414882772680938e-05,
|
| 311 |
+
"loss": 1.206,
|
| 312 |
+
"step": 420
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.9657142857142857,
|
| 316 |
+
"grad_norm": 7.633674621582031,
|
| 317 |
+
"learning_rate": 3.363914373088685e-05,
|
| 318 |
+
"loss": 1.2161,
|
| 319 |
+
"step": 430
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.9977142857142858,
|
| 323 |
+
"eval_accuracy": 0.6915782907049283,
|
| 324 |
+
"eval_loss": 1.0157994031906128,
|
| 325 |
+
"eval_runtime": 74.9081,
|
| 326 |
+
"eval_samples_per_second": 106.998,
|
| 327 |
+
"eval_steps_per_second": 3.351,
|
| 328 |
+
"step": 437
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.0114285714285716,
|
| 332 |
+
"grad_norm": 8.755953788757324,
|
| 333 |
+
"learning_rate": 3.3129459734964325e-05,
|
| 334 |
+
"loss": 1.1613,
|
| 335 |
+
"step": 440
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.057142857142857,
|
| 339 |
+
"grad_norm": 7.305307865142822,
|
| 340 |
+
"learning_rate": 3.26197757390418e-05,
|
| 341 |
+
"loss": 1.0986,
|
| 342 |
+
"step": 450
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.1028571428571428,
|
| 346 |
+
"grad_norm": 8.31972885131836,
|
| 347 |
+
"learning_rate": 3.211009174311927e-05,
|
| 348 |
+
"loss": 1.1722,
|
| 349 |
+
"step": 460
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 2.1485714285714286,
|
| 353 |
+
"grad_norm": 8.749483108520508,
|
| 354 |
+
"learning_rate": 3.160040774719674e-05,
|
| 355 |
+
"loss": 1.1203,
|
| 356 |
+
"step": 470
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.1942857142857144,
|
| 360 |
+
"grad_norm": 6.05934476852417,
|
| 361 |
+
"learning_rate": 3.1090723751274206e-05,
|
| 362 |
+
"loss": 1.1225,
|
| 363 |
+
"step": 480
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.24,
|
| 367 |
+
"grad_norm": 10.040249824523926,
|
| 368 |
+
"learning_rate": 3.0581039755351684e-05,
|
| 369 |
+
"loss": 1.2017,
|
| 370 |
+
"step": 490
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 2.2857142857142856,
|
| 374 |
+
"grad_norm": 8.599287986755371,
|
| 375 |
+
"learning_rate": 3.0071355759429153e-05,
|
| 376 |
+
"loss": 1.1669,
|
| 377 |
+
"step": 500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.3314285714285714,
|
| 381 |
+
"grad_norm": 7.941746711730957,
|
| 382 |
+
"learning_rate": 2.9561671763506628e-05,
|
| 383 |
+
"loss": 1.1189,
|
| 384 |
+
"step": 510
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.3771428571428572,
|
| 388 |
+
"grad_norm": 13.228888511657715,
|
| 389 |
+
"learning_rate": 2.90519877675841e-05,
|
| 390 |
+
"loss": 1.1882,
|
| 391 |
+
"step": 520
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.422857142857143,
|
| 395 |
+
"grad_norm": 7.503291130065918,
|
| 396 |
+
"learning_rate": 2.854230377166157e-05,
|
| 397 |
+
"loss": 1.185,
|
| 398 |
+
"step": 530
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.4685714285714284,
|
| 402 |
+
"grad_norm": 7.043280124664307,
|
| 403 |
+
"learning_rate": 2.8032619775739043e-05,
|
| 404 |
+
"loss": 1.1577,
|
| 405 |
+
"step": 540
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.5142857142857142,
|
| 409 |
+
"grad_norm": 8.937540054321289,
|
| 410 |
+
"learning_rate": 2.7522935779816515e-05,
|
| 411 |
+
"loss": 1.1539,
|
| 412 |
+
"step": 550
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 2.56,
|
| 416 |
+
"grad_norm": 8.076375961303711,
|
| 417 |
+
"learning_rate": 2.701325178389399e-05,
|
| 418 |
+
"loss": 1.1124,
|
| 419 |
+
"step": 560
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.605714285714286,
|
| 423 |
+
"grad_norm": 7.4971442222595215,
|
| 424 |
+
"learning_rate": 2.6503567787971462e-05,
|
| 425 |
+
"loss": 1.0799,
|
| 426 |
+
"step": 570
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.6514285714285712,
|
| 430 |
+
"grad_norm": 9.443109512329102,
|
| 431 |
+
"learning_rate": 2.5993883792048927e-05,
|
| 432 |
+
"loss": 1.0234,
|
| 433 |
+
"step": 580
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 2.697142857142857,
|
| 437 |
+
"grad_norm": 6.811278820037842,
|
| 438 |
+
"learning_rate": 2.5484199796126402e-05,
|
| 439 |
+
"loss": 1.0423,
|
| 440 |
+
"step": 590
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 2.742857142857143,
|
| 444 |
+
"grad_norm": 8.011152267456055,
|
| 445 |
+
"learning_rate": 2.4974515800203874e-05,
|
| 446 |
+
"loss": 1.1336,
|
| 447 |
+
"step": 600
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.7885714285714287,
|
| 451 |
+
"grad_norm": 6.383072853088379,
|
| 452 |
+
"learning_rate": 2.4464831804281346e-05,
|
| 453 |
+
"loss": 1.1203,
|
| 454 |
+
"step": 610
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.8342857142857145,
|
| 458 |
+
"grad_norm": 9.429741859436035,
|
| 459 |
+
"learning_rate": 2.3955147808358818e-05,
|
| 460 |
+
"loss": 1.091,
|
| 461 |
+
"step": 620
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 2.88,
|
| 465 |
+
"grad_norm": 6.606307506561279,
|
| 466 |
+
"learning_rate": 2.3445463812436293e-05,
|
| 467 |
+
"loss": 1.058,
|
| 468 |
+
"step": 630
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.9257142857142857,
|
| 472 |
+
"grad_norm": 8.834084510803223,
|
| 473 |
+
"learning_rate": 2.2935779816513765e-05,
|
| 474 |
+
"loss": 1.1266,
|
| 475 |
+
"step": 640
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 2.9714285714285715,
|
| 479 |
+
"grad_norm": 7.674890518188477,
|
| 480 |
+
"learning_rate": 2.2426095820591233e-05,
|
| 481 |
+
"loss": 1.0807,
|
| 482 |
+
"step": 650
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.998857142857143,
|
| 486 |
+
"eval_accuracy": 0.727386150966937,
|
| 487 |
+
"eval_loss": 0.9072983264923096,
|
| 488 |
+
"eval_runtime": 74.6334,
|
| 489 |
+
"eval_samples_per_second": 107.392,
|
| 490 |
+
"eval_steps_per_second": 3.363,
|
| 491 |
+
"step": 656
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.0171428571428573,
|
| 495 |
+
"grad_norm": 7.3469624519348145,
|
| 496 |
+
"learning_rate": 2.1916411824668705e-05,
|
| 497 |
+
"loss": 1.0863,
|
| 498 |
+
"step": 660
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 3.0628571428571427,
|
| 502 |
+
"grad_norm": 7.052463531494141,
|
| 503 |
+
"learning_rate": 2.140672782874618e-05,
|
| 504 |
+
"loss": 1.0188,
|
| 505 |
+
"step": 670
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 3.1085714285714285,
|
| 509 |
+
"grad_norm": 7.651565074920654,
|
| 510 |
+
"learning_rate": 2.0897043832823652e-05,
|
| 511 |
+
"loss": 0.9874,
|
| 512 |
+
"step": 680
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.1542857142857144,
|
| 516 |
+
"grad_norm": 11.283343315124512,
|
| 517 |
+
"learning_rate": 2.038735983690112e-05,
|
| 518 |
+
"loss": 1.0876,
|
| 519 |
+
"step": 690
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.2,
|
| 523 |
+
"grad_norm": 7.787779331207275,
|
| 524 |
+
"learning_rate": 1.9877675840978592e-05,
|
| 525 |
+
"loss": 1.0479,
|
| 526 |
+
"step": 700
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.2457142857142856,
|
| 530 |
+
"grad_norm": 8.140477180480957,
|
| 531 |
+
"learning_rate": 1.9367991845056068e-05,
|
| 532 |
+
"loss": 1.009,
|
| 533 |
+
"step": 710
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.2914285714285714,
|
| 537 |
+
"grad_norm": 8.705915451049805,
|
| 538 |
+
"learning_rate": 1.885830784913354e-05,
|
| 539 |
+
"loss": 1.056,
|
| 540 |
+
"step": 720
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 3.337142857142857,
|
| 544 |
+
"grad_norm": 7.134729385375977,
|
| 545 |
+
"learning_rate": 1.834862385321101e-05,
|
| 546 |
+
"loss": 1.0378,
|
| 547 |
+
"step": 730
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 3.382857142857143,
|
| 551 |
+
"grad_norm": 7.588448524475098,
|
| 552 |
+
"learning_rate": 1.783893985728848e-05,
|
| 553 |
+
"loss": 1.0508,
|
| 554 |
+
"step": 740
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 3.4285714285714284,
|
| 558 |
+
"grad_norm": 7.663401126861572,
|
| 559 |
+
"learning_rate": 1.7329255861365955e-05,
|
| 560 |
+
"loss": 1.0207,
|
| 561 |
+
"step": 750
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.474285714285714,
|
| 565 |
+
"grad_norm": 6.939538955688477,
|
| 566 |
+
"learning_rate": 1.6819571865443427e-05,
|
| 567 |
+
"loss": 1.1108,
|
| 568 |
+
"step": 760
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 3.52,
|
| 572 |
+
"grad_norm": 7.630512714385986,
|
| 573 |
+
"learning_rate": 1.63098878695209e-05,
|
| 574 |
+
"loss": 0.9975,
|
| 575 |
+
"step": 770
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 3.565714285714286,
|
| 579 |
+
"grad_norm": 9.263322830200195,
|
| 580 |
+
"learning_rate": 1.580020387359837e-05,
|
| 581 |
+
"loss": 1.0372,
|
| 582 |
+
"step": 780
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 3.611428571428571,
|
| 586 |
+
"grad_norm": 8.593291282653809,
|
| 587 |
+
"learning_rate": 1.5290519877675842e-05,
|
| 588 |
+
"loss": 1.0863,
|
| 589 |
+
"step": 790
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 3.657142857142857,
|
| 593 |
+
"grad_norm": 7.475892543792725,
|
| 594 |
+
"learning_rate": 1.4780835881753314e-05,
|
| 595 |
+
"loss": 1.0021,
|
| 596 |
+
"step": 800
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 3.702857142857143,
|
| 600 |
+
"grad_norm": 9.492719650268555,
|
| 601 |
+
"learning_rate": 1.4271151885830786e-05,
|
| 602 |
+
"loss": 1.0572,
|
| 603 |
+
"step": 810
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 3.7485714285714287,
|
| 607 |
+
"grad_norm": 11.467538833618164,
|
| 608 |
+
"learning_rate": 1.3761467889908258e-05,
|
| 609 |
+
"loss": 1.0216,
|
| 610 |
+
"step": 820
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 3.7942857142857145,
|
| 614 |
+
"grad_norm": 7.794005870819092,
|
| 615 |
+
"learning_rate": 1.3251783893985731e-05,
|
| 616 |
+
"loss": 1.0205,
|
| 617 |
+
"step": 830
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 3.84,
|
| 621 |
+
"grad_norm": 6.822214603424072,
|
| 622 |
+
"learning_rate": 1.2742099898063201e-05,
|
| 623 |
+
"loss": 1.0281,
|
| 624 |
+
"step": 840
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.8857142857142857,
|
| 628 |
+
"grad_norm": 6.747819423675537,
|
| 629 |
+
"learning_rate": 1.2232415902140673e-05,
|
| 630 |
+
"loss": 1.0237,
|
| 631 |
+
"step": 850
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 3.9314285714285715,
|
| 635 |
+
"grad_norm": 7.018404006958008,
|
| 636 |
+
"learning_rate": 1.1722731906218146e-05,
|
| 637 |
+
"loss": 1.0446,
|
| 638 |
+
"step": 860
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 3.977142857142857,
|
| 642 |
+
"grad_norm": 7.553677082061768,
|
| 643 |
+
"learning_rate": 1.1213047910295617e-05,
|
| 644 |
+
"loss": 0.9977,
|
| 645 |
+
"step": 870
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 4.0,
|
| 649 |
+
"eval_accuracy": 0.7456019962570181,
|
| 650 |
+
"eval_loss": 0.8551267385482788,
|
| 651 |
+
"eval_runtime": 74.4356,
|
| 652 |
+
"eval_samples_per_second": 107.677,
|
| 653 |
+
"eval_steps_per_second": 3.372,
|
| 654 |
+
"step": 875
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 4.022857142857143,
|
| 658 |
+
"grad_norm": 8.310098648071289,
|
| 659 |
+
"learning_rate": 1.070336391437309e-05,
|
| 660 |
+
"loss": 1.0626,
|
| 661 |
+
"step": 880
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 4.0685714285714285,
|
| 665 |
+
"grad_norm": 7.971031665802002,
|
| 666 |
+
"learning_rate": 1.019367991845056e-05,
|
| 667 |
+
"loss": 1.0294,
|
| 668 |
+
"step": 890
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 4.114285714285714,
|
| 672 |
+
"grad_norm": 8.742574691772461,
|
| 673 |
+
"learning_rate": 9.683995922528034e-06,
|
| 674 |
+
"loss": 1.0401,
|
| 675 |
+
"step": 900
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.16,
|
| 679 |
+
"grad_norm": 7.5797553062438965,
|
| 680 |
+
"learning_rate": 9.174311926605506e-06,
|
| 681 |
+
"loss": 1.0194,
|
| 682 |
+
"step": 910
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 4.2057142857142855,
|
| 686 |
+
"grad_norm": 10.30631160736084,
|
| 687 |
+
"learning_rate": 8.664627930682977e-06,
|
| 688 |
+
"loss": 1.0334,
|
| 689 |
+
"step": 920
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 4.251428571428572,
|
| 693 |
+
"grad_norm": 8.128324508666992,
|
| 694 |
+
"learning_rate": 8.15494393476045e-06,
|
| 695 |
+
"loss": 1.0027,
|
| 696 |
+
"step": 930
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 4.297142857142857,
|
| 700 |
+
"grad_norm": 7.413013935089111,
|
| 701 |
+
"learning_rate": 7.645259938837921e-06,
|
| 702 |
+
"loss": 0.9981,
|
| 703 |
+
"step": 940
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 4.3428571428571425,
|
| 707 |
+
"grad_norm": 7.2544426918029785,
|
| 708 |
+
"learning_rate": 7.135575942915393e-06,
|
| 709 |
+
"loss": 0.9954,
|
| 710 |
+
"step": 950
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 4.388571428571429,
|
| 714 |
+
"grad_norm": 6.764856815338135,
|
| 715 |
+
"learning_rate": 6.6258919469928655e-06,
|
| 716 |
+
"loss": 0.9712,
|
| 717 |
+
"step": 960
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 4.434285714285714,
|
| 721 |
+
"grad_norm": 7.096611022949219,
|
| 722 |
+
"learning_rate": 6.1162079510703365e-06,
|
| 723 |
+
"loss": 0.9646,
|
| 724 |
+
"step": 970
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 4.48,
|
| 728 |
+
"grad_norm": 7.685801982879639,
|
| 729 |
+
"learning_rate": 5.606523955147808e-06,
|
| 730 |
+
"loss": 1.0362,
|
| 731 |
+
"step": 980
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 4.525714285714286,
|
| 735 |
+
"grad_norm": 7.416136741638184,
|
| 736 |
+
"learning_rate": 5.09683995922528e-06,
|
| 737 |
+
"loss": 1.0596,
|
| 738 |
+
"step": 990
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 4.571428571428571,
|
| 742 |
+
"grad_norm": 7.128796100616455,
|
| 743 |
+
"learning_rate": 4.587155963302753e-06,
|
| 744 |
+
"loss": 0.9479,
|
| 745 |
+
"step": 1000
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 4.617142857142857,
|
| 749 |
+
"grad_norm": 6.152943134307861,
|
| 750 |
+
"learning_rate": 4.077471967380225e-06,
|
| 751 |
+
"loss": 0.9503,
|
| 752 |
+
"step": 1010
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 4.662857142857143,
|
| 756 |
+
"grad_norm": 8.12887191772461,
|
| 757 |
+
"learning_rate": 3.5677879714576964e-06,
|
| 758 |
+
"loss": 1.0071,
|
| 759 |
+
"step": 1020
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 4.708571428571428,
|
| 763 |
+
"grad_norm": 5.702174186706543,
|
| 764 |
+
"learning_rate": 3.0581039755351682e-06,
|
| 765 |
+
"loss": 1.0149,
|
| 766 |
+
"step": 1030
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 4.7542857142857144,
|
| 770 |
+
"grad_norm": 6.921694278717041,
|
| 771 |
+
"learning_rate": 2.54841997961264e-06,
|
| 772 |
+
"loss": 0.981,
|
| 773 |
+
"step": 1040
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 4.8,
|
| 777 |
+
"grad_norm": 9.122147560119629,
|
| 778 |
+
"learning_rate": 2.0387359836901123e-06,
|
| 779 |
+
"loss": 0.9939,
|
| 780 |
+
"step": 1050
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 4.845714285714286,
|
| 784 |
+
"grad_norm": 8.022086143493652,
|
| 785 |
+
"learning_rate": 1.5290519877675841e-06,
|
| 786 |
+
"loss": 0.8968,
|
| 787 |
+
"step": 1060
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 4.8914285714285715,
|
| 791 |
+
"grad_norm": 7.508928298950195,
|
| 792 |
+
"learning_rate": 1.0193679918450562e-06,
|
| 793 |
+
"loss": 0.9698,
|
| 794 |
+
"step": 1070
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 4.937142857142857,
|
| 798 |
+
"grad_norm": 9.233248710632324,
|
| 799 |
+
"learning_rate": 5.096839959225281e-07,
|
| 800 |
+
"loss": 0.9584,
|
| 801 |
+
"step": 1080
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 4.982857142857143,
|
| 805 |
+
"grad_norm": 7.491491794586182,
|
| 806 |
+
"learning_rate": 0.0,
|
| 807 |
+
"loss": 0.9737,
|
| 808 |
+
"step": 1090
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 4.982857142857143,
|
| 812 |
+
"eval_accuracy": 0.7499688084840923,
|
| 813 |
+
"eval_loss": 0.8480741381645203,
|
| 814 |
+
"eval_runtime": 71.0957,
|
| 815 |
+
"eval_samples_per_second": 112.735,
|
| 816 |
+
"eval_steps_per_second": 3.53,
|
| 817 |
+
"step": 1090
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 4.982857142857143,
|
| 821 |
+
"step": 1090,
|
| 822 |
+
"total_flos": 3.468170115610067e+18,
|
| 823 |
+
"train_loss": 1.2575330436776537,
|
| 824 |
+
"train_runtime": 2631.6311,
|
| 825 |
+
"train_samples_per_second": 53.178,
|
| 826 |
+
"train_steps_per_second": 0.414
|
| 827 |
+
}
|
| 828 |
+
],
|
| 829 |
+
"logging_steps": 10,
|
| 830 |
+
"max_steps": 1090,
|
| 831 |
+
"num_input_tokens_seen": 0,
|
| 832 |
+
"num_train_epochs": 5,
|
| 833 |
+
"save_steps": 500,
|
| 834 |
+
"stateful_callbacks": {
|
| 835 |
+
"TrainerControl": {
|
| 836 |
+
"args": {
|
| 837 |
+
"should_epoch_stop": false,
|
| 838 |
+
"should_evaluate": false,
|
| 839 |
+
"should_log": false,
|
| 840 |
+
"should_save": true,
|
| 841 |
+
"should_training_stop": true
|
| 842 |
+
},
|
| 843 |
+
"attributes": {}
|
| 844 |
+
}
|
| 845 |
+
},
|
| 846 |
+
"total_flos": 3.468170115610067e+18,
|
| 847 |
+
"train_batch_size": 32,
|
| 848 |
+
"trial_name": null,
|
| 849 |
+
"trial_params": null
|
| 850 |
+
}
|